The extended utilization of picture-enhancing or manipulating tools has led to ease of manipulating multimedia data which includes digital images. These manipulations will disturb the truthfulness and lawfulness of images, resulting in misapprehension, and might disturb social security. The image forensic approach has been employed for detecting whether or not an image has been manipulated with the usage of positive attacks which includes splicing, and copy-move. This paper provides a competent tampering detection technique using resampling features and convolution neural network (CNN). In this model range spatial filtering (RSF)-CNN, throughout preprocessing the image is divided into consistent patches. Then, within every patch, the resampling features are extracted by utilizing affine transformation and the Laplacian operator. Then, the extracted features are accumulated for creating descriptors by using CNN. A wide-ranging analysis is performed for assessing tampering detection and tampered region segmentation accuracies of proposed RSF-CNN based tampering detection procedures considering various falsifications and post-processing attacks which include joint photographic expert group (JPEG) compression, scaling, rotations, noise additions, and more than one manipulation. From the achieved results, it can be visible the RSF-CNN primarily based tampering detection with adequately higher accurateness than existing tampering detection methodologies.
Extraction of image resampling using correlation aware convolution neural ne...IJECEIAES
Detecting hybrid tampering attacks in an image is extremely difficult; especially when copy-clone tampered segments exhibit identical illumination and contrast level about genuine objects. The existing method fails to detect tampering when the image undergoes hybrid transformation such as scaling, rotation, compression, and also fails to detect under smallsmooth tampering. The existing resampling feature extraction using the Deep learning techniques fails to obtain a good correlation among neighboring pixels in both horizontal and vertical directions. This work presents correlation aware convolution neural network (CA-CNN) for extracting resampling features for detecting hybrid tampering attacks. Here the image is resized for detecting tampering under a small-smooth region. The CA-CNN is composed of a three-layer horizontal, vertical, and correlated layer. The correlated layer is used for obtaining correlated resampling feature among horizontal sequence and vertical sequence. Then feature is aggregated and the descriptor is built. An experiment is conducted to evaluate the performance of the CA-CNN model over existing tampering detection methodologies considering the various datasets. From the result achieved it can be seen the CA-CNN is efficient considering various distortions and post-processing attacks such joint photographic expert group (JPEG) compression, and scaling. This model achieves much better accuracies, recall, precision, false positive rate (FPR), and F-measure compared existing methodologies.
Machine learning based augmented reality for improved learning application th...IJECEIAES
Detection of objects and their location in an image are important elements of current research in computer vision. In May 2020, Meta released its state-ofthe-art object-detection model based on a transformer architecture called detection transformer (DETR). There are several object-detection models such as region-based convolutional neural network (R-CNN), you only look once (YOLO) and single shot detectors (SSD), but none have used a transformer to accomplish this task. These models mentioned earlier, use all sorts of hyperparameters and layers. However, the advantages of using a transformer pattern make the architecture simple and easy to implement. In this paper, we determine the name of a chemical experiment through two steps: firstly, by building a DETR model, trained on a customized dataset, and then integrate it into an augmented reality mobile application. By detecting the objects used during the realization of an experiment, we can predict the name of the experiment using a multi-class classification approach. The combination of various computer vision techniques with augmented reality is indeed promising and offers a better user experience.
Deep hypersphere embedding for real-time face recognitionTELKOMNIKA JOURNAL
With the advancement of human-computer interaction capabilities of robots, computer vision surveillance systems involving security yields a large impact in the research industry by helping in digitalization of certain security processes. Recognizing a face in the computer vision involves identification and classification of which faces belongs to the same person by means of comparing face embedding vectors. In an organization that has a large and diverse labelled dataset on a large number of epoch, oftentimes, creates a training difficulties involving incompatibility in different versions of face embedding that leads to poor face recognition accuracy. In this paper, we will design and implement robotic vision security surveillance system incorporating hybrid combination of MTCNN for face detection, and FaceNet as the unified embedding for face recognition and clustering.
A hybrid approach for face recognition using a convolutional neural network c...IAESIJAI
Facial recognition technology has been used in many fields such as security,
biometric identification, robotics, video surveillance, health, and commerce
due to its ease of implementation and minimal data processing time.
However, this technology is influenced by the presence of variations such as
pose, lighting, or occlusion. In this paper, we propose a new approach to
improve the accuracy rate of face recognition in the presence of variation or
occlusion, by combining feature extraction with a histogram of oriented
gradient (HOG), scale invariant feature transform (SIFT), Gabor, and the
Canny contour detector techniques, as well as a convolutional neural
network (CNN) architecture, tested with several combinations of the
activation function used (Softmax and Segmoïd) and the optimization
algorithm used during training (adam, Adamax, RMSprop, and stochastic
gradient descent (SGD)). For this, a preprocessing was performed on two
databases of our database of faces (ORL) and Sheffield faces used, then we
perform a feature extraction operation with the mentioned techniques and
then pass them to our used CNN architecture. The results of our simulations
show a high performance of the SIFT+CNN combination, in the case of the
presence of variations with an accuracy rate up to 100%.
Extraction of image resampling using correlation aware convolution neural ne...IJECEIAES
Detecting hybrid tampering attacks in an image is extremely difficult; especially when copy-clone tampered segments exhibit identical illumination and contrast level about genuine objects. The existing method fails to detect tampering when the image undergoes hybrid transformation such as scaling, rotation, compression, and also fails to detect under smallsmooth tampering. The existing resampling feature extraction using the Deep learning techniques fails to obtain a good correlation among neighboring pixels in both horizontal and vertical directions. This work presents correlation aware convolution neural network (CA-CNN) for extracting resampling features for detecting hybrid tampering attacks. Here the image is resized for detecting tampering under a small-smooth region. The CA-CNN is composed of a three-layer horizontal, vertical, and correlated layer. The correlated layer is used for obtaining correlated resampling feature among horizontal sequence and vertical sequence. Then feature is aggregated and the descriptor is built. An experiment is conducted to evaluate the performance of the CA-CNN model over existing tampering detection methodologies considering the various datasets. From the result achieved it can be seen the CA-CNN is efficient considering various distortions and post-processing attacks such joint photographic expert group (JPEG) compression, and scaling. This model achieves much better accuracies, recall, precision, false positive rate (FPR), and F-measure compared existing methodologies.
Machine learning based augmented reality for improved learning application th...IJECEIAES
Detection of objects and their location in an image are important elements of current research in computer vision. In May 2020, Meta released its state-ofthe-art object-detection model based on a transformer architecture called detection transformer (DETR). There are several object-detection models such as region-based convolutional neural network (R-CNN), you only look once (YOLO) and single shot detectors (SSD), but none have used a transformer to accomplish this task. These models mentioned earlier, use all sorts of hyperparameters and layers. However, the advantages of using a transformer pattern make the architecture simple and easy to implement. In this paper, we determine the name of a chemical experiment through two steps: firstly, by building a DETR model, trained on a customized dataset, and then integrate it into an augmented reality mobile application. By detecting the objects used during the realization of an experiment, we can predict the name of the experiment using a multi-class classification approach. The combination of various computer vision techniques with augmented reality is indeed promising and offers a better user experience.
Deep hypersphere embedding for real-time face recognitionTELKOMNIKA JOURNAL
With the advancement of human-computer interaction capabilities of robots, computer vision surveillance systems involving security yields a large impact in the research industry by helping in digitalization of certain security processes. Recognizing a face in the computer vision involves identification and classification of which faces belongs to the same person by means of comparing face embedding vectors. In an organization that has a large and diverse labelled dataset on a large number of epoch, oftentimes, creates a training difficulties involving incompatibility in different versions of face embedding that leads to poor face recognition accuracy. In this paper, we will design and implement robotic vision security surveillance system incorporating hybrid combination of MTCNN for face detection, and FaceNet as the unified embedding for face recognition and clustering.
A hybrid approach for face recognition using a convolutional neural network c...IAESIJAI
Facial recognition technology has been used in many fields such as security,
biometric identification, robotics, video surveillance, health, and commerce
due to its ease of implementation and minimal data processing time.
However, this technology is influenced by the presence of variations such as
pose, lighting, or occlusion. In this paper, we propose a new approach to
improve the accuracy rate of face recognition in the presence of variation or
occlusion, by combining feature extraction with a histogram of oriented
gradient (HOG), scale invariant feature transform (SIFT), Gabor, and the
Canny contour detector techniques, as well as a convolutional neural
network (CNN) architecture, tested with several combinations of the
activation function used (Softmax and Segmoïd) and the optimization
algorithm used during training (adam, Adamax, RMSprop, and stochastic
gradient descent (SGD)). For this, a preprocessing was performed on two
databases of our database of faces (ORL) and Sheffield faces used, then we
perform a feature extraction operation with the mentioned techniques and
then pass them to our used CNN architecture. The results of our simulations
show a high performance of the SIFT+CNN combination, in the case of the
presence of variations with an accuracy rate up to 100%.
A deep learning based stereo matching model for autonomous vehicleIAESIJAI
Autonomous vehicle is one the prominent area of research in computer
vision. In today’s AI world, the concept of autonomous vehicles has become
popular largely to avoid accidents due to negligence of driver. Perceiving the
depth of the surrounding region accurately is a challenging task in
autonomous vehicles. Sensors like light detection and ranging can be used
for depth estimation but these sensors are expensive. Hence stereo matching
is an alternate solution to estimate the depth. The main difficulties observed
in stereo matching is to minimize mismatches in the ill-posed regions, like
occluded, texture less and discontinuous regions. This paper presents an
efficient deep stereo matching technique for estimating disparity map from
stereo images in ill-posed regions. The images from Middlebury stereo data
set are used to assess the efficacy of the model proposed. The experimental
outcome dipicts that the proposed model generates reliable results in the
occluded, texture less and discontinuous regions as compared to the existing
techniques.
Semantic Concept Detection in Video Using Hybrid Model of CNN and SVM Classif...CSCJournals
In today's era of digitization and fast internet, many video are uploaded on websites, a mechanism is required to access this video accurately and efficiently. Semantic concept detection achieve this task accurately and is used in many application like multimedia annotation, video summarization, annotation, indexing and retrieval. Video retrieval based on semantic concept is efficient and challenging research area. Semantic concept detection bridges the semantic gap between low level extraction of features from key-frame or shot of video and high level interpretation of the same as semantics. Semantic Concept detection automatically assigns labels to video from predefined vocabulary. This task is considered as supervised machine learning problem. Support vector machine (SVM) emerged as default classifier choice for this task. But recently Deep Convolutional Neural Network (CNN) has shown exceptional performance in this area. CNN requires large dataset for training. In this paper, we present framework for semantic concept detection using hybrid model of SVM and CNN. Global features like color moment, HSV histogram, wavelet transform, grey level co-occurrence matrix and edge orientation histogram are selected as low level features extracted from annotated groundtruth video dataset of TRECVID. In second pipeline, deep features are extracted using pretrained CNN. Dataset is partitioned in three segments to deal with data imbalance issue. Two classifiers are separately trained on all segments and fusion of scores is performed to detect the concepts in test dataset. The system performance is evaluated using Mean Average Precision for multi-label dataset. The performance of the proposed framework using hybrid model of SVM and CNN is comparable to existing approaches.
CNN FEATURES ARE ALSO GREAT AT UNSUPERVISED CLASSIFICATION cscpconf
This paper aims at providing insight on the transferability of deep CNN features to
unsupervised problems. We study the impact of different pretrained CNN feature extractors on
the problem of image set clustering for object classification as well as fine-grained
classification. We propose a rather straightforward pipeline combining deep-feature extraction
using a CNN pretrained on ImageNet and a classic clustering algorithm to classify sets of
images. This approach is compared to state-of-the-art algorithms in image-clustering and
provides better results. These results strengthen the belief that supervised training of deep CNN
on large datasets, with a large variability of classes, extracts better features than most carefully
designed engineering approaches, even for unsupervised tasks. We also validate our approach
on a robotic application, consisting in sorting and storing objects smartly based on clustering
Comparative Performance of Image Scrambling in Transform Domain using Sinusoi...CSCJournals
With the rapid development of technology, and the popularization of internet, communication is been greatly promoted. The communication is not limited only to information but also includes multimedia information like digital Images. Therefore, the security of digital images has become a very important and practical issue, and appropriate security technology is used for those digital images containing confidential or private information especially. In this paper a novel approach of Image scrambling has been proposed which includes both spatial as well as Transform domain. Experimental results prove that correlation obtained in scrambled images is much lesser then the one obtained in transformed images.
Robust Malware Detection using Residual Attention NetworkShamika Ganesan
In this paper, we explore the use of an attention based mechanism known as Residual Attention for malware detection and compare this with existing CNN based methods and conventional Machine Learning algorithms with the help of GIST features. The proposed method outperformed traditional malware detection methods which use Machine Learning and CNN based Deep Learning algorithms, by demonstrating an accuracy of 99.25%.
This paper has been accepted in the International Conference of Consumer Electronics (ICCE 2021).
In this paper, an attempt has been made to extract texture
features from facial images using an improved method of
Illumination Invariant Feature Descriptor. The proposed local
ternary Pattern based feature extractor viz., Steady Illumination
Local Ternary Pattern (SIcLTP) has been used to extract texture
features from Indian face database. The similarity matching
between two extracted feature sets has been obtained using Zero
Mean Sum of Squared Differences (ZSSD). The RGB facial images
are first converted into the YIQ colour space to reduce the
redundancy of the RGB images. The result obtained has been
analysed using Receiver Operating Characteristic curve, and is
found to be promising. Finally the results are validated with
standard local binary pattern (LBP) extractor.
Facial recognition based on enhanced neural networkIAESIJAI
Accurate automatic face recognition (FR) has only become a practical goal of biometrics research in recent years. Detection and recognition are the primary steps for identifying faces in this research, and The Viola-Jones algorithm implements to discover faces in images. This paper presents a neural network solution called modify bidirectional associative memory (MBAM). The basic idea is to recognize the image of a human's face, extract the face image, enter it into the MBAM, and identify it. The output ID for the face image from the network should be similar to the ID for the image entered previously in the training phase. The tests have conducted using the suggested model using 100 images. Results show that FR accuracy is 100% for all images used, and the accuracy after adding noise is the proportions that differ between the images used according to the noise ratio. Recognition results for the mobile camera images were more satisfactory than those for the Face94 dataset.
Development of 3D convolutional neural network to recognize human activities ...journalBEEI
Human activity recognition (HAR) is recently used in numerous applications including smart homes to monitor human behavior, automate homes according to human activities, entertainment, falling detection, violence detection, and people care. Vision-based recognition is the most powerful method widely used in HAR systems implementation due to its characteristics in recognizing complex human activities. This paper addresses the design of a 3D convolutional neural network (3D-CNN) model that can be used in smart homes to identify several numbers of activities. The model is trained using KTH dataset that contains activities like (walking, running, jogging, handwaving handclapping, boxing). Despite the challenges of this method due to the effectiveness of the lamination, background variation, and human body variety, the proposed model reached an accuracy of 93.33%. The model was implemented, trained and tested using moderate computation machine and the results show that the proposal was successfully capable to recognize human activities with reasonable computations.
Residual balanced attention network for real-time traffic scene semantic segm...IJECEIAES
Intelligent transportation systems (ITS) are among the most focused research in this century. Actually, autonomous driving provides very advanced tasks in terms of road safety monitoring which include identifying dangers on the road and protecting pedestrians. In the last few years, deep learning (DL) approaches and especially convolutional neural networks (CNNs) have been extensively used to solve ITS problems such as traffic scene semantic segmentation and traffic signs classification. Semantic segmentation is an important task that has been addressed in computer vision (CV). Indeed, traffic scene semantic segmentation using CNNs requires high precision with few computational resources to perceive and segment the scene in real-time. However, we often find related work focusing only on one aspect, the precision, or the number of computational parameters. In this regard, we propose RBANet, a robust and lightweight CNN which uses a new proposed balanced attention module, and a new proposed residual module. Afterward, we have simulated our proposed RBANet using three loss functions to get the best combination using only 0.74M parameters. The RBANet has been evaluated on CamVid, the most used dataset in semantic segmentation, and it has performed well in terms of parameters’ requirements and precision compared to related work.
Slantlet transform used for faults diagnosis in robot armIJEECSIAES
The robot arm systems are the most target systems in the fields of faults detection and diagnosis which are electrical and the mechanical systems in many fields. Fault detection and diagnosis study is presented for two robot arms. The disturbance due to the faults at robot's joints causes oscillations at the tip of the robot arm. The acceleration in multi-direction is analysed to extract the features of the faults. Simulations for planar and space robots are presented. Two types of feature (faults) detection methods are used in this paper. The first one is the discrete wavelet transform, which is applied in many research's works before. The second type, is the Slantlet transform, which represents an improved model of the discrete wavelet transform. The multi-layer perceptron artificial neural network is used for the purpose of faults allocation and classification. According to the obtained results, the Slantlet transform with the multi-layer perceptron artificial neural network appear to possess best performance (4.7088e-05), lower consuming time (71.017308 sec) and higher accuracy (100%) than the results obtained when applying discrete wavelet transform and artificial neural network for the same purpose.
The IoT and registration of MRI brain diagnosis based on genetic algorithm an...IJEECSIAES
The technology of the multimodal brain image registration is the key method for accurate and rapid diagnosis and treatment of brain diseases. For achieving high-resolution image registration, a fast sub pixel registration algorithm is used based on single-step discrete wavelet transform (DWT) combined with phase convolution neural network (CNN) to classify the registration of brain tumors. In this work apply the genetic algorithm and CNN clasifcation in registration of magnetic resonance imaging (MRI) image. This approach follows eight steps, reading the source of MRI brain image and loading the reference image, enhencment all MRI images by bilateral filter, transforming DWT image by applying the DWT2, evaluating (fitness function) each MRI image by using entropy, applying the genetic algorithm, by selecting the two images based on rollout wheel and crossover of the two images, the CNN classify the result of subtraction to normal or abnormal, “in the eighth one,” the Arduino and global system for mobile (GSM) 8080 are applied to send the message to patient. The proposed model is tested on MRI Medical City Hospital in Baghdad database consist 550 normal and 350 abnormal and split to 80% training and 20 testing, the proposed model result achieves the 98.8% accuracy.
More Related Content
Similar to Efficient resampling features and convolution neural network model for image forgery detection
A deep learning based stereo matching model for autonomous vehicleIAESIJAI
Autonomous vehicle is one the prominent area of research in computer
vision. In today’s AI world, the concept of autonomous vehicles has become
popular largely to avoid accidents due to negligence of driver. Perceiving the
depth of the surrounding region accurately is a challenging task in
autonomous vehicles. Sensors like light detection and ranging can be used
for depth estimation but these sensors are expensive. Hence stereo matching
is an alternate solution to estimate the depth. The main difficulties observed
in stereo matching is to minimize mismatches in the ill-posed regions, like
occluded, texture less and discontinuous regions. This paper presents an
efficient deep stereo matching technique for estimating disparity map from
stereo images in ill-posed regions. The images from Middlebury stereo data
set are used to assess the efficacy of the model proposed. The experimental
outcome dipicts that the proposed model generates reliable results in the
occluded, texture less and discontinuous regions as compared to the existing
techniques.
Semantic Concept Detection in Video Using Hybrid Model of CNN and SVM Classif...CSCJournals
In today's era of digitization and fast internet, many video are uploaded on websites, a mechanism is required to access this video accurately and efficiently. Semantic concept detection achieve this task accurately and is used in many application like multimedia annotation, video summarization, annotation, indexing and retrieval. Video retrieval based on semantic concept is efficient and challenging research area. Semantic concept detection bridges the semantic gap between low level extraction of features from key-frame or shot of video and high level interpretation of the same as semantics. Semantic Concept detection automatically assigns labels to video from predefined vocabulary. This task is considered as supervised machine learning problem. Support vector machine (SVM) emerged as default classifier choice for this task. But recently Deep Convolutional Neural Network (CNN) has shown exceptional performance in this area. CNN requires large dataset for training. In this paper, we present framework for semantic concept detection using hybrid model of SVM and CNN. Global features like color moment, HSV histogram, wavelet transform, grey level co-occurrence matrix and edge orientation histogram are selected as low level features extracted from annotated groundtruth video dataset of TRECVID. In second pipeline, deep features are extracted using pretrained CNN. Dataset is partitioned in three segments to deal with data imbalance issue. Two classifiers are separately trained on all segments and fusion of scores is performed to detect the concepts in test dataset. The system performance is evaluated using Mean Average Precision for multi-label dataset. The performance of the proposed framework using hybrid model of SVM and CNN is comparable to existing approaches.
CNN FEATURES ARE ALSO GREAT AT UNSUPERVISED CLASSIFICATION cscpconf
This paper aims at providing insight on the transferability of deep CNN features to
unsupervised problems. We study the impact of different pretrained CNN feature extractors on
the problem of image set clustering for object classification as well as fine-grained
classification. We propose a rather straightforward pipeline combining deep-feature extraction
using a CNN pretrained on ImageNet and a classic clustering algorithm to classify sets of
images. This approach is compared to state-of-the-art algorithms in image-clustering and
provides better results. These results strengthen the belief that supervised training of deep CNN
on large datasets, with a large variability of classes, extracts better features than most carefully
designed engineering approaches, even for unsupervised tasks. We also validate our approach
on a robotic application, consisting in sorting and storing objects smartly based on clustering
Comparative Performance of Image Scrambling in Transform Domain using Sinusoi...CSCJournals
With the rapid development of technology, and the popularization of internet, communication is been greatly promoted. The communication is not limited only to information but also includes multimedia information like digital Images. Therefore, the security of digital images has become a very important and practical issue, and appropriate security technology is used for those digital images containing confidential or private information especially. In this paper a novel approach of Image scrambling has been proposed which includes both spatial as well as Transform domain. Experimental results prove that correlation obtained in scrambled images is much lesser then the one obtained in transformed images.
Robust Malware Detection using Residual Attention NetworkShamika Ganesan
In this paper, we explore the use of an attention based mechanism known as Residual Attention for malware detection and compare this with existing CNN based methods and conventional Machine Learning algorithms with the help of GIST features. The proposed method outperformed traditional malware detection methods which use Machine Learning and CNN based Deep Learning algorithms, by demonstrating an accuracy of 99.25%.
This paper has been accepted in the International Conference of Consumer Electronics (ICCE 2021).
In this paper, an attempt has been made to extract texture
features from facial images using an improved method of
Illumination Invariant Feature Descriptor. The proposed local
ternary Pattern based feature extractor viz., Steady Illumination
Local Ternary Pattern (SIcLTP) has been used to extract texture
features from Indian face database. The similarity matching
between two extracted feature sets has been obtained using Zero
Mean Sum of Squared Differences (ZSSD). The RGB facial images
are first converted into the YIQ colour space to reduce the
redundancy of the RGB images. The result obtained has been
analysed using Receiver Operating Characteristic curve, and is
found to be promising. Finally the results are validated with
standard local binary pattern (LBP) extractor.
Facial recognition based on enhanced neural networkIAESIJAI
Accurate automatic face recognition (FR) has only become a practical goal of biometrics research in recent years. Detection and recognition are the primary steps for identifying faces in this research, and The Viola-Jones algorithm implements to discover faces in images. This paper presents a neural network solution called modify bidirectional associative memory (MBAM). The basic idea is to recognize the image of a human's face, extract the face image, enter it into the MBAM, and identify it. The output ID for the face image from the network should be similar to the ID for the image entered previously in the training phase. The tests have conducted using the suggested model using 100 images. Results show that FR accuracy is 100% for all images used, and the accuracy after adding noise is the proportions that differ between the images used according to the noise ratio. Recognition results for the mobile camera images were more satisfactory than those for the Face94 dataset.
Development of 3D convolutional neural network to recognize human activities ...journalBEEI
Human activity recognition (HAR) is recently used in numerous applications including smart homes to monitor human behavior, automate homes according to human activities, entertainment, falling detection, violence detection, and people care. Vision-based recognition is the most powerful method widely used in HAR systems implementation due to its characteristics in recognizing complex human activities. This paper addresses the design of a 3D convolutional neural network (3D-CNN) model that can be used in smart homes to identify several numbers of activities. The model is trained using KTH dataset that contains activities like (walking, running, jogging, handwaving handclapping, boxing). Despite the challenges of this method due to the effectiveness of the lamination, background variation, and human body variety, the proposed model reached an accuracy of 93.33%. The model was implemented, trained and tested using moderate computation machine and the results show that the proposal was successfully capable to recognize human activities with reasonable computations.
Residual balanced attention network for real-time traffic scene semantic segm...IJECEIAES
Intelligent transportation systems (ITS) are among the most focused research in this century. Actually, autonomous driving provides very advanced tasks in terms of road safety monitoring which include identifying dangers on the road and protecting pedestrians. In the last few years, deep learning (DL) approaches and especially convolutional neural networks (CNNs) have been extensively used to solve ITS problems such as traffic scene semantic segmentation and traffic signs classification. Semantic segmentation is an important task that has been addressed in computer vision (CV). Indeed, traffic scene semantic segmentation using CNNs requires high precision with few computational resources to perceive and segment the scene in real-time. However, we often find related work focusing only on one aspect, the precision, or the number of computational parameters. In this regard, we propose RBANet, a robust and lightweight CNN which uses a new proposed balanced attention module, and a new proposed residual module. Afterward, we have simulated our proposed RBANet using three loss functions to get the best combination using only 0.74M parameters. The RBANet has been evaluated on CamVid, the most used dataset in semantic segmentation, and it has performed well in terms of parameters’ requirements and precision compared to related work.
Similar to Efficient resampling features and convolution neural network model for image forgery detection (20)
Slantlet transform used for faults diagnosis in robot armIJEECSIAES
The robot arm systems are the most target systems in the fields of faults detection and diagnosis which are electrical and the mechanical systems in many fields. Fault detection and diagnosis study is presented for two robot arms. The disturbance due to the faults at robot's joints causes oscillations at the tip of the robot arm. The acceleration in multi-direction is analysed to extract the features of the faults. Simulations for planar and space robots are presented. Two types of feature (faults) detection methods are used in this paper. The first one is the discrete wavelet transform, which is applied in many research's works before. The second type, is the Slantlet transform, which represents an improved model of the discrete wavelet transform. The multi-layer perceptron artificial neural network is used for the purpose of faults allocation and classification. According to the obtained results, the Slantlet transform with the multi-layer perceptron artificial neural network appear to possess best performance (4.7088e-05), lower consuming time (71.017308 sec) and higher accuracy (100%) than the results obtained when applying discrete wavelet transform and artificial neural network for the same purpose.
The IoT and registration of MRI brain diagnosis based on genetic algorithm an...IJEECSIAES
The technology of the multimodal brain image registration is the key method for accurate and rapid diagnosis and treatment of brain diseases. For achieving high-resolution image registration, a fast sub pixel registration algorithm is used based on single-step discrete wavelet transform (DWT) combined with phase convolution neural network (CNN) to classify the registration of brain tumors. In this work apply the genetic algorithm and CNN clasifcation in registration of magnetic resonance imaging (MRI) image. This approach follows eight steps, reading the source of MRI brain image and loading the reference image, enhencment all MRI images by bilateral filter, transforming DWT image by applying the DWT2, evaluating (fitness function) each MRI image by using entropy, applying the genetic algorithm, by selecting the two images based on rollout wheel and crossover of the two images, the CNN classify the result of subtraction to normal or abnormal, “in the eighth one,” the Arduino and global system for mobile (GSM) 8080 are applied to send the message to patient. The proposed model is tested on MRI Medical City Hospital in Baghdad database consist 550 normal and 350 abnormal and split to 80% training and 20 testing, the proposed model result achieves the 98.8% accuracy.
A comparative study for the assessment of Ikonos satellite image-fusion techn...IJEECSIAES
Image-fusion provide users with detailed information about the urban and rural environment, which is useful for applications such as urban planning and management when higher spatial resolution images are not available. There are different image fusion methods. This paper implements, evaluates, and compares six satellite image-fusion methods, namely wavelet 2D-M transform, gram schmidt, high-frequency modulation, high pass filter (HPF) transform, simple mean value, and PCA. An Ikonos image (PanchromaticPAN and multispectral-MULTI) showing the northwest of Bogotá (Colombia) is used to generate six fused images: MULTIWavelet 2D-M, MULTIG-S, MULTIMHF, MULTIHPF, MULTISMV, and MULTIPCA. In order to assess the efficiency of the six image-fusion methods, the resulting images were evaluated in terms of both spatial quality and spectral quality. To this end, four metrics were applied, namely the correlation index, erreur relative globale adimensionnelle de synthese (ERGAS), relative average spectral error (RASE) and the Q index. The best results were obtained for the MULTISMV image, which exhibited spectral correlation higher than 0.85, a Q index of 0.84, and the highest scores in spectral assessment according to ERGAS and RASE, 4.36% and 17.39% respectively.
Elliptical curve cryptography image encryption scheme with aid of optimizatio...IJEECSIAES
Image encryption enables users to safely transmit digital photographs via a wireless medium while maintaining enhanced anonymity and validity. Numerous studies are being conducted to strengthen picture encryption systems. Elliptical curve cryptography (ECC) is an effective tool for safely transferring images and recovering them at the receiver end in asymmetric cryptosystems. This method's key generation generates a public and private key pair that is used to encrypt and decrypt a picture. They use a public key to encrypt the picture before sending it to the intended user. When the receiver receives the image, they use their private key to decrypt it. This paper proposes an ECC-dependent image encryption scheme utilizing an enhancement strategy based on the gravitational search algorithm (GSA) algorithm. The private key generation step of the ECC system uses a GSAbased optimization process to boost the efficiency of picture encryption. The image's output is used as a health attribute in the optimization phase, such as the peak signal to noise ratio (PSNR) value, which demonstrates the efficacy of the proposed approach. As comparison to the ECC method, it has been discovered that the suggested encryption scheme offers better optimal PSNR values.
Design secure multi-level communication system based on duffing chaotic map a...IJEECSIAES
Cryptography and steganography are among the most important sciences that have been properly used to keep confidential data from potential spies and hackers. They can be used separately or together. Encryption involves the basic principle of instantaneous conversion of valuable information into a specific form that unauthorized persons will not understand to decrypt it. While steganography is the science of embedding confidential data inside a cover, in a way that cannot be recognized or seen by the human eye. This paper presents a high-resolution chaotic approach applied to images that hide information. A more secure and reliable system is designed to properly include confidential data transmitted through transmission channels. This is done by working the use of encryption and steganography together. This work proposed a new method that achieves a very high level of hidden information based on non-uniform systems by generating a random index vector (RIV) for hidden data within least significant bit (LSB) image pixels. This method prevents the reduction of image quality. The simulation results also show that the peak signal to noise ratio (PSNR) is up to 74.87 dB and the mean square error (MSE) values is up to 0.0828, which sufficiently indicates the effectiveness of the proposed algorithm.
A new function of stereo matching algorithm based on hybrid convolutional neu...IJEECSIAES
This paper proposes a new hybrid method between the learning-based and handcrafted methods for a stereo matching algorithm. The main purpose of the stereo matching algorithm is to produce a disparity map. This map is essential for many applications, including three-dimensional (3D) reconstruction. The raw disparity map computed by a convolutional neural network (CNN) is still prone to errors in the low texture region. The algorithm is set to improve the matching cost computation stage with hybrid CNN-based combined with truncated directional intensity computation. The difference in truncated directional intensity value is employed to decrease radiometric errors. The proposed method’s raw matching cost went through the cost aggregation step using the bilateral filter (BF) to improve accuracy. The winner-take-all (WTA) optimization uses the aggregated cost volume to produce an initial disparity map. Finally, a series of refinement processes enhance the initial disparity map for a more accurate final disparity map. This paper verified the performance of the algorithm using the Middlebury online stereo benchmarking system. The proposed algorithm achieves the objective of generating a more accurate and smooth disparity map with different depths at low texture regions through better matching cost quality.
An internet of things-based automatic brain tumor detection systemIJEECSIAES
Due to the advances in information and communication technologies, the usage of the internet of things (IoT) has reached an evolutionary process in the development of the modern health care environment. In the recent human health care analysis system, the amount of brain tumor patients has increased severely and placed in the 10th position of the leading cause of death. Previous state-of-the-art techniques based on magnetic resonance imaging (MRI) faces challenges in brain tumor detection as it requires accurate image segmentation. A wide variety of algorithms were developed earlier to classify MRI images which are computationally very complex and expensive. In this paper, a cost-effective stochastic method for the automatic detection of brain tumors using the IoT is proposed. The proposed system uses the physical activities of the brain to detect brain tumors. To track the daily brain activities, a portable wrist band named Mi Band 2, temperature, and blood pressure monitoring sensors embedded with Arduino-Uno are used and the system achieved an accuracy of 99.3%. Experimental results show the effectiveness of the designed method in detecting brain tumors automatically and produce better accuracy in comparison to previous approaches.
A YOLO and convolutional neural network for the detection and classification ...IJEECSIAES
The developing of deep learning systems that used for chronic diseases diagnosing is challenge. Furthermore, the localization and identification of objects like white blood cells (WBCs) in leukemia without preprocessing or traditional hand segmentation of cells is a challenging matter due to irregular and distorted of nucleus. This paper proposed a system for computer-aided detection depend completely on deep learning with three models computeraided detection (CAD3) to detect and classify three types of WBC which is fundamentals of leukemia diagnosing. The system used modified you only look once (YOLO v2) algorithm and convolutional neural network (CNN). The proposed system trained and evaluated on dataset created and prepared specially for the addressed problem without any traditional segmentation or preprocessing on microscopic images. The study proved that dividing of addressed problem into sub-problems will achieve better performance and accuracy. Furthermore, the results show that the CAD3 achieved an average precision (AP) up to 96% in the detection of leukocytes and accuracy 94.3% in leukocytes classification. Moreover, the CAD3 gives report contain a complete information of WBC. Finally, the CAD3 proved its efficiency on the other dataset such as acute lymphoblastic leukemia image database (ALL-IBD1) and blood cell count dataset (BCCD).
Development of smart machine for sorting of deceased onionsIJEECSIAES
Today, we are thinking to raise farmer’s income through various means and measures. Implementation of new crop patterns, technology inclusion and promoting the eshtablishment of numerous agro processing industries will play a major role in agriculture sector. The labour issue is also one of the main concerns in many of the agricultural activities. In this paper we propose a technological evolvement in onion detection process, where we apply image processing and sensory mechanism to identify sprouted and rotten onions respectively. This will yield to quick, accurate and prompt supply of goods to the market, irrespective of lack of consistent but costly manpower. The efficiency of this prototype in identifying the sprouted onions with the help of camera is observed to be upto 87% and also the response of Gas sensing system in detecting rooten onions under prescribed chamber dimensions is analysed and obtained encouraging results.
State and fault estimation based on fuzzy observer for a class of Takagi-Suge...IJEECSIAES
Singular nonlinear systems have received wide attention in recent years, and can be found in various applications of engineering practice. On the basis of the Takagi-Sugeno (T-S) formalism, which represents a powerful tool allowing the study and the treatment of nonlinear systems, many control and diagnostic problems have been treated in the literature. In this work, we aim to present a new approach making it possible to estimate simultaneously both non-measurable states and unknown faults in the actuators and sensors for a class of continuous-time Takagi-Sugeno singular model (CTSSM). Firstly, the considered class of CTSSM is represented in the case of premise variables which are non-measurable, and is subjected to actuator and sensor faults. Secondly, the suggested observer is synthesized based on the decomposition approach. Next, the observer’s gain matrices are determined using the Lyapunov theory and the constraints are defined as linear matrix inequalities (LMIs). Finally, a numerical simulation on an application example is given to demonstrate the usefulness and the good performance of the proposed dynamic system.
Hunting strategy for multi-robot based on wolf swarm algorithm and artificial...IJEECSIAES
The cooperation and coordination in multi-robot systems is a popular topic in the field of robotics and artificial intelligence, thanks to its important role in solving problems that are better solved by several robots compared to a single robot. Cooperative hunting is one of the important problems that exist in many areas such as military and industry, requiring cooperation between robots in order to accomplish the hunting process effectively. This paper proposed a cooperative hunting strategy for a multi-robot system based on wolf swarm algorithm (WSA) and artificial potential field (APF) in order to hunt by several robots a dynamic target whose behavior is unexpected. The formation of the robots within the multi-robot system contains three types of roles: the leader, the follower, and the antagonist. Each role is characterized by a different cognitive behavior. The robots arrive at the hunting point accurately and rapidly while avoiding static and dynamic obstacles through the artificial potential field algorithm to hunt the moving target. Simulation results are given in this paper to demonstrate the validity and the effectiveness of the proposed strategy.
Agricultural harvesting using integrated robot systemIJEECSIAES
In today's competitive world, robot designs are developed to simplify and improve quality wherever necessary. The rise in technology and modernization has led people from the unskilled sector to shift to the skilled sector. The agricultural sector's solution for harvesting fruits and vegetables is manual labor and a few other agro bots that are expensive and have various limitations when it comes to harvesting. Although robots present may achieve harvesting, the affordability of such designs may not be possible by small and medium-scale producers. The integrated robot system is designed to solve this problem, and when compared with the existing manual methods, this seems to be the most cost-effective, efficient, and viable solution. The robot uses deep learning for image detection, and the object is acquired using robotic manipulators. The robot uses a Cartesian and articulated configuration to perform the picking action. In the end, the robot is operated where carrots and cantaloupes were harvested. The data of the harvested crops are used to arrive at the conclusion of the robot's accuracy.
Characterization of silicon tunnel field effect transistor based on charge pl...IJEECSIAES
The aim of the proposed paper is an analytical model and realization of the characteristics for tunnel field-effect transistor (TFET) based on charge plasma (CP). One of the most applications of the TFET device which operates based on CP technique is the biosensor. CP-TFET is to be used as an effective device to detect the uncharged molecules of the bio-sample solution. Charge plasma is one of some techniques that recently invited to induce charge carriers inside the devices. In this proposed paper we use a high work function in the source (ϕ=5.93 eV) to induce hole charges and we use a lower work function in drain (ϕ=3.90 eV) to induce electron charges. Many electrical characterizations in this paper are considered to study the performance of this device like a current drain (ID) versus voltage gate (Vgs), ION/IOFF ratio, threshold voltage (VT) transconductance (gm), and subthreshold swing (SS). The signification of this paper comes into view enhancement the performance of the device. Results show that high dielectric (K=12), oxide thickness (Tox=1 nm), channel length (Lch=42 nm), and higher work function for the gate (ϕ=4.5 eV) tend to best charge plasma silicon tunnel field-effect transistor characterization.
A new smart approach of an efficient energy consumption management by using a...IJEECSIAES
Many consumers of electric power have excesses in their electric power consumptions that exceed the permissible limit by the electrical power distribution stations, and then we proposed a validation approach that works intelligently by applying machine learning (ML) technology to teach electrical consumers how to properly consume without wasting energy expended. The validation approach is one of a large combination of intelligent processes related to energy consumption which is called the efficient energy consumption management (EECM) approaches, and it connected with the internet of things (IoT) technology to be linked to Google Firebase Cloud where a utility center used to check whether the consumption of the efficient energy is satisfied. It divides the measured data for actual power (Ap) of the electrical model into two portions: the training portion is selected for different maximum actual powers, and the validation portion is determined based on the minimum output power consumption and then used for comparison with the actual required input power. Simulation results show the energy expenditure problem can be solved with good accuracy in energy consumption by reducing the maximum rate (Ap) in a given time (24) hours for a single house, as well as electricity’s bill cost, is reduced.
Parameter selection in data-driven fault detection and diagnosis of the air c...IJEECSIAES
Data-driven fault detection and diagnosis system (FDD) has been proven as simple yet powerful to identify soft and abrupt faults in the air conditioning system, leading to energy saving. However, the challenge is to obtain reliable operation data from the actual building. Therefore, a lab-scaled centralized chilled water air conditioning system was successfully developed in this paper. All necessary sensors were installed to generate reliable operation data for the data-driven FDD. Nevertheless, if a practical system is considered, the number of sensors required would be extensive as it depends on the number of rooms in the building. Hence, parameters impact in the dataset were also investigated to identify critical parameters for fault classifications. The analysis results had identified four critical parameters for data-driven FDD: the rooms' temperature (TTCx), supplied chilled water temperature (TCHWS), supplied chilled water flow rate (VCHWS) and supplied cooled water temperature (TCWS). Results showed that the data-driven FDD successfully diagnosed all six conditions correctly with the proposed parameters for more than 92.3% accuracy; only 0.6-3.4% differed from the original dataset's accuracy. Therefore, the proposed parameters can reduce the number of sensors used for practical buildings, thus reducing installation costs without compromising the FDD accuracy.
Electrical load forecasting through long short term memoryIJEECSIAES
For a power supplier, meeting demand-supply equilibrium is of utmost importance. Electrical energy must be generated according to demand, as a
large amount of electrical energy cannot be stored. For the proper
functioning of a power supply system, an adequate model for predicting load is a necessity. In the present world, in almost every industry, whether it be healthcare, agriculture, and consulting, growing digitization and automation is a prominent feature. As a result, large sets of data related to these industries are being generated, which when subjected to rigorous analysis,
yield out-of-the-box methods to optimize the business and services offered. This paper aims to ascertain the viability of long short term memory (LSTM)
neural networks, a recurrent neural network capable of handling both longterm and short-term dependencies of data sets, for predicting load that is to
be met by a Dispatch Center located in a major city. The result shows appreciable accuracy in forecasting future demand.
A simple faulted phase-based fault distance estimation algorithm for a loop d...IJEECSIAES
This paper presents a single ended faulted phase-based traveling wave fault localization algorithm for loop distribution grids which is that the sensor can
get many reflected signals from the fault point to face the complexity of localization. This localization algorithm uses a band pass filter to remove noise from the corrupted signal. The arriving times of the faulted phasebased filtered signals can be obtained by using phase-modal and discrete
wavelet transformations. The estimated fault distance can be calculated using the traveling wave method. The proposed algorithm presents detail
level analysis using three detail levels coefficients. The proposed algorithm is tested with MATLAB simulation single line to ground fault in a 10 kV grounded loop distribution system. The simulation result shows that the
faulted phase time delay can give better accuracy than using conventional time delays. The proposed algorithm can give fault distance estimation accuracy up to 99.7% with 30 dB contaminated signal-to-noise ratio (SNR)
for the nearest lines from the measured terminal.
NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...Amil Baba Dawood bangali
Contact with Dawood Bhai Just call on +92322-6382012 and we'll help you. We'll solve all your problems within 12 to 24 hours and with 101% guarantee and with astrology systematic. If you want to take any personal or professional advice then also you can call us on +92322-6382012 , ONLINE LOVE PROBLEM & Other all types of Daily Life Problem's.Then CALL or WHATSAPP us on +92322-6382012 and Get all these problems solutions here by Amil Baba DAWOOD BANGALI
#vashikaranspecialist #astrologer #palmistry #amliyaat #taweez #manpasandshadi #horoscope #spiritual #lovelife #lovespell #marriagespell#aamilbabainpakistan #amilbabainkarachi #powerfullblackmagicspell #kalajadumantarspecialist #realamilbaba #AmilbabainPakistan #astrologerincanada #astrologerindubai #lovespellsmaster #kalajaduspecialist #lovespellsthatwork #aamilbabainlahore#blackmagicformarriage #aamilbaba #kalajadu #kalailam #taweez #wazifaexpert #jadumantar #vashikaranspecialist #astrologer #palmistry #amliyaat #taweez #manpasandshadi #horoscope #spiritual #lovelife #lovespell #marriagespell#aamilbabainpakistan #amilbabainkarachi #powerfullblackmagicspell #kalajadumantarspecialist #realamilbaba #AmilbabainPakistan #astrologerincanada #astrologerindubai #lovespellsmaster #kalajaduspecialist #lovespellsthatwork #aamilbabainlahore #blackmagicforlove #blackmagicformarriage #aamilbaba #kalajadu #kalailam #taweez #wazifaexpert #jadumantar #vashikaranspecialist #astrologer #palmistry #amliyaat #taweez #manpasandshadi #horoscope #spiritual #lovelife #lovespell #marriagespell#aamilbabainpakistan #amilbabainkarachi #powerfullblackmagicspell #kalajadumantarspecialist #realamilbaba #AmilbabainPakistan #astrologerincanada #astrologerindubai #lovespellsmaster #kalajaduspecialist #lovespellsthatwork #aamilbabainlahore #Amilbabainuk #amilbabainspain #amilbabaindubai #Amilbabainnorway #amilbabainkrachi #amilbabainlahore #amilbabaingujranwalan #amilbabainislamabad
Immunizing Image Classifiers Against Localized Adversary Attacksgerogepatton
This paper addresses the vulnerability of deep learning models, particularly convolutional neural networks
(CNN)s, to adversarial attacks and presents a proactive training technique designed to counter them. We
introduce a novel volumization algorithm, which transforms 2D images into 3D volumetric representations.
When combined with 3D convolution and deep curriculum learning optimization (CLO), itsignificantly improves
the immunity of models against localized universal attacks by up to 40%. We evaluate our proposed approach
using contemporary CNN architectures and the modified Canadian Institute for Advanced Research (CIFAR-10
and CIFAR-100) and ImageNet Large Scale Visual Recognition Challenge (ILSVRC12) datasets, showcasing
accuracy improvements over previous techniques. The results indicate that the combination of the volumetric
input and curriculum learning holds significant promise for mitigating adversarial attacks without necessitating
adversary training.
Vaccine management system project report documentation..pdfKamal Acharya
The Division of Vaccine and Immunization is facing increasing difficulty monitoring vaccines and other commodities distribution once they have been distributed from the national stores. With the introduction of new vaccines, more challenges have been anticipated with this additions posing serious threat to the already over strained vaccine supply chain system in Kenya.
CFD Simulation of By-pass Flow in a HRSG module by R&R Consult.pptxR&R Consult
CFD analysis is incredibly effective at solving mysteries and improving the performance of complex systems!
Here's a great example: At a large natural gas-fired power plant, where they use waste heat to generate steam and energy, they were puzzled that their boiler wasn't producing as much steam as expected.
R&R and Tetra Engineering Group Inc. were asked to solve the issue with reduced steam production.
An inspection had shown that a significant amount of hot flue gas was bypassing the boiler tubes, where the heat was supposed to be transferred.
R&R Consult conducted a CFD analysis, which revealed that 6.3% of the flue gas was bypassing the boiler tubes without transferring heat. The analysis also showed that the flue gas was instead being directed along the sides of the boiler and between the modules that were supposed to capture the heat. This was the cause of the reduced performance.
Based on our results, Tetra Engineering installed covering plates to reduce the bypass flow. This improved the boiler's performance and increased electricity production.
It is always satisfying when we can help solve complex challenges like this. Do your systems also need a check-up or optimization? Give us a call!
Work done in cooperation with James Malloy and David Moelling from Tetra Engineering.
More examples of our work https://www.r-r-consult.dk/en/cases-en/
Cosmetic shop management system project report.pdfKamal Acharya
Buying new cosmetic products is difficult. It can even be scary for those who have sensitive skin and are prone to skin trouble. The information needed to alleviate this problem is on the back of each product, but it's thought to interpret those ingredient lists unless you have a background in chemistry.
Instead of buying and hoping for the best, we can use data science to help us predict which products may be good fits for us. It includes various function programs to do the above mentioned tasks.
Data file handling has been effectively used in the program.
The automated cosmetic shop management system should deal with the automation of general workflow and administration process of the shop. The main processes of the system focus on customer's request where the system is able to search the most appropriate products and deliver it to the customers. It should help the employees to quickly identify the list of cosmetic product that have reached the minimum quantity and also keep a track of expired date for each cosmetic product. It should help the employees to find the rack number in which the product is placed.It is also Faster and more efficient way.
Quality defects in TMT Bars, Possible causes and Potential Solutions.PrashantGoswami42
Maintaining high-quality standards in the production of TMT bars is crucial for ensuring structural integrity in construction. Addressing common defects through careful monitoring, standardized processes, and advanced technology can significantly improve the quality of TMT bars. Continuous training and adherence to quality control measures will also play a pivotal role in minimizing these defects.
Final project report on grocery store management system..pdfKamal Acharya
In today’s fast-changing business environment, it’s extremely important to be able to respond to client needs in the most effective and timely manner. If your customers wish to see your business online and have instant access to your products or services.
Online Grocery Store is an e-commerce website, which retails various grocery products. This project allows viewing various products available enables registered users to purchase desired products instantly using Paytm, UPI payment processor (Instant Pay) and also can place order by using Cash on Delivery (Pay Later) option. This project provides an easy access to Administrators and Managers to view orders placed using Pay Later and Instant Pay options.
In order to develop an e-commerce website, a number of Technologies must be studied and understood. These include multi-tiered architecture, server and client-side scripting techniques, implementation technologies, programming language (such as PHP, HTML, CSS, JavaScript) and MySQL relational databases. This is a project with the objective to develop a basic website where a consumer is provided with a shopping cart website and also to know about the technologies used to develop such a website.
This document will discuss each of the underlying technologies to create and implement an e- commerce website.
Automobile Management System Project Report.pdfKamal Acharya
The proposed project is developed to manage the automobile in the automobile dealer company. The main module in this project is login, automobile management, customer management, sales, complaints and reports. The first module is the login. The automobile showroom owner should login to the project for usage. The username and password are verified and if it is correct, next form opens. If the username and password are not correct, it shows the error message.
When a customer search for a automobile, if the automobile is available, they will be taken to a page that shows the details of the automobile including automobile name, automobile ID, quantity, price etc. “Automobile Management System” is useful for maintaining automobiles, customers effectively and hence helps for establishing good relation between customer and automobile organization. It contains various customized modules for effectively maintaining automobiles and stock information accurately and safely.
When the automobile is sold to the customer, stock will be reduced automatically. When a new purchase is made, stock will be increased automatically. While selecting automobiles for sale, the proposed software will automatically check for total number of available stock of that particular item, if the total stock of that particular item is less than 5, software will notify the user to purchase the particular item.
Also when the user tries to sale items which are not in stock, the system will prompt the user that the stock is not enough. Customers of this system can search for a automobile; can purchase a automobile easily by selecting fast. On the other hand the stock of automobiles can be maintained perfectly by the automobile shop manager overcoming the drawbacks of existing system.
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...Dr.Costas Sachpazis
Terzaghi's soil bearing capacity theory, developed by Karl Terzaghi, is a fundamental principle in geotechnical engineering used to determine the bearing capacity of shallow foundations. This theory provides a method to calculate the ultimate bearing capacity of soil, which is the maximum load per unit area that the soil can support without undergoing shear failure. The Calculation HTML Code included.
Overview of the fundamental roles in Hydropower generation and the components involved in wider Electrical Engineering.
This paper presents the design and construction of hydroelectric dams from the hydrologist’s survey of the valley before construction, all aspects and involved disciplines, fluid dynamics, structural engineering, generation and mains frequency regulation to the very transmission of power through the network in the United Kingdom.
Author: Robbie Edward Sayers
Collaborators and co editors: Charlie Sims and Connor Healey.
(C) 2024 Robbie E. Sayers
Courier management system project report.pdfKamal Acharya
It is now-a-days very important for the people to send or receive articles like imported furniture, electronic items, gifts, business goods and the like. People depend vastly on different transport systems which mostly use the manual way of receiving and delivering the articles. There is no way to track the articles till they are received and there is no way to let the customer know what happened in transit, once he booked some articles. In such a situation, we need a system which completely computerizes the cargo activities including time to time tracking of the articles sent. This need is fulfilled by Courier Management System software which is online software for the cargo management people that enables them to receive the goods from a source and send them to a required destination and track their status from time to time.
Efficient resampling features and convolution neural network model for image forgery detection
1. Indonesian Journal of Electrical Engineering and Computer Science
Vol. 25, No. 1, January 2022, pp. 183~190
ISSN: 2502-4752, DOI: 10.11591/ijeecs.v25.i1.pp183-190 183
Journal homepage: http://ijeecs.iaescore.com
Efficient resampling features and convolution neural network
model for image forgery detection
Manjunatha S1,2
, Malini M. Patil3
1
Department of Information Science and Engineering, Global Academy of Technology, Bengaluru, India
2
Department of CSE, J.S.S Academy of Technical Education Bengaluru, Bengaluru, India
3
Department of Information Science and Engineering, J.S.S Academy of Technical Education, Bengaluru, India
Article Info ABSTRACT
Article history:
Received Feb 4, 2021
Revised Oct 31, 2021
Accepted Dec 1, 2021
The extended utilization of picture-enhancing or manipulating tools has led
to ease of manipulating multimedia data which includes digital images.
These manipulations will disturb the truthfulness and lawfulness of images,
resulting in misapprehension, and might disturb social security. The image
forensic approach has been employed for detecting whether or not an image
has been manipulated with the usage of positive attacks which includes
splicing, and copy-move. This paper provides a competent tampering
detection technique using resampling features and convolution neural
network (CNN). In this model range spatial filtering (RSF)-CNN,
throughout preprocessing the image is divided into consistent patches. Then,
within every patch, the resampling features are extracted by utilizing affine
transformation and the Laplacian operator. Then, the extracted features are
accumulated for creating descriptors by using CNN. A wide-ranging analysis
is performed for assessing tampering detection and tampered region
segmentation accuracies of proposed RSF-CNN based tampering detection
procedures considering various falsifications and post-processing attacks
which include joint photographic expert group (JPEG) compression, scaling,
rotations, noise additions, and more than one manipulation. From the
achieved results, it can be visible the RSF-CNN primarily based tampering
detection with adequately higher accurateness than existing tampering
detection methodologies.
Keywords:
Convolution neural network
Deep learning
Hybrid attack
Image tampering detection
Resampling feature
This is an open access article under the CC BY-SA license.
Corresponding Author:
Manjunatha S
Department of Information Science and Engineering, Global Academy of Technology
Bengaluru, India
Email: manju.dvg2020@gmail.com
1. INTRODUCTION
The emergence of social networks in our daily life spurs the advent of various digital image editing
tools which leads to belief issues of multimedia content being circulated. Designing high-quality tampering
employing a machine learning model that is visually undetectable through the human eye [1], [2] is well
within reach of the user through the following tool such as FaceApp [3], Adobe Sensiei [4], DeepPhoto
editor [5], and Adobe sky Replace [6]. Thus, classifying an image as tampered with or not is becoming an
extremely difficult task. In general, tampering is classified into content preserving and changing [7]. The
copy-clone, splicing, and object exclusion are widely used primary attacks, and blurring, compression, and
contrast adjustment are commonly used secondary attacks. However, the primary attack will result in a
semantical illustration of an image. Thus, this work focuses on detecting the primary attack.
2. ISSN: 2502-4752
Indonesian J Elec Eng & Comp Sci, Vol. 25, No. 1, January 2022: 183-190
184
Recently, several methods have been presented for detecting primary tampering attacks [8]-[10].
However, these methods have focused on detecting whether an image is tampered or not and failed to
localize tampered region within a tampered image [11], [12]. In [13], [14] able to localize the patches that are
tampered with through the employment of frequency domain [15], [16]. Sudiatmika et al. [17] used noisy
information [18] of compressed image for establishing whether the image has tampered or not. In recent
times deep learning techniques have achieved a very good result in computer vision applications [19]-[21]
including image tampering detection and classification [22], [23]. The autoencoder [24] and convolution
neural network (CNN) [22], [23] have been widely used to detect primary attacks such as splicing [24], [25]
and copy-move [26]. However, these model fails to provide good accuracy when image undergoes a hybrid
transformation as they are designed to detect either splicing or copy-move. Further, the traditional fully-
connected CNN-based framework [27] fails to generalize different noise induced through different tampering
detection methods; thus, poor tampering region localization outcome is achieved.
In addressing such issues in this work presented an improved tampering detection model employing
improved preprocessing, feature aggregation, and CNN architecture [11]. Bunk et al. [11] showed, image
tampering induces noise because of periodic interpolation among adjacent pixels which can be understood
through resampling features [13], [28]. Here affine transformation and Laplacian function is used for
extracting resampling features and descriptor is built through CNN. The significance of range spatial filtering
(RSF)-tamper detection (TD) is described next. The paper presented a CNN-based tampering detection
method by learning Resampling Features. The RSF-TD can extract useful features among adjacent pixels of
both horizontal and vertical directions with better accuracy. The RSF-TD can be used for detecting tampered
image that has undergone multiple tampering. The RSF-TD achieves very good precision, F1-score, and
recall performance in comparison with the recent tampering detection method.
The paper is arranged as follows: the proposed resampling feature tampering detection method
through CNN is conferred in section 2. The overall outcome achieved using the RSF-TD method over
different tampering detection model are given. The last section discusses the significance of RSF-TD and
also discusses the future direction of research work.
2. EFFICIENT RSF AND CNN MODEL FOR IMAGE FORGERY DETECTION
Here the tampering detection through resampling feature extraction and CNN descriptor is
presented. For detecting tampering and segmenting tampered region efficiently the following design is
presented in Figure 1. This RSF-TD architecture is having six steps. First, the image is segmented into
different patches. Then, the feature is extracted using a scale-invariant descriptor for establishing the
duplicated region even under the small and smooth region.
Figure 1. Methodology of proposed RSF-CNN Model for tampering detection
2.1. Preprocessing and resampling feature detection and extraction
Generally, the tampered images will have significant impacts on the statistical properties along the
edges. Similar to methodologies presented in [29] in this work the resampling feature are extracted through
affine transformation and Laplacian function. Here the image is divided into non-overlapping with patch size
set to 64. The dimension of the patch will be 64*64 for considering an image size of 512*512. Keeping the
size to 512*512 significantly reduce the computation time in detecting and localizing tampering region. To
predict the linear error Laplacian function is utilized [13]. Affine transformation matrices are used for
collecting the errors considering various directions and angels. Finally, Fourier transformation is used for
3. Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752
Efficient resampling features and convolution neural network model for …. (Manjunatha S.)
185
extracting RSF considering diverse tampering transformation. The existing method performs localization at
pixel level; however, the RSF-TD performs localization at the pixel level. Here the patch size is set to 32*32
and block size is kept to 8*8 for extracting local RSF with less redundancy. Once RSF is extracted they are
trained using CNN; achieving higher accuracy depends on the order of feature aggregation. The existing
model arranges the features in the horizontal or vertical direction; thus, poor correlation among different
directions significantly impacts accuracy. To address the aforementioned issues in this space-filling
curve [30]-[33] is used. The spacing filling is very effective in converting the multi-dimension problems to a
single dimension [34]-[37].
2.2. Aggregation of features and decision using CNN for tampering detection
The previous section extracts efficient RSF features; these features must be cumulated to build an
efficient learning mechanism for the classification of the image is undergone hybrid primary attacks.
Different aggregation methods are modeled in (1) to (4) [38]. The (1) defines the minimum aggregation
model,
𝐺↓ = min
𝑗=1,…, 𝑂𝑞
𝐺𝑗 (1)
The (2) defines the maximum aggregation function
𝐺↑ = max
𝑗=1,…,𝐺𝑞
𝐺𝑗 (2)
The (3) defines the mean aggregation function
𝐺→ =
1
𝑂𝑞
∑ 𝐺𝑗
𝑂𝑞
𝑗=1
(3)
The (4) squared mean aggregation function
𝐺← =
1
𝑂𝑞
∑ 𝐺𝑗
2
𝑂𝑞
𝑗=1
(4)
In (1) to (4), the parameter 𝐺𝑗 = [𝐺𝑗,1, … , 𝐺𝑗,𝐷] defines the amount of feature extracted in the 𝑗𝑡ℎ
patch and 𝑂𝑞 defines the maximum patch size used. In this work we used different pooling methods; then
features are aggregated for eliminating spatial dependency. Let 𝜃 defines CNN parameter, 𝑀 defines loss
function of CNN structure sued, and 𝐺𝑎𝑔𝑟 defines cumulated features. Then, the gradient of 𝑀 concerning 𝜃
is defined as follows:
𝜕𝑀
𝜕𝜃
= ∑
𝜕𝑀
𝜕𝐺𝑎𝑔𝑟,𝑑
𝐷
𝑑=1
𝜕𝐺𝑎𝑔𝑟,𝑑
𝜕𝜃
(5)
with,
𝜕𝐺𝑎𝑔𝑟,𝑑
𝜕𝜃
=
{
𝜕𝐺𝑗,𝑑
𝜕𝜃
∙ 𝜇𝑗,𝑗↑(𝑑) max pooling
𝜕𝐺𝑗,𝑑
𝜕𝜃
∙ 𝜇𝑗,𝑗↓(𝑑) min pooling
1
𝑂𝑞
∑
𝜕𝐺𝑗,𝑑
𝜕𝜃
𝑂𝑞
𝑗=1
average pooling
1
𝑂𝑞
∑ 2𝐺𝑗,𝑑
𝜕𝐺𝑗,𝑑
𝜕𝜃
𝑂𝑞
𝑗=1
avg. sqr pooling
(6)
Using (6) we can state that 𝜇𝑗,𝑘 = 1 provided 𝑗 = 𝑘; otherwise 𝜇𝑗,𝑘 = 0. The parameters 𝑗↓(𝑑) and 𝑗↑(𝑑)
define the feature vector with the smallest and largest 𝑑𝑡ℎ
component. Using all pooling together assures that
gradient weights can be optimized more efficiently; thus, aiding in achieving better tampering detection
accuracy. Finally, the optimized RSF are aggregated for building descriptor 𝐺 using two-layer fully
connected CNN [39]; this assures higher accuracy with minimal computation time.
2.3. Training of CNN
The RSF-TD without the need for any fixed network can be trained end-to-end by using the whole
image. Then, the decision is taken to classify image is tampered or not. Here loss functions are back-
4. ISSN: 2502-4752
Indonesian J Elec Eng & Comp Sci, Vol. 25, No. 1, January 2022: 183-190
186
propagated within the network of respective patches; this helps in obtaining better-correlated features in an
adaptive nature. Thus, assures RSF-TD achieves better detection accuracy with less misclassification.
3. RESULTS AND DISCUSSIONS
Here experiment is conducted to study the tampering detection outcome of RSF-TD and standard
tampering detection methods. The experiment is conducted using window 10 operating system with Intel
quad processor and 16 GB Ram. The RSF-TD has been implemented through Matlab, C++, and the
Anaconda python 3 frameworks. The experiment is done using two standard datasets such as media
integration and communication center (MICC) and D0 dataset because it has undergone a hybrid
transformation attack. More detail of the dataset used for the experiment is given in Table 1. The metric used
for evaluating the performance of RSF-TD and the existing tampering detection model [40], [41] is recall,
F1-score, and false-positive rate.
Table 1. Dataset description
Dataset Image size Scaling and rotation Compression
MICC 600 Yes No
D0 50 yes Yes
3.1. Evaluation of performance on MICC dataset
The MICC is composed of 300 genuine images and 300 images undergo a hybrid transformation
attack. Therefore, a total of 600 images are available. On average 1.2% of patches are tampered with. Thus, it
is extremely difficult in detecting tampering. The Figure 2(a) shows the original image, Figure 2(b) shows
respective segmentation outcome using RSF-TD respectively. In Figure 2, the tampering region segmentation
outcome achieved using RSF-TD is shown. Similarly, The Figure 3(a) shows the original image, Figure 3(b)
shows respective ground truth of tampered region, segmentation outcome achieved using existing and
RSF-TD model is shown in Figure 3(c) and Figure 3(d), respectively. Figure 3 shows the segmentation
outcome achieved using RSF-TD and the existing segmentation model [26]. Table 2, shows the recall, false
positive rate (FPR), and F1-Score performance achieved using RSF-TD and the existing tampering detection
model [40]. From the result, we can state the RSF-TD archives have much better detection accuracies under
hybrid attack in comparison with existing tampering detection models.
(a)
(b)
Figure 2. The results of the proposed RSF-CNN model; (a) original image and (b) segmentation outcome
using RSF-TD
3.2. Evaluation of performance on D0 dataset
In this section performance of RSF-TD and the existing tampering, detection methods are studied
using the D0 dataset. The dataset undergoes hybrid transformation such as scaling and rotation with JPEG
compression. Thus, tampering detection makes very challenging. The Figure 4(a) shows the original image,
5. Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752
Efficient resampling features and convolution neural network model for …. (Manjunatha S.)
187
Figure 4(b) shows respective ground truth of tampered region, feature extraction and segmentation outcome
achieved using existing and RSF-TD model is shown in Figure 4(c), and Figure 4(d), respectively. In Figure 4,
the segmentation outcome of the tampered region using RSF-TD is shown. The detection accuracy using
RSF-TD and the existing model on D0 dataset is shown in Table 3. Through the outcome achieved we can
state that RSF-TD performs significantly better than the existing tampering detection model in terms of
F1-score, FPR, precision, and recall.
(a) (b)
(c) (d)
Figure 3. Tampering region segmentation outcome using RSF-TD and existing tampering region
segmentation model; (a) original image, (b) ground truth, (c) existing model [26], and (d) RSF-TD
Table 2. RSF-TD and existing method tampering detection accuracy for MICC dataset
Model Recall/TPR FPR F1-Score
Raju and Nair 2018 [40] 89.14 - 92.6
RSF-TD 97.5 1.4 97.7
(a) (b) (c) (d)
Figure 4. Segmentation outcome of RSF-TD model; (a) input image, (b) ground truth, (c) feature extraction,
and (d) segmentation outcome
6. ISSN: 2502-4752
Indonesian J Elec Eng & Comp Sci, Vol. 25, No. 1, January 2022: 183-190
188
Table 3. RSF-TD and existing tampering detection model for D0 dataset
Model Recall Precision FPR F1
Huang and Ciou [41] 84.88 92.81 3.39 88.67
RSF-CNN 98.08 98.84 1.68 99.28
3.3. Computation complexity performance evaluation
Here experiment is done to study the computation complexity of different tampering detection
methods. The computation complexity is measured in terms of the amount of time taken to detect tampering
regions. Figure 5 shows the computation time taken to process MICC-600 using the deep learning-based
method [42], the hierarchical method, and RSF-TD. Similarly, Figure 6 shows the computation time taken to
process MICC-2000 using the deep learning-based method [42] and RSF-TD. Figure 7 shows the
computation time taken to process the CoMoFoD dataset using the DBSCAN-based method, and RSF-TD.
From Figures 5 to 7 we can see the RSF-TD significantly reduces computation time in comparison with the
deep learning-based method, hierarchical method, and DBSCAN-based method.
Figure 5. Computation time for MICC-600 Figure 6. Computation time for MICC-2000
Figure 7. Computation time for CoMoFoD
4. CONCLUSION
Here the paper emphasized using the resampling feature through effective preprocessing and CNN
model. The RSF-TD brings in good tradeoffs between achieving higher detection accuracies with minimal
computation time. Further, able to achieve higher detection accuracies with better segmentation outcomes in
comparison with the standard tampering detection model. The result proves the RSF-TD can extract highly
correlated features and eliminate spatial dependencies. The experiment conducted on two datasets with
hybrid tampering attack transformation shows the RSF-TD achieves much better accuracies in comparison
with the recent tampering detection model which is measured through the following metrics such as F1-score,
FPR, TPR, precision, and recall. Despite achieving superior detection accuracies the model can be further
improved in the future by eliminating outliers through effective design of CNN framework that is robust
against noise. Further, validate the model using more diverse datasets.
REFERENCES
[1] F. Luan, S. Paris, E. Shechtman, and K. Bala, “Deep Photo Style Transfer,” 2017 IEEE Conference on Computer Vision and
Pattern Recognition (CVPR), 2017, pp. 6997-7005, doi: 10.1109/CVPR.2017.740.
[2] N. P. Joglekar and P. N. Chatur, “A Compressive Survey on Active and Passive Methods for Image Forgery Detection,”
International Journal of Engineering and Computer Science, vol. 4, no. 1, pp. 10187-10190, 2015.
7. Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752
Efficient resampling features and convolution neural network model for …. (Manjunatha S.)
189
[3] J. Bunk et al., “Detection and Localization of Image Forgeries Using Resampling Features and Deep Learning,” 2017 IEEE
Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2017, pp. 1881-1889, doi:
10.1109/CVPRW.2017.235.
[4] V. Schetinger, M. M. Oliveira, R. DaSilva, and T. J. Carvalho, “Humans are easily fooled by digital images,” Computers &
Graphics, vol. 68, pp. 142-151, 2015, doi: 10.1016/j.cag.2017.08.010.
[5] R. M. Joseph and A. S. Chithra, “Literature survey on image manipulation detection,” International Research Journal of
Engineering and Technology (IRJET), vol. 2, no. 04, pp. 740-744, 2015.
[6] M. T. Shashidhar and K. B. Ramesh, “A novel framework for optimized digital forensic for mitigating complex image attacks,”
International Journal of Electrical and Computer Engineering (IJECE), vol. 10, no. 5, pp. 5198-5207 2020, doi:
10.11591/ijece.v10i5.pp5198-5207.
[7] H. Li, W. Luo, X. Qiu, and J. Huang, “Image Forgery Localization via Integrating Tampering Possibility Maps,” inIEEE
Transactions on Information Forensics and Security, vol. 12, no. 5, pp. 1240-1252, May 2017, doi: 10.1109/TIFS.2017.2656823.
[8] A. Pourkashani, A. Shahbahrami, and A. Akoushideh, “Copy-move forgery detection using convolutional neural network and K-
mean clustering,” International Journal of Electrical and Computer Engineering (IJECE), vol. 11, no. 3, pp. 2604-2612, 2021,
doi: 10.11591/ijece.v11i3.pp2604-2612.
[9] J. H. Bappy, A. K. Roy-Chowdhury, J. Bunk, L. Nataraj, and B. S. Manjunath, “Exploiting Spatial Structure for Localizing
Manipulated Image Regions,” 2017 IEEE International Conference on Computer Vision (ICCV), 2017, pp. 4980-4989, doi:
10.1109/ICCV.2017.532.
[10] L. Bondi, S. Lameri, D. Güera, P. Bestagini, E. J. Delp, and S. Tubaro, “Tampering Detection and Localization Through
Clustering of Camera-Based CNN Features,” 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops
(CVPRW), 2017, pp. 1855-1864, doi: 10.1109/CVPRW.2017.232.
[11] R. Thakur and R. Rohilla, “Recent advances in digital image manipulation detection techniques: A brief review,” Forensic
Science International, vol. 312, 2020, doi: 10.1016/j.forsciint.2020.110311.
[12] Y. Liu, Q. Guan, X. Zhao, and Y. Cao, “Image forgery localization based on multi-scale convolutional neural networks,”
IH&MMSec '18: Proceedings of the 6th ACM Workshop on Information Hiding and Multimedia Security, 2018, pp. 85-90, doi:
10.1145/3206004.3206010.
[13] B. Mahdian and S. Saic, “Detection of copy-move forgery using a method based on blur moment invariants,” Forensic science
international, vol. 171, no. 2, pp. 180-189, 2007, doi: 10.1016/j.forsciint.2006.11.002.
[14] W. Wang, J. Dong, and T. Tan, “Exploring DCT Coefficient Quantization Effects for Local Tampering Detection,” in IEEE
Transactions on Information Forensics and Security, vol. 9, no. 10, pp. 1653-1666, Oct. 2014, doi: 10.1109/TIFS.2014.2345479.
[15] I.-C. Chang, J. C. Yu, and C.-C. Chang, “A forgery detection algorithm for exemplar-based inpainting images using multi-region
relation,” Image and Vision Computing, vol. 31, no. 1, pp. 57-71, 2013, doi: 10.1016/j.imavis.2012.09.002.
[16] A. Gupta, “New Copy Move Forgery DetectionTechnique Using Adaptive Over-segmentation and Feature Point Matching,”
Bulletin of Electrical Engineering and Informatics (BEEI), vol. 7, no. 3, pp. 345-349, 2018, doi: 10.11591/eei.v7i3.754.
[17] I. B. K. Sudiatmika, F. Rahman, Trisno, and Suyoto, “Image forgery detection using error level analysis and deep learning,”
TELKOMNIKA (Telecommunication, Computing, Electronics and Control), vol. 17, no. 2, pp. 653-659, 2019, doi:
10.12928/TELKOMNIKA.v17i2.8976.
[18] J. Long, E. Shelhamer, and T. Darrell, “Fully convolutional networks for semantic segmentation,” 2015 IEEE Conference on
Computer Vision and Pattern Recognition (CVPR), 2015, pp. 3431-3440, doi: 10.1109/CVPR.2015.7298965.
[19] B. Zhou, A. Lapedriza, J. Xiao, A. Torralba, and A. Oliva, “Learning deep features for scene recognition using a places database,”
In Advances in neural information processing systems, 2014.
[20] B. Bayar and M. C. Stamm, “A deep learning approach to universal image manipulation detection using a new convolutional
layer,” In Proceedings of the 4th ACM Workshop on Information Hiding and Multimedia Security, 2016, pp 5-10, doi:
10.1145/2909827.2930786.
[21] Y. Rao and J. Ni, “A deep learning approach to detection of splicing and copy-move forgeries in images,” 2016 IEEE
International Workshop on Information Forensics and Security (WIFS), 2016, pp. 1-6, doi: 10.1109/WIFS.2016.7823911.
[22] Y. Zhang, J. Goh, L. L. Win, and V. L. Thing, “Image region forgery detection: A deep learning approach,” In Proceedings of the
Singapore Cyber-Security Conference (SG-CRC), 2016, vol. 14, pp. 1-11, doi: 10.3233/978-1-61499-617-0-1.
[23] V. T. Manu and B. M. Mehtre, “Visual artifacts based image splicing detection in uncompressed images,” 2015 IEEE
International Conference on Computer Graphics, Vision and Information Security (CGVIS), 2015, pp. 145-150, doi:
10.1109/CGVIS.2015.7449911.
[24] J. Li, X. Li, B. Yang, and X. Sun, “Segmentation-Based Image Copy-Move Forgery Detection Scheme,” in IEEE Transactions on
Information Forensics and Security, vol. 10, no. 3, pp. 507-518, March 2015, doi: 10.1109/TIFS.2014.2381872.
[25] V. Badrinarayanan, A. Kendall, and R. Cipolla, “SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image
Segmentation,” in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 12, pp. 2481-2495, Dec. 2017,
doi: 10.1109/TPAMI.2016.2644615.
[26] S.-J. Ryu and H.-K. Lee, “Estimation of linear transformation by analyzing the periodicity of interpolation,” Pattern Recognition
Letters, vol. 36, pp. 89-99, 2014, doi: 10.1016/j.patrec.2013.09.028.
[27] B. Mahdian and S. Saic, “Blind Authentication Using Periodic Properties of Interpolation,” in IEEE Transactions on Information
Forensics and Security, vol. 3, no. 3, pp. 529-538, Sept. 2008, doi: 10.1109/TIFS.2004.924603.
[28] D. Voorhies, “Space-filling curves and pace-filling curves and a measure of coherence,” Graphics Gems II, pp. 26-30, 1991, doi:
10.1016/B978-0-08-050754-5.50018-9.
[29] B. Moon, H. V. Jagadish, C. Faloutsos, and J. H. Saltz, “Analysis of the clustering properties of the Hilbert space-filling curve,” in
IEEE Transactions on Knowledge and Data Engineering, vol. 13, no. 1, pp. 124-141, Jan.-Feb. 2001, doi: 10.1109/69.908985.
[30] T. M. Mohammed et al., “Boosting image forgery detection using resampling detection and copy-move analysis,” Society for
Imaging Science and Technology, vol. 7, pp. 118-1-118-7, 2018, doi: 10.2352/ISSN.2470-1173.2018.07.MWSF-118.
[31] B. Bayar and M. C. Stamm, “On the robustness of constrained convolutional neural networks to JPEG post-compression for
image resampling detection,” 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2017,
pp. 2152-2156, doi: 10.1109/ICASSP.2017.7952537.
[32] L. Jiao and J. Zhao, “A Survey on the New Generation of Deep Learning in Image Processing,” in IEEE Access, vol. 7, pp.
172231-172263, 2019, doi: 10.1109/ACCESS.2019.2956508.
8. ISSN: 2502-4752
Indonesian J Elec Eng & Comp Sci, Vol. 25, No. 1, January 2022: 183-190
190
[33] Z. J. Barad and M. M. Goswami, “Image Forgery Detection using Deep Learning: A Survey,” 2020 6th International Conference
on Advanced Computing and Communication Systems (ICACCS), 2020, pp. 571-576, doi: 10.1109/ICACCS48705.2020.9074408.
[34] A. Kuznetsov, “Digital image forgery detection using deep learning approach,” Journal of Physics: Conference Series, vol. 1368,
no. 3, 2019, doi: 10.1088/1742-6596/1368/3/032028.
[35] R. Huang, F. Fang, H. H. Nguyen, J. Yamagishi, and I. Echizen, “A Method for Identifying Origin of Digital Images Using a
Convolutional Neural Network,” 2020 Asia-Pacific Signal and Information Processing Association Annual Summit and
Conference (APSIPA ASC), 2020, pp. 1293-1299.
[36] A. Flenner, L. Peterson, J. Bunk, T. M. Mohammed, L. Nataraj, and B. S. Manjunath, “Resampling Forgery Detection Using
Deep Learning and A-Contrario Analysis,” Society for Imaging Science and Technology, vol. 7, pp. 212-1-212-7, 2018, doi:
10.2352/ISSN.2470-1173.2018.07.MWSF-212.
[37] F. Marra, D. Gragnaniello, L. Verdoliva, and G. Poggi, “A Full-Image Full-Resolution End-to-End-Trainable CNN Framework
for Image Forgery Detection,” in IEEE Access, vol. 8, pp. 133488-133502, 2020, doi: 10.1109/ACCESS.2020.3009877.
[38] C. Pun, X. Yuan, and X. Bi, “Image Forgery Detection Using Adaptive Oversegmentation and Feature Point Matching,” in IEEE
Transactions on Information Forensics and Security, vol. 10, no. 8, pp. 1705-1716, Aug. 2015, doi: 10.1109/TIFS.2015.2423261.
[39] A. Kuznetsov and V. Myasnikov, “A new copy-move forgery detection algorithm using image preprocessing procedure,”
Procedia Engineering, vol. 201, pp. 436-444, 2017, doi: 10.1016/j.proeng.2017.09.671.
[40] P. M. Raju and M. S. Nair, “Copy-move forgery detection using binary discriminant features,” Journal of King Saud University -
Computer and Information Sciences, 2018, doi: 10.1016/j.jksuci.2018.11.004.
[41] H-Y. Huang and A-J. Ciou, “Copy-move forgery detection for image forensics using the superpixel segmentation and the Helmert
transformation,” EURASIP Journal on Image and Video Processing, vol. 68, 2019, doi: 10.1186/s13640-019-0469-9.
[42] M. Bilal, H. A. Habib, Z. Mehmood, T. Saba, and M. Rashid “Single and Multiple Copy–Move Forgery Detection and
Localization in Digital Images Based on the Sparsely Encoded Distinctive Features and DBSCAN Clustering,” Arabian journal
for science and engineering, vol. 45, pp. 2975-2992, 2019, doi: 10.1007/s13369-019-04238-2.
BIOGRAPHIES OF AUTHORS
Manjunatha S is a Research Scholar at JSSATE Research Centre, Dept. of CSE,
JSSATE, affiliated to VTU Belagavi. He completed M. Tech Computer Science and
Engineering from NMAMIT Nitte Mangalore, affiliated to VTU Belagavi, Karnataka. Now he
is working as an Associate Professor Dept. Of ISE, Global Academy of Technology,
Bengaluru. He can be contacted at email: manjunaths@gat.ac.in.
Dr. Malini M. Patil is presently working as an Associate Professor in the
Department of Information Science and Engineering at J.S.S. Academy of Technical
Education, Bangalore, Karnataka, India. She received her Ph.D. Degree from Bharathiar
University in the year 2015. Her research interests are big data analytics, bioinformatics, cloud
computing, image processing. She has published more than 50 research papers in reputed peer-
reviewed international journals. She is a member of IEEE, IEI, ISTE, CSI. She has attended
and presented papers in many international conferences in India and Abroad. Presently she is
guiding 3 research scholars and one research scholar has completed the Ph.D. He can be
contacted at email: drmalinimpatil@gmail.com.