This document presents a proposed system to detect sign language gestures from hearing impaired individuals and convert them to text or speech that can be understood by others. The system uses a convolutional neural network trained on a dataset of images of hand gestures to classify signs in real-time. When a sign is detected, the system sends a corresponding text or speech output. The goal is to help hearing impaired people more easily communicate with those who do not understand sign language. The proposed system achieved over 99% accuracy in testing and could potentially be developed into a wearable device.
IRJET- American Sign Language ClassificationIRJET Journal
This document describes a system for classifying American Sign Language (ASL) gestures into their corresponding English alphabet letters and then converting those letters to audio using text-to-speech. The system uses convolutional neural networks (CNNs) with transfer learning using the VGG16 model to classify ASL images. It achieved 0.99 precision on classification. The CNN model was trained on a dataset of ASL alphabet gestures. Once a gesture is classified, the letter is converted to audio using the gTTS library to play the sound of the classified letter. The system provides a pipeline from image classification to audio output to help facilitate communication between deaf and hearing communities.
IRJET - Hand Gesture Recognition to Perform System OperationsIRJET Journal
This document describes a hand gesture recognition system that uses deep learning and convolutional neural networks. The system is trained on a dataset of over 50,000 images to recognize 19 different gestures. It first calibrates the background, segments the hand from the image, and recognizes the gesture. The model achieves 86.39% accuracy on the test set after training for 20 epochs with a batch size of 64 using an Adam optimizer.
IRJET- Mango Classification using Convolutional Neural NetworksIRJET Journal
This document presents research on classifying different types of mangoes using convolutional neural networks (CNNs). The researchers collected a dataset of over 5000 mango images across 5 classes. They used transfer learning with the Inception v3 CNN model pre-trained on ImageNet, removing the final classification layer and retraining a new one for the mango classes. The CNN achieved over 99% accuracy on the test set at classifying mango types, demonstrating that CNNs can effectively perform fine-grained image classification of mangoes and distinguish between similar types.
IRJET-Multiclass Classification Method Based On Deep Learning For Leaf Identi...IRJET Journal
This document discusses a multiclass classification method using deep learning for leaf identification to help farmers. It proposes using a convolutional neural network (CNN) model for feature extraction and classification of leaf images. The CNN model is trained on labeled leaf image data and can then be used to classify new unlabeled leaf images. The method involves preprocessing leaf images, extracting features using the CNN model, and classifying the leaves into different plant categories. The researchers tested their method on 13 plant leaf categories and 4 disease categories, achieving 95.25% accuracy. They conclude CNNs are well-suited for leaf identification and classification tasks due to their ability to handle large image datasets.
FACE COUNTING USING OPEN CV & PYTHON FOR ANALYZING UNUSUAL EVENTS IN CROWDSIRJET Journal
The document discusses face counting using OpenCV and Python by analyzing unusual events in crowds. It proposes using the Haar cascade algorithm for face detection and counting. Feature extraction is performed using gray-level co-occurrence matrix (GLCM) to extract texture and edge features. Discriminant analysis is then used to differentiate between samples accurately. The system aims to correctly detect and count faces in images using Python tools like OpenCV for digital image processing tasks and feature extraction algorithms like GLCM and discrete wavelet transform (DWT). It is intended to have good recognition accuracy compared to previous methods.
IRJET- Image Classification – Cat and Dog ImagesIRJET Journal
This document describes a study that used a convolutional neural network to classify images of cats and dogs. The network was trained on a dataset of 10,000 images divided into training and test sets. Data augmentation techniques were used to increase the size of the training dataset. The network architecture included convolutional, max pooling, dropout and dense layers. It was trained for 15 epochs using early stopping to prevent overfitting, and achieved a test accuracy of 90.1% for classifying images as cats or dogs. The results demonstrate that convolutional neural networks can effectively perform image classification tasks.
IRJET- Implementation of Gender Detection with Notice Board using Raspberry PiIRJET Journal
1) The document describes a system that uses a Raspberry Pi device with a camera module to implement gender detection.
2) Images captured by the camera are processed through a convolutional neural network to extract facial features and predict gender.
3) The system is intended to address limitations of existing gender detection technologies and provide a low-cost hardware solution using a Raspberry Pi single-board computer.
The document describes a traffic sign recognition model that uses a convolutional neural network (CNN) algorithm to recognize traffic signs from images with 94.98% accuracy. The model was trained on the German Traffic Sign Recognition Benchmark dataset containing over 50,000 images split into 80% for training and 20% for testing. The CNN model extracts features from the images and classifies the traffic signs. The results show the network can accurately classify traffic signs and could be integrated into advanced driver assistance systems to help vehicles recognize road signs.
IRJET- American Sign Language ClassificationIRJET Journal
This document describes a system for classifying American Sign Language (ASL) gestures into their corresponding English alphabet letters and then converting those letters to audio using text-to-speech. The system uses convolutional neural networks (CNNs) with transfer learning using the VGG16 model to classify ASL images. It achieved 0.99 precision on classification. The CNN model was trained on a dataset of ASL alphabet gestures. Once a gesture is classified, the letter is converted to audio using the gTTS library to play the sound of the classified letter. The system provides a pipeline from image classification to audio output to help facilitate communication between deaf and hearing communities.
IRJET - Hand Gesture Recognition to Perform System OperationsIRJET Journal
This document describes a hand gesture recognition system that uses deep learning and convolutional neural networks. The system is trained on a dataset of over 50,000 images to recognize 19 different gestures. It first calibrates the background, segments the hand from the image, and recognizes the gesture. The model achieves 86.39% accuracy on the test set after training for 20 epochs with a batch size of 64 using an Adam optimizer.
IRJET- Mango Classification using Convolutional Neural NetworksIRJET Journal
This document presents research on classifying different types of mangoes using convolutional neural networks (CNNs). The researchers collected a dataset of over 5000 mango images across 5 classes. They used transfer learning with the Inception v3 CNN model pre-trained on ImageNet, removing the final classification layer and retraining a new one for the mango classes. The CNN achieved over 99% accuracy on the test set at classifying mango types, demonstrating that CNNs can effectively perform fine-grained image classification of mangoes and distinguish between similar types.
IRJET-Multiclass Classification Method Based On Deep Learning For Leaf Identi...IRJET Journal
This document discusses a multiclass classification method using deep learning for leaf identification to help farmers. It proposes using a convolutional neural network (CNN) model for feature extraction and classification of leaf images. The CNN model is trained on labeled leaf image data and can then be used to classify new unlabeled leaf images. The method involves preprocessing leaf images, extracting features using the CNN model, and classifying the leaves into different plant categories. The researchers tested their method on 13 plant leaf categories and 4 disease categories, achieving 95.25% accuracy. They conclude CNNs are well-suited for leaf identification and classification tasks due to their ability to handle large image datasets.
FACE COUNTING USING OPEN CV & PYTHON FOR ANALYZING UNUSUAL EVENTS IN CROWDSIRJET Journal
The document discusses face counting using OpenCV and Python by analyzing unusual events in crowds. It proposes using the Haar cascade algorithm for face detection and counting. Feature extraction is performed using gray-level co-occurrence matrix (GLCM) to extract texture and edge features. Discriminant analysis is then used to differentiate between samples accurately. The system aims to correctly detect and count faces in images using Python tools like OpenCV for digital image processing tasks and feature extraction algorithms like GLCM and discrete wavelet transform (DWT). It is intended to have good recognition accuracy compared to previous methods.
IRJET- Image Classification – Cat and Dog ImagesIRJET Journal
This document describes a study that used a convolutional neural network to classify images of cats and dogs. The network was trained on a dataset of 10,000 images divided into training and test sets. Data augmentation techniques were used to increase the size of the training dataset. The network architecture included convolutional, max pooling, dropout and dense layers. It was trained for 15 epochs using early stopping to prevent overfitting, and achieved a test accuracy of 90.1% for classifying images as cats or dogs. The results demonstrate that convolutional neural networks can effectively perform image classification tasks.
IRJET- Implementation of Gender Detection with Notice Board using Raspberry PiIRJET Journal
1) The document describes a system that uses a Raspberry Pi device with a camera module to implement gender detection.
2) Images captured by the camera are processed through a convolutional neural network to extract facial features and predict gender.
3) The system is intended to address limitations of existing gender detection technologies and provide a low-cost hardware solution using a Raspberry Pi single-board computer.
The document describes a traffic sign recognition model that uses a convolutional neural network (CNN) algorithm to recognize traffic signs from images with 94.98% accuracy. The model was trained on the German Traffic Sign Recognition Benchmark dataset containing over 50,000 images split into 80% for training and 20% for testing. The CNN model extracts features from the images and classifies the traffic signs. The results show the network can accurately classify traffic signs and could be integrated into advanced driver assistance systems to help vehicles recognize road signs.
IRJET- Automated Student’s Attendance Management using Convolutional Neural N...IRJET Journal
This document describes a proposed system to automate student attendance management using convolutional neural networks and face recognition. The system would take attendance automatically by detecting faces in the classroom and comparing them to a database of student faces. This would make the attendance process more efficient than current manual methods like calling roll numbers or paper sign-ins. The system would use a CNN algorithm and face detection/recognition techniques like PCA to detect and identify student faces during lectures and automatically update attendance records.
Image Classification using Deep LearningIRJET Journal
1) The document discusses using a convolutional neural network (CNN) model for image classification of cats and dogs.
2) The CNN model architecture includes convolution, ReLU, max pooling, flatten, and fully connected layers to extract features and classify images.
3) The model was trained on a dataset of cat and dog images from Kaggle and tested on sample images, accurately classifying them as cats or dogs.
IRJET- Advanced Control Strategies for Mold Level ProcessIRJET Journal
This document discusses advanced control strategies for mold level process in continuous casting of steel. It proposes using a fuzzy controller to help control mold level during abnormal conditions like nozzle clogging/unclogging, which can be difficult for a proportional-integral-derivative controller to handle alone. The fuzzy controller design is based on operator expertise. Image processing techniques are also used to monitor mold level in real-time, including image segmentation using fuzzy clustering and classification using support vector machines. Simulation and online implementation results indicate the effectiveness of these advanced control methods in maintaining stable mold level during transient disturbances in the continuous casting process.
This document presents a traffic sign recognition system using a convolutional neural network (CNN) model. The authors train the CNN model on a German traffic sign dataset containing over 50,000 images across 43 classes. The proposed CNN architecture contains 4 VGGNet blocks with convolutional, max pooling, dropout and batch normalization layers. The model is trained for 45 epochs and achieves 96.9% accuracy and 11.4% test loss on the test set, outperforming other baseline models. The trained CNN model can accurately classify traffic sign images to assist with applications like self-driving cars.
IRJET- A Vision based Hand Gesture Recognition System using Convolutional...IRJET Journal
This document describes a vision-based hand gesture recognition system using convolutional neural networks. The system captures images of hand gestures using a camera, pre-processes the images, and classifies the gestures using a CNN model. The CNN architecture includes convolutional layers, max pooling layers, dropout layers, and fully connected layers. The system was trained on a dataset of images representing 7 different hand gestures. Testing achieved over 90% accuracy in recognizing the gestures. This vision-based approach allows for natural human-computer interaction without physical devices.
IRJET- Face Recognition using Machine LearningIRJET Journal
This document presents a modified CNN architecture for face recognition that adds two batch normalization operations to improve performance. The CNN extracts facial features using convolutional layers and max pooling, and classifies faces using a softmax classifier. The proposed approach was tested on a face database containing images of 4 individuals with varying lighting conditions. Experimental results showed the modified CNN with batch normalization achieved better recognition results than traditional methods.
This document discusses various techniques for image segmentation. It begins with an abstract discussing image segmentation and its importance in image processing. It then discusses different types of image segmentation like semantic and instance segmentation.
The document then discusses implementation of different image segmentation techniques. It implements region-based segmentation using Mask R-CNN. It performs thresholding-based segmentation using simple thresholding, Otsu's automatic thresholding. It also implements clustering-based segmentation using K-means and Fuzzy C-means. Furthermore, it implements edge-based segmentation using gradient-based techniques like Sobel and Prewitt, and Gaussian-based techniques like Laplacian and Canny edge detectors. Code snippets and output images are provided.
VIDEO BASED SIGN LANGUAGE RECOGNITION USING CNN-LSTMIRJET Journal
This document presents a proposed method for video-based sign language recognition using convolutional neural networks (CNN) and long short-term memory (LSTM). The method uses CNN to extract spatial features from video frames of sign language and LSTM to analyze the temporal characteristics of the frames to recognize the sign. Color segmentation is used to isolate the hands from video frames by detecting colored gloves worn by the signer. CNN is trained on spatial features from frames to classify signs, and LSTM is used to analyze the sequential features from CNN to recognize signs in full videos. The proposed method achieved 94% accuracy on sign recognition in testing.
Deep Learning for Natural Language ProcessingIRJET Journal
This document discusses the use of deep learning techniques in natural language processing. It begins by defining deep learning as a set of machine learning algorithms that use multiple layered models like neural networks to learn inputs. Deep learning aims to process complex data like text in a way that mimics the human brain. The document then discusses several deep learning methods that have been applied to natural language processing tasks, including stacked autoencoders, deep Boltzmann machines, and transfer learning. It provides examples of how these techniques are used to perform tasks like object recognition from text and speech recognition.
IMAGE CAPTION GENERATOR USING DEEP LEARNINGIRJET Journal
The document describes a proposed model for generating image captions using deep learning techniques. The model uses a convolutional neural network (CNN) to extract visual features from an input image, and then uses those features as input to a long short-term memory (LSTM) network to generate a descriptive caption. The model is trained on the Flickr8K dataset and evaluated based on its ability to generate accurate captions for images from that dataset. Evaluation metrics like BLEU score are used to compare the model's captions to reference human-written captions.
Blood Cell Image Classification for Detecting Malaria using CNNIRJET Journal
The document describes a study that used a convolutional neural network (CNN) to classify blood cell images and detect malaria. The CNN model achieved 93.39% accuracy in classifying images as either infected with malaria or uninfected. The study used a dataset of over 27,000 blood cell images. The images were preprocessed and divided into training, validation, and test sets. The CNN model architecture included convolutional layers, max pooling, dropout, and ReLU activation. The model was trained on the preprocessed images and evaluated using metrics like accuracy, precision, and recall. The results demonstrated high performance but also room for improving the model to reduce false predictions.
IRJET- Emotion and Gender Classification in Real-TimeIRJET Journal
This document proposes a convolutional neural network (CNN) model for real-time face detection, gender classification, and emotion recognition. Separate models are trained for each task and then combined into a single pipeline. The architecture is designed to have high performance even on low-end systems. Two CNN models are developed - the first removes fully connected layers and the second further reduces parameters using depthwise separable convolutions. The combined pipeline can process images in 0.15 milliseconds with over 80% fewer parameters than baseline models, achieving human-level accuracy for gender classification and emotion recognition.
DEEP LEARNING BASED BRAIN STROKE DETECTIONIRJET Journal
This document discusses using deep learning and convolutional neural networks to detect brain strokes in CT scan images. It proposes a CNN model with four layers - convolution, pooling, flatten, and fully connected layers - to classify brain CT images as normal or showing signs of stroke. The CNN model was trained on brain CT images and able to accurately diagnose hemorrhages in the brain and detect strokes. This early detection of strokes using deep learning could help reduce death rates by enabling faster treatment.
SIGN LANGUAGE INTERFACE SYSTEM FOR HEARING IMPAIRED PEOPLEIRJET Journal
The document describes a proposed sign language interface system for hearing impaired people. The system aims to use machine learning algorithms like convolutional neural networks to classify hand gestures captured by a webcam into corresponding letters or words. The system would preprocess the images, extract features, then use a trained CNN model to predict the sign and output it as text and speech for better understanding by users. The goal is to help bridge communication between deaf/mute and normal people without requiring specialized gloves or sensors.
IRJET- Low Light Image Enhancement using Convolutional Neural NetworkIRJET Journal
This document presents a study on enhancing low light images using a convolutional neural network. It begins with an introduction to the importance of image quality and challenges of low light images. It then describes the proposed system which uses a convolutional neural network with three layers - gamma correction, multiple convolutional layers, and color restoration. The results show that the convolutional layers help enhance edges in grayscale images. Finally, it concludes the CNN approach is effective for low light image enhancement.
IRJET - Effective Workflow for High-Performance Recognition of Fruits using M...IRJET Journal
This document presents research on fruit recognition using machine learning approaches. The researchers used the fruit-360 dataset containing 74,572 images of 109 fruit classes. They applied feature extraction techniques including HU moments, Haralick texture, and color histogram. Several machine learning classifiers were then trained on the extracted features, including decision tree, K-nearest neighbors, linear discriminant analysis, logistic regression, naive Bayes, random forest, and support vector machine. The models were evaluated using metrics like sensitivity, specificity, precision, F1-score, and accuracy. The results found that K-nearest neighbors and random forest classifiers achieved the best performance with a false positive rate of 0% and high accuracy, outperforming previous fruit recognition studies.
Car Steering Angle Prediction Using Deep LearningIRJET Journal
This document discusses two deep learning models for predicting steering angles of a self-driving car from camera images. The first model uses 3D convolutional layers followed by LSTM layers to incorporate temporal information. The second model uses transfer learning with pre-trained CNN models. Both models were tested in simulation and showed potential for real-world autonomous driving applications by accurately predicting steering angles from camera images alone.
Comparative Study of Pre-Trained Neural Network Models in Detection of GlaucomaIRJET Journal
The document presents a comparative study of various pre-trained neural network models for the early detection of glaucoma from fundus images. Six pre-trained models - Inception, Xception, ResNet50, MobileNetV3, DenseNet121 and DenseNet169 - were analyzed based on their accuracy, loss graphs, confusion matrices and performance metrics like precision, recall, F1 score and specificity. The DenseNet169 model achieved the best results among the models based on these evaluation parameters.
IRJET- Identification of Scene Images using Convolutional Neural Networks - A...IRJET Journal
This document summarizes research on using convolutional neural networks (CNNs) for scene image identification. It first discusses traditional object detection methods and their limitations. CNNs are presented as an improved approach, with convolutional, pooling and fully connected layers to extract features and classify images. Several popular CNN-based object detection algorithms are then summarized, including R-CNN, Fast R-CNN, Faster R-CNN and YOLO. The document concludes that CNN methods provide more accurate object identification compared to traditional algorithms due to their ability to learn from large datasets.
IRJET- Semantic Segmentation using Deep LearningIRJET Journal
The document discusses semantic image segmentation using deep learning techniques. It summarizes several state-of-the-art semantic segmentation models like U-Net, Dilated U-Net, PSPNet, Fully Convolutional DenseNets, Global Convolutional Network (GCN), DeepLabV3, and proposes an optimized FRRN model. It implements these models on the CamVid dataset and evaluates their performance using the intersection-over-union score, finding that the optimized FRRN model achieves a score of 0.87.
TUNNELING IN HIMALAYAS WITH NATM METHOD: A SPECIAL REFERENCES TO SUNGAL TUNNE...IRJET Journal
1) The document discusses the Sungal Tunnel project in Jammu and Kashmir, India, which is being constructed using the New Austrian Tunneling Method (NATM).
2) NATM involves continuous monitoring during construction to adapt to changing ground conditions, and makes extensive use of shotcrete for temporary tunnel support.
3) The methodology section outlines the systematic geotechnical design process for tunnels according to Austrian guidelines, and describes the various steps of NATM tunnel construction including initial and secondary tunnel support.
STUDY THE EFFECT OF RESPONSE REDUCTION FACTOR ON RC FRAMED STRUCTUREIRJET Journal
This study examines the effect of response reduction factors (R factors) on reinforced concrete (RC) framed structures through nonlinear dynamic analysis. Three RC frame models with varying heights (4, 8, and 12 stories) were analyzed in ETABS software under different R factors ranging from 1 to 5. The results showed that displacement increased as the R factor decreased, indicating less linear behavior for lower R factors. Drift also decreased proportionally with increasing R factors from 1 to 5. Shear forces in the frames decreased with higher R factors. In general, R factors of 3 to 5 produced more satisfactory performance with less displacement and drift. The displacement variations between different building heights were consistent at different R factors. This study evaluated how R factors influence
IRJET- Automated Student’s Attendance Management using Convolutional Neural N...IRJET Journal
This document describes a proposed system to automate student attendance management using convolutional neural networks and face recognition. The system would take attendance automatically by detecting faces in the classroom and comparing them to a database of student faces. This would make the attendance process more efficient than current manual methods like calling roll numbers or paper sign-ins. The system would use a CNN algorithm and face detection/recognition techniques like PCA to detect and identify student faces during lectures and automatically update attendance records.
Image Classification using Deep LearningIRJET Journal
1) The document discusses using a convolutional neural network (CNN) model for image classification of cats and dogs.
2) The CNN model architecture includes convolution, ReLU, max pooling, flatten, and fully connected layers to extract features and classify images.
3) The model was trained on a dataset of cat and dog images from Kaggle and tested on sample images, accurately classifying them as cats or dogs.
IRJET- Advanced Control Strategies for Mold Level ProcessIRJET Journal
This document discusses advanced control strategies for mold level process in continuous casting of steel. It proposes using a fuzzy controller to help control mold level during abnormal conditions like nozzle clogging/unclogging, which can be difficult for a proportional-integral-derivative controller to handle alone. The fuzzy controller design is based on operator expertise. Image processing techniques are also used to monitor mold level in real-time, including image segmentation using fuzzy clustering and classification using support vector machines. Simulation and online implementation results indicate the effectiveness of these advanced control methods in maintaining stable mold level during transient disturbances in the continuous casting process.
This document presents a traffic sign recognition system using a convolutional neural network (CNN) model. The authors train the CNN model on a German traffic sign dataset containing over 50,000 images across 43 classes. The proposed CNN architecture contains 4 VGGNet blocks with convolutional, max pooling, dropout and batch normalization layers. The model is trained for 45 epochs and achieves 96.9% accuracy and 11.4% test loss on the test set, outperforming other baseline models. The trained CNN model can accurately classify traffic sign images to assist with applications like self-driving cars.
IRJET- A Vision based Hand Gesture Recognition System using Convolutional...IRJET Journal
This document describes a vision-based hand gesture recognition system using convolutional neural networks. The system captures images of hand gestures using a camera, pre-processes the images, and classifies the gestures using a CNN model. The CNN architecture includes convolutional layers, max pooling layers, dropout layers, and fully connected layers. The system was trained on a dataset of images representing 7 different hand gestures. Testing achieved over 90% accuracy in recognizing the gestures. This vision-based approach allows for natural human-computer interaction without physical devices.
IRJET- Face Recognition using Machine LearningIRJET Journal
This document presents a modified CNN architecture for face recognition that adds two batch normalization operations to improve performance. The CNN extracts facial features using convolutional layers and max pooling, and classifies faces using a softmax classifier. The proposed approach was tested on a face database containing images of 4 individuals with varying lighting conditions. Experimental results showed the modified CNN with batch normalization achieved better recognition results than traditional methods.
This document discusses various techniques for image segmentation. It begins with an abstract discussing image segmentation and its importance in image processing. It then discusses different types of image segmentation like semantic and instance segmentation.
The document then discusses implementation of different image segmentation techniques. It implements region-based segmentation using Mask R-CNN. It performs thresholding-based segmentation using simple thresholding, Otsu's automatic thresholding. It also implements clustering-based segmentation using K-means and Fuzzy C-means. Furthermore, it implements edge-based segmentation using gradient-based techniques like Sobel and Prewitt, and Gaussian-based techniques like Laplacian and Canny edge detectors. Code snippets and output images are provided.
VIDEO BASED SIGN LANGUAGE RECOGNITION USING CNN-LSTMIRJET Journal
This document presents a proposed method for video-based sign language recognition using convolutional neural networks (CNN) and long short-term memory (LSTM). The method uses CNN to extract spatial features from video frames of sign language and LSTM to analyze the temporal characteristics of the frames to recognize the sign. Color segmentation is used to isolate the hands from video frames by detecting colored gloves worn by the signer. CNN is trained on spatial features from frames to classify signs, and LSTM is used to analyze the sequential features from CNN to recognize signs in full videos. The proposed method achieved 94% accuracy on sign recognition in testing.
Deep Learning for Natural Language ProcessingIRJET Journal
This document discusses the use of deep learning techniques in natural language processing. It begins by defining deep learning as a set of machine learning algorithms that use multiple layered models like neural networks to learn inputs. Deep learning aims to process complex data like text in a way that mimics the human brain. The document then discusses several deep learning methods that have been applied to natural language processing tasks, including stacked autoencoders, deep Boltzmann machines, and transfer learning. It provides examples of how these techniques are used to perform tasks like object recognition from text and speech recognition.
IMAGE CAPTION GENERATOR USING DEEP LEARNINGIRJET Journal
The document describes a proposed model for generating image captions using deep learning techniques. The model uses a convolutional neural network (CNN) to extract visual features from an input image, and then uses those features as input to a long short-term memory (LSTM) network to generate a descriptive caption. The model is trained on the Flickr8K dataset and evaluated based on its ability to generate accurate captions for images from that dataset. Evaluation metrics like BLEU score are used to compare the model's captions to reference human-written captions.
Blood Cell Image Classification for Detecting Malaria using CNNIRJET Journal
The document describes a study that used a convolutional neural network (CNN) to classify blood cell images and detect malaria. The CNN model achieved 93.39% accuracy in classifying images as either infected with malaria or uninfected. The study used a dataset of over 27,000 blood cell images. The images were preprocessed and divided into training, validation, and test sets. The CNN model architecture included convolutional layers, max pooling, dropout, and ReLU activation. The model was trained on the preprocessed images and evaluated using metrics like accuracy, precision, and recall. The results demonstrated high performance but also room for improving the model to reduce false predictions.
IRJET- Emotion and Gender Classification in Real-TimeIRJET Journal
This document proposes a convolutional neural network (CNN) model for real-time face detection, gender classification, and emotion recognition. Separate models are trained for each task and then combined into a single pipeline. The architecture is designed to have high performance even on low-end systems. Two CNN models are developed - the first removes fully connected layers and the second further reduces parameters using depthwise separable convolutions. The combined pipeline can process images in 0.15 milliseconds with over 80% fewer parameters than baseline models, achieving human-level accuracy for gender classification and emotion recognition.
DEEP LEARNING BASED BRAIN STROKE DETECTIONIRJET Journal
This document discusses using deep learning and convolutional neural networks to detect brain strokes in CT scan images. It proposes a CNN model with four layers - convolution, pooling, flatten, and fully connected layers - to classify brain CT images as normal or showing signs of stroke. The CNN model was trained on brain CT images and able to accurately diagnose hemorrhages in the brain and detect strokes. This early detection of strokes using deep learning could help reduce death rates by enabling faster treatment.
SIGN LANGUAGE INTERFACE SYSTEM FOR HEARING IMPAIRED PEOPLEIRJET Journal
The document describes a proposed sign language interface system for hearing impaired people. The system aims to use machine learning algorithms like convolutional neural networks to classify hand gestures captured by a webcam into corresponding letters or words. The system would preprocess the images, extract features, then use a trained CNN model to predict the sign and output it as text and speech for better understanding by users. The goal is to help bridge communication between deaf/mute and normal people without requiring specialized gloves or sensors.
IRJET- Low Light Image Enhancement using Convolutional Neural NetworkIRJET Journal
This document presents a study on enhancing low light images using a convolutional neural network. It begins with an introduction to the importance of image quality and challenges of low light images. It then describes the proposed system which uses a convolutional neural network with three layers - gamma correction, multiple convolutional layers, and color restoration. The results show that the convolutional layers help enhance edges in grayscale images. Finally, it concludes the CNN approach is effective for low light image enhancement.
IRJET - Effective Workflow for High-Performance Recognition of Fruits using M...IRJET Journal
This document presents research on fruit recognition using machine learning approaches. The researchers used the fruit-360 dataset containing 74,572 images of 109 fruit classes. They applied feature extraction techniques including HU moments, Haralick texture, and color histogram. Several machine learning classifiers were then trained on the extracted features, including decision tree, K-nearest neighbors, linear discriminant analysis, logistic regression, naive Bayes, random forest, and support vector machine. The models were evaluated using metrics like sensitivity, specificity, precision, F1-score, and accuracy. The results found that K-nearest neighbors and random forest classifiers achieved the best performance with a false positive rate of 0% and high accuracy, outperforming previous fruit recognition studies.
Car Steering Angle Prediction Using Deep LearningIRJET Journal
This document discusses two deep learning models for predicting steering angles of a self-driving car from camera images. The first model uses 3D convolutional layers followed by LSTM layers to incorporate temporal information. The second model uses transfer learning with pre-trained CNN models. Both models were tested in simulation and showed potential for real-world autonomous driving applications by accurately predicting steering angles from camera images alone.
Comparative Study of Pre-Trained Neural Network Models in Detection of GlaucomaIRJET Journal
The document presents a comparative study of various pre-trained neural network models for the early detection of glaucoma from fundus images. Six pre-trained models - Inception, Xception, ResNet50, MobileNetV3, DenseNet121 and DenseNet169 - were analyzed based on their accuracy, loss graphs, confusion matrices and performance metrics like precision, recall, F1 score and specificity. The DenseNet169 model achieved the best results among the models based on these evaluation parameters.
IRJET- Identification of Scene Images using Convolutional Neural Networks - A...IRJET Journal
This document summarizes research on using convolutional neural networks (CNNs) for scene image identification. It first discusses traditional object detection methods and their limitations. CNNs are presented as an improved approach, with convolutional, pooling and fully connected layers to extract features and classify images. Several popular CNN-based object detection algorithms are then summarized, including R-CNN, Fast R-CNN, Faster R-CNN and YOLO. The document concludes that CNN methods provide more accurate object identification compared to traditional algorithms due to their ability to learn from large datasets.
IRJET- Semantic Segmentation using Deep LearningIRJET Journal
The document discusses semantic image segmentation using deep learning techniques. It summarizes several state-of-the-art semantic segmentation models like U-Net, Dilated U-Net, PSPNet, Fully Convolutional DenseNets, Global Convolutional Network (GCN), DeepLabV3, and proposes an optimized FRRN model. It implements these models on the CamVid dataset and evaluates their performance using the intersection-over-union score, finding that the optimized FRRN model achieves a score of 0.87.
Similar to Sign Detection from Hearing Impaired (20)
TUNNELING IN HIMALAYAS WITH NATM METHOD: A SPECIAL REFERENCES TO SUNGAL TUNNE...IRJET Journal
1) The document discusses the Sungal Tunnel project in Jammu and Kashmir, India, which is being constructed using the New Austrian Tunneling Method (NATM).
2) NATM involves continuous monitoring during construction to adapt to changing ground conditions, and makes extensive use of shotcrete for temporary tunnel support.
3) The methodology section outlines the systematic geotechnical design process for tunnels according to Austrian guidelines, and describes the various steps of NATM tunnel construction including initial and secondary tunnel support.
STUDY THE EFFECT OF RESPONSE REDUCTION FACTOR ON RC FRAMED STRUCTUREIRJET Journal
This study examines the effect of response reduction factors (R factors) on reinforced concrete (RC) framed structures through nonlinear dynamic analysis. Three RC frame models with varying heights (4, 8, and 12 stories) were analyzed in ETABS software under different R factors ranging from 1 to 5. The results showed that displacement increased as the R factor decreased, indicating less linear behavior for lower R factors. Drift also decreased proportionally with increasing R factors from 1 to 5. Shear forces in the frames decreased with higher R factors. In general, R factors of 3 to 5 produced more satisfactory performance with less displacement and drift. The displacement variations between different building heights were consistent at different R factors. This study evaluated how R factors influence
A COMPARATIVE ANALYSIS OF RCC ELEMENT OF SLAB WITH STARK STEEL (HYSD STEEL) A...IRJET Journal
This study compares the use of Stark Steel and TMT Steel as reinforcement materials in a two-way reinforced concrete slab. Mechanical testing is conducted to determine the tensile strength, yield strength, and other properties of each material. A two-way slab design adhering to codes and standards is executed with both materials. The performance is analyzed in terms of deflection, stability under loads, and displacement. Cost analyses accounting for material, durability, maintenance, and life cycle costs are also conducted. The findings provide insights into the economic and structural implications of each material for reinforcement selection and recommendations on the most suitable material based on the analysis.
Effect of Camber and Angles of Attack on Airfoil CharacteristicsIRJET Journal
This document discusses a study analyzing the effect of camber, position of camber, and angle of attack on the aerodynamic characteristics of airfoils. Sixteen modified asymmetric NACA airfoils were analyzed using computational fluid dynamics (CFD) by varying the camber, camber position, and angle of attack. The results showed the relationship between these parameters and the lift coefficient, drag coefficient, and lift to drag ratio. This provides insight into how changes in airfoil geometry impact aerodynamic performance.
A Review on the Progress and Challenges of Aluminum-Based Metal Matrix Compos...IRJET Journal
This document reviews the progress and challenges of aluminum-based metal matrix composites (MMCs), focusing on their fabrication processes and applications. It discusses how various aluminum MMCs have been developed using reinforcements like borides, carbides, oxides, and nitrides to improve mechanical and wear properties. These composites have gained prominence for their lightweight, high-strength and corrosion resistance properties. The document also examines recent advancements in fabrication techniques for aluminum MMCs and their growing applications in industries such as aerospace and automotive. However, it notes that challenges remain around issues like improper mixing of reinforcements and reducing reinforcement agglomeration.
Dynamic Urban Transit Optimization: A Graph Neural Network Approach for Real-...IRJET Journal
This document discusses research on using graph neural networks (GNNs) for dynamic optimization of public transportation networks in real-time. GNNs represent transit networks as graphs with nodes as stops and edges as connections. The GNN model aims to optimize networks using real-time data on vehicle locations, arrival times, and passenger loads. This helps increase mobility, decrease traffic, and improve efficiency. The system continuously trains and infers to adapt to changing transit conditions, providing decision support tools. While research has focused on performance, more work is needed on security, socio-economic impacts, contextual generalization of models, continuous learning approaches, and effective real-time visualization.
Structural Analysis and Design of Multi-Storey Symmetric and Asymmetric Shape...IRJET Journal
This document summarizes a research project that aims to compare the structural performance of conventional slab and grid slab systems in multi-story buildings using ETABS software. The study will analyze both symmetric and asymmetric building models under various loading conditions. Parameters like deflections, moments, shears, and stresses will be examined to evaluate the structural effectiveness of each slab type. The results will provide insights into the comparative behavior of conventional and grid slabs to help engineers and architects select appropriate slab systems based on building layouts and design requirements.
A Review of “Seismic Response of RC Structures Having Plan and Vertical Irreg...IRJET Journal
This document summarizes and reviews a research paper on the seismic response of reinforced concrete (RC) structures with plan and vertical irregularities, with and without infill walls. It discusses how infill walls can improve or reduce the seismic performance of RC buildings, depending on factors like wall layout, height distribution, connection to the frame, and relative stiffness of walls and frames. The reviewed research paper analyzes the behavior of infill walls, effects of vertical irregularities, and seismic performance of high-rise structures under linear static and dynamic analysis. It studies response characteristics like story drift, deflection and shear. The document also provides literature on similar research investigating the effects of infill walls, soft stories, plan irregularities, and different
This document provides a review of machine learning techniques used in Advanced Driver Assistance Systems (ADAS). It begins with an abstract that summarizes key applications of machine learning in ADAS, including object detection, recognition, and decision-making. The introduction discusses the integration of machine learning in ADAS and how it is transforming vehicle safety. The literature review then examines several research papers on topics like lightweight deep learning models for object detection and lane detection models using image processing. It concludes by discussing challenges and opportunities in the field, such as improving algorithm robustness and adaptability.
Long Term Trend Analysis of Precipitation and Temperature for Asosa district,...IRJET Journal
The document analyzes temperature and precipitation trends in Asosa District, Benishangul Gumuz Region, Ethiopia from 1993 to 2022 based on data from the local meteorological station. The results show:
1) The average maximum and minimum annual temperatures have generally decreased over time, with maximum temperatures decreasing by a factor of -0.0341 and minimum by -0.0152.
2) Mann-Kendall tests found the decreasing temperature trends to be statistically significant for annual maximum temperatures but not for annual minimum temperatures.
3) Annual precipitation in Asosa District showed a statistically significant increasing trend.
The conclusions recommend development planners account for rising summer precipitation and declining temperatures in
P.E.B. Framed Structure Design and Analysis Using STAAD ProIRJET Journal
This document discusses the design and analysis of pre-engineered building (PEB) framed structures using STAAD Pro software. It provides an overview of PEBs, including that they are designed off-site with building trusses and beams produced in a factory. STAAD Pro is identified as a key tool for modeling, analyzing, and designing PEBs to ensure their performance and safety under various load scenarios. The document outlines modeling structural parts in STAAD Pro, evaluating structural reactions, assigning loads, and following international design codes and standards. In summary, STAAD Pro is used to design and analyze PEB framed structures to ensure safety and code compliance.
A Review on Innovative Fiber Integration for Enhanced Reinforcement of Concre...IRJET Journal
This document provides a review of research on innovative fiber integration methods for reinforcing concrete structures. It discusses studies that have explored using carbon fiber reinforced polymer (CFRP) composites with recycled plastic aggregates to develop more sustainable strengthening techniques. It also examines using ultra-high performance fiber reinforced concrete to improve shear strength in beams. Additional topics covered include the dynamic responses of FRP-strengthened beams under static and impact loads, and the performance of preloaded CFRP-strengthened fiber reinforced concrete beams. The review highlights the potential of fiber composites to enable more sustainable and resilient construction practices.
Survey Paper on Cloud-Based Secured Healthcare SystemIRJET Journal
This document summarizes a survey on securing patient healthcare data in cloud-based systems. It discusses using technologies like facial recognition, smart cards, and cloud computing combined with strong encryption to securely store patient data. The survey found that healthcare professionals believe digitizing patient records and storing them in a centralized cloud system would improve access during emergencies and enable more efficient care compared to paper-based systems. However, ensuring privacy and security of patient data is paramount as healthcare incorporates these digital technologies.
Review on studies and research on widening of existing concrete bridgesIRJET Journal
This document summarizes several studies that have been conducted on widening existing concrete bridges. It describes a study from China that examined load distribution factors for a bridge widened with composite steel-concrete girders. It also outlines challenges and solutions for widening a bridge in the UAE, including replacing bearings and stitching the new and existing structures. Additionally, it discusses two bridge widening projects in New Zealand that involved adding precast beams and stitching to connect structures. Finally, safety measures and challenges for strengthening a historic bridge in Switzerland under live traffic are presented.
React based fullstack edtech web applicationIRJET Journal
The document describes the architecture of an educational technology web application built using the MERN stack. It discusses the frontend developed with ReactJS, backend with NodeJS and ExpressJS, and MongoDB database. The frontend provides dynamic user interfaces, while the backend offers APIs for authentication, course management, and other functions. MongoDB enables flexible data storage. The architecture aims to provide a scalable, responsive platform for online learning.
A Comprehensive Review of Integrating IoT and Blockchain Technologies in the ...IRJET Journal
This paper proposes integrating Internet of Things (IoT) and blockchain technologies to help implement objectives of India's National Education Policy (NEP) in the education sector. The paper discusses how blockchain could be used for secure student data management, credential verification, and decentralized learning platforms. IoT devices could create smart classrooms, automate attendance tracking, and enable real-time monitoring. Blockchain would ensure integrity of exam processes and resource allocation, while smart contracts automate agreements. The paper argues this integration has potential to revolutionize education by making it more secure, transparent and efficient, in alignment with NEP goals. However, challenges like infrastructure needs, data privacy, and collaborative efforts are also discussed.
A REVIEW ON THE PERFORMANCE OF COCONUT FIBRE REINFORCED CONCRETE.IRJET Journal
This document provides a review of research on the performance of coconut fibre reinforced concrete. It summarizes several studies that tested different volume fractions and lengths of coconut fibres in concrete mixtures with varying compressive strengths. The studies found that coconut fibre improved properties like tensile strength, toughness, crack resistance, and spalling resistance compared to plain concrete. Volume fractions of 2-5% and fibre lengths of 20-50mm produced the best results. The document concludes that using a 4-5% volume fraction of coconut fibres 30-40mm in length with M30-M60 grade concrete would provide benefits based on previous research.
Optimizing Business Management Process Workflows: The Dynamic Influence of Mi...IRJET Journal
The document discusses optimizing business management processes through automation using Microsoft Power Automate and artificial intelligence. It provides an overview of Power Automate's key components and features for automating workflows across various apps and services. The document then presents several scenarios applying automation solutions to common business processes like data entry, monitoring, HR, finance, customer support, and more. It estimates the potential time and cost savings from implementing automation for each scenario. Finally, the conclusion emphasizes the transformative impact of AI and automation tools on business processes and the need for ongoing optimization.
Multistoried and Multi Bay Steel Building Frame by using Seismic DesignIRJET Journal
The document describes the seismic design of a G+5 steel building frame located in Roorkee, India according to Indian codes IS 1893-2002 and IS 800. The frame was analyzed using the equivalent static load method and response spectrum method, and its response in terms of displacements and shear forces were compared. Based on the analysis, the frame was designed as a seismic-resistant steel structure according to IS 800:2007. The software STAAD Pro was used for the analysis and design.
Cost Optimization of Construction Using Plastic Waste as a Sustainable Constr...IRJET Journal
This research paper explores using plastic waste as a sustainable and cost-effective construction material. The study focuses on manufacturing pavers and bricks using recycled plastic and partially replacing concrete with plastic alternatives. Initial results found that pavers and bricks made from recycled plastic demonstrate comparable strength and durability to traditional materials while providing environmental and cost benefits. Additionally, preliminary research indicates incorporating plastic waste as a partial concrete replacement significantly reduces construction costs without compromising structural integrity. The outcomes suggest adopting plastic waste in construction can address plastic pollution while optimizing costs, promoting more sustainable building practices.
Literature Review Basics and Understanding Reference Management.pptxDr Ramhari Poudyal
Three-day training on academic research focuses on analytical tools at United Technical College, supported by the University Grant Commission, Nepal. 24-26 May 2024
Understanding Inductive Bias in Machine LearningSUTEJAS
This presentation explores the concept of inductive bias in machine learning. It explains how algorithms come with built-in assumptions and preferences that guide the learning process. You'll learn about the different types of inductive bias and how they can impact the performance and generalizability of machine learning models.
The presentation also covers the positive and negative aspects of inductive bias, along with strategies for mitigating potential drawbacks. We'll explore examples of how bias manifests in algorithms like neural networks and decision trees.
By understanding inductive bias, you can gain valuable insights into how machine learning models work and make informed decisions when building and deploying them.
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024Sinan KOZAK
Sinan from the Delivery Hero mobile infrastructure engineering team shares a deep dive into performance acceleration with Gradle build cache optimizations. Sinan shares their journey into solving complex build-cache problems that affect Gradle builds. By understanding the challenges and solutions found in our journey, we aim to demonstrate the possibilities for faster builds. The case study reveals how overlapping outputs and cache misconfigurations led to significant increases in build times, especially as the project scaled up with numerous modules using Paparazzi tests. The journey from diagnosing to defeating cache issues offers invaluable lessons on maintaining cache integrity without sacrificing functionality.
Introduction- e - waste – definition - sources of e-waste– hazardous substances in e-waste - effects of e-waste on environment and human health- need for e-waste management– e-waste handling rules - waste minimization techniques for managing e-waste – recycling of e-waste - disposal treatment methods of e- waste – mechanism of extraction of precious metal from leaching solution-global Scenario of E-waste – E-waste in India- case studies.
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODELgerogepatton
As digital technology becomes more deeply embedded in power systems, protecting the communication
networks of Smart Grids (SG) has emerged as a critical concern. Distributed Network Protocol 3 (DNP3)
represents a multi-tiered application layer protocol extensively utilized in Supervisory Control and Data
Acquisition (SCADA)-based smart grids to facilitate real-time data gathering and control functionalities.
Robust Intrusion Detection Systems (IDS) are necessary for early threat detection and mitigation because
of the interconnection of these networks, which makes them vulnerable to a variety of cyberattacks. To
solve this issue, this paper develops a hybrid Deep Learning (DL) model specifically designed for intrusion
detection in smart grids. The proposed approach is a combination of the Convolutional Neural Network
(CNN) and the Long-Short-Term Memory algorithms (LSTM). We employed a recent intrusion detection
dataset (DNP3), which focuses on unauthorized commands and Denial of Service (DoS) cyberattacks, to
train and test our model. The results of our experiments show that our CNN-LSTM method is much better
at finding smart grid intrusions than other deep learning algorithms used for classification. In addition,
our proposed approach improves accuracy, precision, recall, and F1 score, achieving a high detection
accuracy rate of 99.50%.
Harnessing WebAssembly for Real-time Stateless Streaming PipelinesChristina Lin
Traditionally, dealing with real-time data pipelines has involved significant overhead, even for straightforward tasks like data transformation or masking. However, in this talk, we’ll venture into the dynamic realm of WebAssembly (WASM) and discover how it can revolutionize the creation of stateless streaming pipelines within a Kafka (Redpanda) broker. These pipelines are adept at managing low-latency, high-data-volume scenarios.
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMSIJNSA Journal
The smart irrigation system represents an innovative approach to optimize water usage in agricultural and landscaping practices. The integration of cutting-edge technologies, including sensors, actuators, and data analysis, empowers this system to provide accurate monitoring and control of irrigation processes by leveraging real-time environmental conditions. The main objective of a smart irrigation system is to optimize water efficiency, minimize expenses, and foster the adoption of sustainable water management methods. This paper conducts a systematic risk assessment by exploring the key components/assets and their functionalities in the smart irrigation system. The crucial role of sensors in gathering data on soil moisture, weather patterns, and plant well-being is emphasized in this system. These sensors enable intelligent decision-making in irrigation scheduling and water distribution, leading to enhanced water efficiency and sustainable water management practices. Actuators enable automated control of irrigation devices, ensuring precise and targeted water delivery to plants. Additionally, the paper addresses the potential threat and vulnerabilities associated with smart irrigation systems. It discusses limitations of the system, such as power constraints and computational capabilities, and calculates the potential security risks. The paper suggests possible risk treatment methods for effective secure system operation. In conclusion, the paper emphasizes the significant benefits of implementing smart irrigation systems, including improved water conservation, increased crop yield, and reduced environmental impact. Additionally, based on the security analysis conducted, the paper recommends the implementation of countermeasures and security approaches to address vulnerabilities and ensure the integrity and reliability of the system. By incorporating these measures, smart irrigation technology can revolutionize water management practices in agriculture, promoting sustainability, resource efficiency, and safeguarding against potential security threats.
Using recycled concrete aggregates (RCA) for pavements is crucial to achieving sustainability. Implementing RCA for new pavement can minimize carbon footprint, conserve natural resources, reduce harmful emissions, and lower life cycle costs. Compared to natural aggregate (NA), RCA pavement has fewer comprehensive studies and sustainability assessments.