SlideShare a Scribd company logo
1 of 18
Aerodynamic Analysis and Design Lab.
WAI NWE TUN
August 16, 2017
Using HOG Descriptors on Superpixels for
Human Detection of UAV Imagery
AADL AI Seminar
Aerodynamic Analysis and Design Lab.
Contents
 Introduction
 System architecture
• Superpixel extraction
• HOG descriptor calculation
• Adaboost classifier
 Performance evaluations
 Dataset
 Evaluation results
 Conclusion
 References
2
Aerodynamic Analysis and Design Lab.
Introduction
3
 Human detection is a challenging task, with many applications such as
pedestrian detection, search and rescue operations, surveillance.
 Object detection process is generally performed in two main steps:
feature extraction and classification.
 Feature extraction maps image windows to a fixed size feature space
that robustly encodes visual form.
 Feature vectors are fed into a classifier to train.
 Using the trained classifier, the images can be determined if desired
object is present or not.
 This system presented human detection using image features by HOG
descriptors for each superpixel by Simple Linear Iterative Clustering
(SLIC) and HOG features descriptors are classified by AdaBoost.
Aerodynamic Analysis and Design Lab.
Introduction
4
 Superpixel algorithms group pixels into perceptually meaningful atomic
regions which can be used to replace the rigid structure of the pixel grid.
 They capture image redundancy, provide a convenient primitive from
which to compute image features, and greatly reduce the complexity of
subsequent image processing tasks. SLIC adapts k-means clustering to
generate superpixels.
 Feature descriptor is a representation of an image or an image patch
that simplifies the image by extracting useful information and throwing
away extraneous information.
 In HOG descriptors, the distribution of directions of gradient are used to
calculate features.
 AdaBoost classifier is an ensemble type which uses a combination of
models consisting of a series of k weak classifiers.
Aerodynamic Analysis and Design Lab.
System Architecture
5
Aerodynamic Analysis and Design Lab.
Features Extraction
6
Aerodynamic Analysis and Design Lab.
Feature Extraction
7
 Sobel filters
 Gradient magnitude and
direction are obtained from
converting x, y coordinates
to polar coordinates
Aerodynamic Analysis and Design Lab.
Superpixel Extraction by SLIC
8
 Instead of fixed-sized block in original HOG, superpixels are used in
HOG calculation.
 Superpixels are a group of consecutive pixels which share common
characteristics such as intensity and represent meaningful region.
 Superpixel algorithm
• graph-based : each pixel as a node, node similarity as edge weight, minimize
cost function over graph.
• gradient ascent methods: clustering
 SLIC : gradient ascent method
 Distance measure is calculated by
normalizing the two measures:
color proximity (LAB) and
spatial proximity (XY).
 K-means clustering is performed using the distance measure for each
limited region.
Aerodynamic Analysis and Design Lab.
HOG Descriptors Calculation
9
 Histograms of gradients are created for each superpixel group and
concatenated all. Then, L2 Normalization is performed.
Aerodynamic Analysis and Design Lab.
AdaBoost Classifier: Learning
10
 Given D, a data set of d class-labeled tuples, (𝑋1, 𝑦1), .. (𝑋 𝑑, 𝑦𝑛) where X
= HOG descriptors, y=human or not, d = number of images, n= 2
(number of class labels).
 Adaboost assigns each training tuple an equal weight of
1
𝑑
.
 Generating k classifiers requires k rounds/iterations
 In round i, tuples from D are sampled to form a training set, 𝐷𝑖, of size d
(with replacement).
 Each tuple’s chance of being selected is based on its weight.
 A classifier model, 𝑀𝑖, is derived from the training tuples, 𝐷𝑖 and its error
is calculated.
 If a tuple was incorrectly classified, its weight is increased. Otherwise, it
is decreased. This is to focus more on the misclassified tuples.
𝑒𝑟𝑟𝑜𝑟 𝑀𝑖 =
𝑗
𝑑
𝑤𝑗 × 𝑒𝑟𝑟(𝑋𝑗
Aerodynamic Analysis and Design Lab.
AdaBoost Classifer: Prediction
11
 Instead of an equal vote, a weight to each classifier’s vote is assigned,
based on how well the classifier performed.
 The lower a classifier’s error rate, the more accurate it is and the higher
its weight for voting should be.
 For each class c (e.g., human presence or not), weights of each classifer
that assigned class c to ‘that’ tuple.
 The class with the highest sum is the ‘winner’ and is returned as the
class prediction for ‘that’ tuple.
𝑣𝑜𝑡𝑒 𝑀𝑖 =
1 − 𝑒𝑟𝑟𝑜𝑟 𝑀𝑖
𝑒𝑟𝑟𝑜𝑟 𝑀𝑖
Aerodynamic Analysis and Design Lab.
Performance Evaluations
12
 Performance is evaluated using four measures: accuracy, precision,
recall and computational time.
 Accuracy
• Overall accuracy of a system
 Precision (Correctness)
• Express how many selected items are relevant
• High precision means that a system returned more
relevant results than irrelevant ones
 Recall (Completeness)
• Express how many relevant items are selected
• High recall means that a system returned most of the
relevant results
 Computational time
• Time taken to build a classifier and classify an image
* Wikipedia: Precision and Recall
Aerodynamic Analysis and Design Lab.
Dataset
13
 INRIA person dataset and ImageNet person dataset
 Training stage : number of positive (human presence) and negative (no
human) images is 1000 each.
 Testing stage : 300 images for positive and 200 images for negative.
 All images are downscaled to a resolution of 128x96. (Trial and error
process)
Aerodynamic Analysis and Design Lab.
Evaluation Methods
14
 Confusion matrix is created for each case.
Predicted class
Human detected
Human not
detected
Actual
class
Human True positives (TP) False negatives (FN)
No human False positives (FP) True negatives (TN)
𝐴𝑐𝑐𝑢𝑟𝑎𝑐𝑦 =
𝑇𝑃 + 𝑇𝑁
𝑇𝑃 + 𝐹𝑃 + 𝑇𝑁 + 𝐹𝑁
𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛 =
𝑇𝑃
𝑇𝑃 + 𝐹𝑃
𝑅𝑒𝑐𝑎𝑙𝑙 =
𝑇𝑃
𝑇𝑃 + 𝐹𝑁
Aerodynamic Analysis and Design Lab.
Evaluation Results
15
 Computational time:
• Training
– 10-seg : 23 minutes
– Original HOG : 8 minutes
• Testing
– 10-seg : 0.64 seconds
– Original HOG : 0.04 seconds
71.2
80.3
80.6
79.8
83.7
82.8
76.6
80.7
80.4
78.8
81
83.2
73.2
84.7
74.3
AC C UR AC Y PR EC ISION R EC ALL
%
10-seg 100-seg 200-seg 300-seg original HOG
Aerodynamic Analysis and Design Lab.
Results
16
Aerodynamic Analysis and Design Lab.
Conclusion
17
 It presents an approach to detect human in images uniquely by
superpixelwise HOG calculations each channel of LUV color space.
 HOG descriptors are fed into AdaBoost to classify images into two
categories : human-detected and human-not-detected.
 For performance measure, accuracy, precision, recall and computational
time are used.
 Four experiments including 10-segmented, 100-segmented, 200-
segmented, and 300-segmented are performed to compare results
against the original HOG.
 Except 10-segmented one, the other superpixelwise approaches
outperform the orginal HOG in terms of accuracy by 3% or more.
Aerodynamic Analysis and Design Lab.
References
18
 All references described in my paper
 http://www.learnopencv.com/histogram-of-oriented-gradients/
 https://www.mathsisfun.com/polar-cartesian-coordinates.html

More Related Content

What's hot

Object detection
Object detectionObject detection
Object detectionSomesh Vyas
 
RAIL: Risk-Averse Imitation Learning | Invited talk at Intel AI Workshop at K...
RAIL: Risk-Averse Imitation Learning | Invited talk at Intel AI Workshop at K...RAIL: Risk-Averse Imitation Learning | Invited talk at Intel AI Workshop at K...
RAIL: Risk-Averse Imitation Learning | Invited talk at Intel AI Workshop at K...Anirban Santara
 
Object Tracking By Online Discriminative Feature Selection Algorithm
Object Tracking By Online Discriminative Feature Selection AlgorithmObject Tracking By Online Discriminative Feature Selection Algorithm
Object Tracking By Online Discriminative Feature Selection AlgorithmIRJET Journal
 
Breaking the Nonsmooth Barrier: A Scalable Parallel Method for Composite Opti...
Breaking the Nonsmooth Barrier: A Scalable Parallel Method for Composite Opti...Breaking the Nonsmooth Barrier: A Scalable Parallel Method for Composite Opti...
Breaking the Nonsmooth Barrier: A Scalable Parallel Method for Composite Opti...Fabian Pedregosa
 
IDA 2015: Efficient model selection for regularized classification by exploit...
IDA 2015: Efficient model selection for regularized classification by exploit...IDA 2015: Efficient model selection for regularized classification by exploit...
IDA 2015: Efficient model selection for regularized classification by exploit...George Balikas
 

What's hot (6)

Object detection
Object detectionObject detection
Object detection
 
RAIL: Risk-Averse Imitation Learning | Invited talk at Intel AI Workshop at K...
RAIL: Risk-Averse Imitation Learning | Invited talk at Intel AI Workshop at K...RAIL: Risk-Averse Imitation Learning | Invited talk at Intel AI Workshop at K...
RAIL: Risk-Averse Imitation Learning | Invited talk at Intel AI Workshop at K...
 
Object Tracking By Online Discriminative Feature Selection Algorithm
Object Tracking By Online Discriminative Feature Selection AlgorithmObject Tracking By Online Discriminative Feature Selection Algorithm
Object Tracking By Online Discriminative Feature Selection Algorithm
 
Breaking the Nonsmooth Barrier: A Scalable Parallel Method for Composite Opti...
Breaking the Nonsmooth Barrier: A Scalable Parallel Method for Composite Opti...Breaking the Nonsmooth Barrier: A Scalable Parallel Method for Composite Opti...
Breaking the Nonsmooth Barrier: A Scalable Parallel Method for Composite Opti...
 
IDA 2015: Efficient model selection for regularized classification by exploit...
IDA 2015: Efficient model selection for regularized classification by exploit...IDA 2015: Efficient model selection for regularized classification by exploit...
IDA 2015: Efficient model selection for regularized classification by exploit...
 
Imitation Learning
Imitation LearningImitation Learning
Imitation Learning
 

Similar to Using HOG Descriptors on Superpixels for Human Detection of UAV Imagery

机器学习Adaboost
机器学习Adaboost机器学习Adaboost
机器学习AdaboostShocky1
 
Avihu Efrat's Viola and Jones face detection slides
Avihu Efrat's Viola and Jones face detection slidesAvihu Efrat's Viola and Jones face detection slides
Avihu Efrat's Viola and Jones face detection slideswolf
 
A scalable collaborative filtering framework based on co clustering
A scalable collaborative filtering framework based on co clusteringA scalable collaborative filtering framework based on co clustering
A scalable collaborative filtering framework based on co clusteringAllenWu
 
EE660_Report_YaxinLiu_8448347171
EE660_Report_YaxinLiu_8448347171EE660_Report_YaxinLiu_8448347171
EE660_Report_YaxinLiu_8448347171Yaxin Liu
 
Artem Baklanov - Votes Aggregation Techniques in Geo-Wiki Crowdsourcing Game:...
Artem Baklanov - Votes Aggregation Techniques in Geo-Wiki Crowdsourcing Game:...Artem Baklanov - Votes Aggregation Techniques in Geo-Wiki Crowdsourcing Game:...
Artem Baklanov - Votes Aggregation Techniques in Geo-Wiki Crowdsourcing Game:...AIST
 
Visual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learningVisual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learningBenjamin Bengfort
 
Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners ...
Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners ...Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners ...
Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners ...Simplilearn
 
Super Resolution of Image
Super Resolution of ImageSuper Resolution of Image
Super Resolution of ImageSatheesh K
 
Deep Learning in Computer Vision
Deep Learning in Computer VisionDeep Learning in Computer Vision
Deep Learning in Computer VisionSungjoon Choi
 
Realtime pothole detection system using improved CNN Models
Realtime pothole detection system using improved CNN ModelsRealtime pothole detection system using improved CNN Models
Realtime pothole detection system using improved CNN Modelsnithinsai2992
 
Deep_Learning__INAF_baroncelli.pdf
Deep_Learning__INAF_baroncelli.pdfDeep_Learning__INAF_baroncelli.pdf
Deep_Learning__INAF_baroncelli.pdfasdfasdf214078
 
A multilevel automatic thresholding method based on a genetic algorithm for a...
A multilevel automatic thresholding method based on a genetic algorithm for a...A multilevel automatic thresholding method based on a genetic algorithm for a...
A multilevel automatic thresholding method based on a genetic algorithm for a...Akshit Arora
 
Human action recognition with kinect using a joint motion descriptor
Human action recognition with kinect using a joint motion descriptorHuman action recognition with kinect using a joint motion descriptor
Human action recognition with kinect using a joint motion descriptorSoma Boubou
 
Visualizing the Model Selection Process
Visualizing the Model Selection ProcessVisualizing the Model Selection Process
Visualizing the Model Selection ProcessBenjamin Bengfort
 
Towards Accurate Multi-person Pose Estimation in the Wild (My summery)
Towards Accurate Multi-person Pose Estimation in the Wild (My summery)Towards Accurate Multi-person Pose Estimation in the Wild (My summery)
Towards Accurate Multi-person Pose Estimation in the Wild (My summery)Abdulrahman Kerim
 
Recognition of anaerobic based on machine learning using smart watch sensor data
Recognition of anaerobic based on machine learning using smart watch sensor dataRecognition of anaerobic based on machine learning using smart watch sensor data
Recognition of anaerobic based on machine learning using smart watch sensor dataSuhyun Cho
 
Face Detection System on Ada boost Algorithm Using Haar Classifiers
Face Detection System on Ada boost Algorithm Using Haar ClassifiersFace Detection System on Ada boost Algorithm Using Haar Classifiers
Face Detection System on Ada boost Algorithm Using Haar ClassifiersIJMER
 
Sift based arabic sign language recognition aecia 2014 –november17-19, addis ...
Sift based arabic sign language recognition aecia 2014 –november17-19, addis ...Sift based arabic sign language recognition aecia 2014 –november17-19, addis ...
Sift based arabic sign language recognition aecia 2014 –november17-19, addis ...Tarek Gaber
 

Similar to Using HOG Descriptors on Superpixels for Human Detection of UAV Imagery (20)

机器学习Adaboost
机器学习Adaboost机器学习Adaboost
机器学习Adaboost
 
Avihu Efrat's Viola and Jones face detection slides
Avihu Efrat's Viola and Jones face detection slidesAvihu Efrat's Viola and Jones face detection slides
Avihu Efrat's Viola and Jones face detection slides
 
A scalable collaborative filtering framework based on co clustering
A scalable collaborative filtering framework based on co clusteringA scalable collaborative filtering framework based on co clustering
A scalable collaborative filtering framework based on co clustering
 
EE660_Report_YaxinLiu_8448347171
EE660_Report_YaxinLiu_8448347171EE660_Report_YaxinLiu_8448347171
EE660_Report_YaxinLiu_8448347171
 
Artem Baklanov - Votes Aggregation Techniques in Geo-Wiki Crowdsourcing Game:...
Artem Baklanov - Votes Aggregation Techniques in Geo-Wiki Crowdsourcing Game:...Artem Baklanov - Votes Aggregation Techniques in Geo-Wiki Crowdsourcing Game:...
Artem Baklanov - Votes Aggregation Techniques in Geo-Wiki Crowdsourcing Game:...
 
Visual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learningVisual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learning
 
Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners ...
Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners ...Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners ...
Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners ...
 
Super Resolution of Image
Super Resolution of ImageSuper Resolution of Image
Super Resolution of Image
 
All projects
All projectsAll projects
All projects
 
Deep Learning in Computer Vision
Deep Learning in Computer VisionDeep Learning in Computer Vision
Deep Learning in Computer Vision
 
Realtime pothole detection system using improved CNN Models
Realtime pothole detection system using improved CNN ModelsRealtime pothole detection system using improved CNN Models
Realtime pothole detection system using improved CNN Models
 
Deep_Learning__INAF_baroncelli.pdf
Deep_Learning__INAF_baroncelli.pdfDeep_Learning__INAF_baroncelli.pdf
Deep_Learning__INAF_baroncelli.pdf
 
A multilevel automatic thresholding method based on a genetic algorithm for a...
A multilevel automatic thresholding method based on a genetic algorithm for a...A multilevel automatic thresholding method based on a genetic algorithm for a...
A multilevel automatic thresholding method based on a genetic algorithm for a...
 
Human action recognition with kinect using a joint motion descriptor
Human action recognition with kinect using a joint motion descriptorHuman action recognition with kinect using a joint motion descriptor
Human action recognition with kinect using a joint motion descriptor
 
Visualizing the Model Selection Process
Visualizing the Model Selection ProcessVisualizing the Model Selection Process
Visualizing the Model Selection Process
 
Towards Accurate Multi-person Pose Estimation in the Wild (My summery)
Towards Accurate Multi-person Pose Estimation in the Wild (My summery)Towards Accurate Multi-person Pose Estimation in the Wild (My summery)
Towards Accurate Multi-person Pose Estimation in the Wild (My summery)
 
Recognition of anaerobic based on machine learning using smart watch sensor data
Recognition of anaerobic based on machine learning using smart watch sensor dataRecognition of anaerobic based on machine learning using smart watch sensor data
Recognition of anaerobic based on machine learning using smart watch sensor data
 
Face Detection System on Ada boost Algorithm Using Haar Classifiers
Face Detection System on Ada boost Algorithm Using Haar ClassifiersFace Detection System on Ada boost Algorithm Using Haar Classifiers
Face Detection System on Ada boost Algorithm Using Haar Classifiers
 
Sift based arabic sign language recognition aecia 2014 –november17-19, addis ...
Sift based arabic sign language recognition aecia 2014 –november17-19, addis ...Sift based arabic sign language recognition aecia 2014 –november17-19, addis ...
Sift based arabic sign language recognition aecia 2014 –november17-19, addis ...
 
Neural networks
Neural networksNeural networks
Neural networks
 

Recently uploaded

Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machinePadma Pradeep
 
Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)Allon Mureinik
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure servicePooja Nehwal
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Alan Dix
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
Benefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksBenefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksSoftradix Technologies
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Patryk Bandurski
 
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your Budget
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your BudgetHyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your Budget
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your BudgetEnjoy Anytime
 
Key Features Of Token Development (1).pptx
Key  Features Of Token  Development (1).pptxKey  Features Of Token  Development (1).pptx
Key Features Of Token Development (1).pptxLBM Solutions
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersThousandEyes
 
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticsKotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticscarlostorres15106
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024Scott Keck-Warren
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsMemoori
 
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions
 

Recently uploaded (20)

Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
 
Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
Vulnerability_Management_GRC_by Sohang Sengupta.pptx
Vulnerability_Management_GRC_by Sohang Sengupta.pptxVulnerability_Management_GRC_by Sohang Sengupta.pptx
Vulnerability_Management_GRC_by Sohang Sengupta.pptx
 
Benefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksBenefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other Frameworks
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
 
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your Budget
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your BudgetHyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your Budget
Hyderabad Call Girls Khairatabad ✨ 7001305949 ✨ Cheap Price Your Budget
 
Key Features Of Token Development (1).pptx
Key  Features Of Token  Development (1).pptxKey  Features Of Token  Development (1).pptx
Key Features Of Token Development (1).pptx
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
 
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptxE-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
 
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticsKotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial Buildings
 
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping Elbows
 

Using HOG Descriptors on Superpixels for Human Detection of UAV Imagery

  • 1. Aerodynamic Analysis and Design Lab. WAI NWE TUN August 16, 2017 Using HOG Descriptors on Superpixels for Human Detection of UAV Imagery AADL AI Seminar
  • 2. Aerodynamic Analysis and Design Lab. Contents  Introduction  System architecture • Superpixel extraction • HOG descriptor calculation • Adaboost classifier  Performance evaluations  Dataset  Evaluation results  Conclusion  References 2
  • 3. Aerodynamic Analysis and Design Lab. Introduction 3  Human detection is a challenging task, with many applications such as pedestrian detection, search and rescue operations, surveillance.  Object detection process is generally performed in two main steps: feature extraction and classification.  Feature extraction maps image windows to a fixed size feature space that robustly encodes visual form.  Feature vectors are fed into a classifier to train.  Using the trained classifier, the images can be determined if desired object is present or not.  This system presented human detection using image features by HOG descriptors for each superpixel by Simple Linear Iterative Clustering (SLIC) and HOG features descriptors are classified by AdaBoost.
  • 4. Aerodynamic Analysis and Design Lab. Introduction 4  Superpixel algorithms group pixels into perceptually meaningful atomic regions which can be used to replace the rigid structure of the pixel grid.  They capture image redundancy, provide a convenient primitive from which to compute image features, and greatly reduce the complexity of subsequent image processing tasks. SLIC adapts k-means clustering to generate superpixels.  Feature descriptor is a representation of an image or an image patch that simplifies the image by extracting useful information and throwing away extraneous information.  In HOG descriptors, the distribution of directions of gradient are used to calculate features.  AdaBoost classifier is an ensemble type which uses a combination of models consisting of a series of k weak classifiers.
  • 5. Aerodynamic Analysis and Design Lab. System Architecture 5
  • 6. Aerodynamic Analysis and Design Lab. Features Extraction 6
  • 7. Aerodynamic Analysis and Design Lab. Feature Extraction 7  Sobel filters  Gradient magnitude and direction are obtained from converting x, y coordinates to polar coordinates
  • 8. Aerodynamic Analysis and Design Lab. Superpixel Extraction by SLIC 8  Instead of fixed-sized block in original HOG, superpixels are used in HOG calculation.  Superpixels are a group of consecutive pixels which share common characteristics such as intensity and represent meaningful region.  Superpixel algorithm • graph-based : each pixel as a node, node similarity as edge weight, minimize cost function over graph. • gradient ascent methods: clustering  SLIC : gradient ascent method  Distance measure is calculated by normalizing the two measures: color proximity (LAB) and spatial proximity (XY).  K-means clustering is performed using the distance measure for each limited region.
  • 9. Aerodynamic Analysis and Design Lab. HOG Descriptors Calculation 9  Histograms of gradients are created for each superpixel group and concatenated all. Then, L2 Normalization is performed.
  • 10. Aerodynamic Analysis and Design Lab. AdaBoost Classifier: Learning 10  Given D, a data set of d class-labeled tuples, (𝑋1, 𝑦1), .. (𝑋 𝑑, 𝑦𝑛) where X = HOG descriptors, y=human or not, d = number of images, n= 2 (number of class labels).  Adaboost assigns each training tuple an equal weight of 1 𝑑 .  Generating k classifiers requires k rounds/iterations  In round i, tuples from D are sampled to form a training set, 𝐷𝑖, of size d (with replacement).  Each tuple’s chance of being selected is based on its weight.  A classifier model, 𝑀𝑖, is derived from the training tuples, 𝐷𝑖 and its error is calculated.  If a tuple was incorrectly classified, its weight is increased. Otherwise, it is decreased. This is to focus more on the misclassified tuples. 𝑒𝑟𝑟𝑜𝑟 𝑀𝑖 = 𝑗 𝑑 𝑤𝑗 × 𝑒𝑟𝑟(𝑋𝑗
  • 11. Aerodynamic Analysis and Design Lab. AdaBoost Classifer: Prediction 11  Instead of an equal vote, a weight to each classifier’s vote is assigned, based on how well the classifier performed.  The lower a classifier’s error rate, the more accurate it is and the higher its weight for voting should be.  For each class c (e.g., human presence or not), weights of each classifer that assigned class c to ‘that’ tuple.  The class with the highest sum is the ‘winner’ and is returned as the class prediction for ‘that’ tuple. 𝑣𝑜𝑡𝑒 𝑀𝑖 = 1 − 𝑒𝑟𝑟𝑜𝑟 𝑀𝑖 𝑒𝑟𝑟𝑜𝑟 𝑀𝑖
  • 12. Aerodynamic Analysis and Design Lab. Performance Evaluations 12  Performance is evaluated using four measures: accuracy, precision, recall and computational time.  Accuracy • Overall accuracy of a system  Precision (Correctness) • Express how many selected items are relevant • High precision means that a system returned more relevant results than irrelevant ones  Recall (Completeness) • Express how many relevant items are selected • High recall means that a system returned most of the relevant results  Computational time • Time taken to build a classifier and classify an image * Wikipedia: Precision and Recall
  • 13. Aerodynamic Analysis and Design Lab. Dataset 13  INRIA person dataset and ImageNet person dataset  Training stage : number of positive (human presence) and negative (no human) images is 1000 each.  Testing stage : 300 images for positive and 200 images for negative.  All images are downscaled to a resolution of 128x96. (Trial and error process)
  • 14. Aerodynamic Analysis and Design Lab. Evaluation Methods 14  Confusion matrix is created for each case. Predicted class Human detected Human not detected Actual class Human True positives (TP) False negatives (FN) No human False positives (FP) True negatives (TN) 𝐴𝑐𝑐𝑢𝑟𝑎𝑐𝑦 = 𝑇𝑃 + 𝑇𝑁 𝑇𝑃 + 𝐹𝑃 + 𝑇𝑁 + 𝐹𝑁 𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛 = 𝑇𝑃 𝑇𝑃 + 𝐹𝑃 𝑅𝑒𝑐𝑎𝑙𝑙 = 𝑇𝑃 𝑇𝑃 + 𝐹𝑁
  • 15. Aerodynamic Analysis and Design Lab. Evaluation Results 15  Computational time: • Training – 10-seg : 23 minutes – Original HOG : 8 minutes • Testing – 10-seg : 0.64 seconds – Original HOG : 0.04 seconds 71.2 80.3 80.6 79.8 83.7 82.8 76.6 80.7 80.4 78.8 81 83.2 73.2 84.7 74.3 AC C UR AC Y PR EC ISION R EC ALL % 10-seg 100-seg 200-seg 300-seg original HOG
  • 16. Aerodynamic Analysis and Design Lab. Results 16
  • 17. Aerodynamic Analysis and Design Lab. Conclusion 17  It presents an approach to detect human in images uniquely by superpixelwise HOG calculations each channel of LUV color space.  HOG descriptors are fed into AdaBoost to classify images into two categories : human-detected and human-not-detected.  For performance measure, accuracy, precision, recall and computational time are used.  Four experiments including 10-segmented, 100-segmented, 200- segmented, and 300-segmented are performed to compare results against the original HOG.  Except 10-segmented one, the other superpixelwise approaches outperform the orginal HOG in terms of accuracy by 3% or more.
  • 18. Aerodynamic Analysis and Design Lab. References 18  All references described in my paper  http://www.learnopencv.com/histogram-of-oriented-gradients/  https://www.mathsisfun.com/polar-cartesian-coordinates.html