scispace - formally typeset
Search or ask a question
Topic

Histogram of oriented gradients

About: Histogram of oriented gradients is a research topic. Over the lifetime, 2037 publications have been published within this topic receiving 55881 citations. The topic is also known as: HOG.


Papers
More filters
Journal ArticleDOI
TL;DR: This paper analyzes the performance of various texture analysis methods for the purpose of reducing the number of false positives in breast cancer detection and proposes the use of local directional number patterns as a new feature extraction method for breast mass detection.
Abstract: In this paper we analyse the performance of various texture analysis methods for the purpose of reducing the number of false positives in breast cancer detection; as a result, the cost of breast cancer diagnosis would be reduced. We consider well-known methods such as local binary patterns, histogram of oriented gradients, co-occurrence matrix features and Gabor filters. Moreover, we propose the use of local directional number patterns as a new feature extraction method for breast mass detection. For each method, different classifiers are trained on the extracted features to predict the class of unknown instances. In order to improve the mass detection capability of each individual method, we use feature combination techniques and classifier majority voting. Some experiments were performed on the images obtained from a public breast cancer database, achieving promising levels of sensitivity and specificity.

21 citations

Posted Content
TL;DR: In this paper, an individualised baseline micro-movement detection method using 3D Histogram of Oriented Gradients (3D HOG) temporal difference method was proposed.
Abstract: Micro-facial expressions are regarded as an important human behavioural event that can highlight emotional deception. Spotting these movements is difficult for humans and machines, however research into using computer vision to detect subtle facial expressions is growing in popularity. This paper proposes an individualised baseline micro-movement detection method using 3D Histogram of Oriented Gradients (3D HOG) temporal difference method. We define a face template consisting of 26 regions based on the Facial Action Coding System (FACS). We extract the temporal features of each region using 3D HOG. Then, we use Chi-square distance to find subtle facial motion in the local regions. Finally, an automatic peak detector is used to detect micro-movements above the newly proposed adaptive baseline threshold. The performance is validated on two FACS coded datasets: SAMM and CASME II. This objective method focuses on the movement of the 26 face regions. When comparing with the ground truth, the best result was an AUC of 0.7512 and 0.7261 on SAMM and CASME II, respectively. The results show that 3D HOG outperformed for micro-movement detection, compared to state-of-the-art feature representations: Local Binary Patterns in Three Orthogonal Planes and Histograms of Oriented Optical Flow.

21 citations

Journal ArticleDOI
TL;DR: This paper extracted the histogram of oriented gradients feature and local binary pattern feature from the original images respectively and K-singular value decomposition was used to extract sparse representation features from the HOG and LBP features.

21 citations

Journal ArticleDOI
TL;DR: The results show that the combination of the tracking algorithm and the face recognition algorithm not only tracks the person but also recognizes the person, which makes it best suit for video surveillance applications.
Abstract: Tracking of human and recognition in public places using surveillance cameras is the topic of research in the area computer vision. Recognition of human and then tracking completes the video surveillance system. A novel algorithm for face recognition and human tracking is presented in this article. Human is tracked using Gaussian mixture model. To track the human in specific, template of GMM is divided into four regions which are placed one above the other and tracked simultaneously. For recognizing the human, the histogram of oriented gradients features of the face region are given to the support vector machine classifier. Three experiments are conducted in taking the training faces. Every $$10{\mathrm{th}}$$ frame, every $$5{\mathrm{th}}$$ frame and every $$3{\mathrm{rd}}$$ frame of the first 100 frames are considered. The other frames in the video are considered for testing using SVM classifier. Three datasets namely AITAM1 (simple), AITAM2 (moderate) and AITAM3 (complex) are used in this work. The experimental results show that as the complexity of dataset increases the performance metrics are getting decreased. The more the number of training faces in preparing a classifier, the better is the face recognition rate. This is experimented for all types of datasets. The Performance results show that the combination of the tracking algorithm and the face recognition algorithm not only tracks the person but also recognizes the person. This unique property of both tracking and recognition makes it best suit for video surveillance applications.

21 citations

Proceedings ArticleDOI
14 Dec 2013
TL;DR: A vehicle and pedestrian detection system is built up by combing Histogram of Oriented Gradients feature and support vector machine (SVM), which provides a reasonable and feature invariant object representation and gives a robust classifier that can control both the training set error and the classifier's complexity.
Abstract: Vehicle and Pedestrian Detection is a key problem in computer vision, with several applications including robotics, surveillance and automotive safety. Much of the progress of the past few years has been driven by the availability of challenging public datasets. In this paper, we build up a vehicle and pedestrian detection system by combing Histogram of Oriented Gradients (HoG) feature and support vector machine (SVM). HoG feature provides a reasonable and feature invariant object representation, while SVM framework gives us a robust classifier that can control both the training set error and the classifier's complexity. A detailed system architecture design is presented and the testing experiments show that high performance in both accuracy and speed can be achieved by the developed system.

21 citations


Network Information
Related Topics (5)
Feature extraction
111.8K papers, 2.1M citations
89% related
Convolutional neural network
74.7K papers, 2M citations
87% related
Deep learning
79.8K papers, 2.1M citations
87% related
Image segmentation
79.6K papers, 1.8M citations
87% related
Feature (computer vision)
128.2K papers, 1.7M citations
86% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
202356
2022181
2021116
2020189
2019179
2018240