• PDF / 484,700 Bytes
• 11 Pages / 439.37 x 666.142 pts Page_size
• 29 Downloads / 198 Views

DOWNLOAD

REPORT

Abstract Human recognition techniques are used in many areas such as video surveillance, human action recognition, automobile industry for pedestrian detection, etc. The research on human recognition is widely going on and is open due to typical challenges in human detection. Histogram-based human detection methods are popular because of its better detection rate than other approaches. Histograms of oriented gradients (HOG) and co-occurrence of histogram-oriented gradients (CoHOG) are used widely for human recognition. A CoHOG is an extension of HOG and it takes a pair of orientations instead of one. Co-occurrence matrix is computed and histograms are calculated. In CoHOG, gradient directions alone are considered and magnitude is ignored. In this paper magnitude details are considered to improve detection rate. Magnitude is included to influence the feature vector to achieve better performance than the existing method. In this paper, weighted co-occurrence histograms of oriented gradients (W-CoHOG) is introduced by calculating weighted co-occurrence matrix to include magnitude factor for feature vector. Experiments are conducted on two benchmark datasets, INRIA and Chrysler pedestrian datasets. The experiment results support our approach and shows that our approach has better detection rate.




Keywords Histograms of oriented gradients (HOG) Co-occurrence histogram of oriented gradients (CoHOG) Weighted co-occurrence histogram of oriented gradients (W-CoHOG) Human detection Pedestrian detection










N. Andavarapu (&)
V.K. Vatsavayi Department of Computer Science and Systems Engineering, Andhra University, Visakhapatnam, India e-mail: [email protected] V.K. Vatsavayi e-mail: [email protected] © Springer Science+Business Media Singapore 2016 M. Pant et al. (eds.), Proceedings of Fifth International Conference on Soft Computing for Problem Solving, Advances in Intelligent Systems and Computing 437, DOI 10.1007/978-981-10-0451-3_26

273

274

N. Andavarapu and V.K. Vatsavayi

1 Introduction Computer vision is a wide and emerging area over the past few years. The analysis of images involving humans comes under computer vision problem. Human detection techniques are used in many areas such as people abnormal behavior monitoring, robots, automobile safety systems, and gait recognition. The main goal of a human detector is to check whether humans are present in the image or not. If human is identified in the particular image then it can be used for further analysis. Human detection is still an open problem. Human detection is one of the active and challenging problems in computer vision, due to different articulations and poses, different types of appearances of clothes and accessories acting as occlusions. In this paper humans are identified in a static image. Identifying humans in a static image is more difficult than in a video sequence because no motion and background information is available to provide clues to approximate human position. In our approach, input of the human detector is an image and output i

Recommend Documents

Glowing Window-Based Feature Extraction Technique for Object Detection

170 28 20MB

Feature Extraction Methods for Real-Time Face Detection and Classification

184 55 1MB

Feature Extraction

237 51 239KB

Human body flexibility fitness test based on image edge detection and feature point extraction

182 74 2MB

Spatial-Temporal Feature Fusion for Human Fall Detection

177 26 54MB

Convergence Feature Extraction

186 19 47MB

Non-local gait feature extraction and human identification

153 67 1MB

Weed Detection Approach Using Feature Extraction and KNN Classification

197 105 44MB

Linear Feature Extraction

221 19 47MB

Audio Feature Extraction

224 115 2MB

Shape Feature Extraction

193 25 1MB

Feature Extraction, Abstract

195 6 239KB