DocumentCode :
1833535
Title :
Histogram of Silhouette Direction code: An efficient HOG-based descriptor for accurate human detection
Author :
Wei Yang ; Zhan Song ; Xinyu Wu
Author_Institution :
Shenzhen Inst. of Adv. Technol., Shenzhen, China
fYear :
2012
fDate :
11-14 Dec. 2012
Firstpage :
330
Lastpage :
335
Abstract :
Histograms of Oriented Gradients (HOG) is an effective means for the human detection from image or video. Based on the HOG principle, this paper presents a more efficient method named Histogram of Silhouette Direction (HSD). To extract the silhouette pattern of human body, the average background model is used. A novel function that combined with brightness and color information is proposed to segment the foreground accurately and adaptively. The histogram of direction code is constructed via Freeman Chain Code of Eight Directions (FCCE) along the extracted silhouette and used as the feature descriptor. Compared with traditional HOG descriptor which calculates gradients of all image pixels, the proposed HSD descriptor can reduce the feature dimension and whole computation greatly. Experimental results with real video sequences show its improvements in both segmentation efficiency and accuracy.
Keywords :
feature extraction; image colour analysis; image segmentation; image sequences; object detection; statistical analysis; video signal processing; FCCE; Freeman chain code-of-eight directions; HOG descriptor; HOG principle; HOG-based descriptor; HSD method; average background model; brightness information; color information; feature descriptor; histogram-of-oriented gradient; histogram-of-silhouette direction; human detection; image pixel; segmentation accuracy; segmentation efficiency; silhouette pattern extraction; video sequence;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Robotics and Biomimetics (ROBIO), 2012 IEEE International Conference on
Conference_Location :
Guangzhou
Print_ISBN :
978-1-4673-2125-9
Type :
conf
DOI :
10.1109/ROBIO.2012.6490988
Filename :
6490988
Link To Document :
بازگشت