Title :
A Combined Pedestrian Detection Method Based on Haar-Like Features and HOG Features
Author :
Yuan Xin ; Shan Xiaosen ; Su Li
Author_Institution :
Coll. of Autom., Harbin Eng. Univ., Harbin, China
Abstract :
Considering the simplicity and fast training speed of Haar-like features, the high detecting precision of HOG features, a combined method is proposed on the basis of the two features. Several rectangular features which can describe local human characteristics based on original features are added. The combined method can retain the precision of HOG features and increase the speed of detection at the same time. The experiments have shown this method can detect the pedestrians quickly and effectively with high precision.
Keywords :
Haar transforms; feature extraction; image classification; learning (artificial intelligence); object detection; support vector machines; traffic engineering computing; AdaBoost learning method; HOG features; Haar-like features; classifier training; detection speed; histograms of oriented gradients; human characteristics; pedestrian detection method; rectangular features; support vector machine; Accuracy; Feature extraction; Histograms; Humans; Pixel; Support vector machines; Training;
Conference_Titel :
Intelligent Systems and Applications (ISA), 2011 3rd International Workshop on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-9855-0
Electronic_ISBN :
978-1-4244-9857-4
DOI :
10.1109/ISA.2011.5873387
