Research for pedestrian detection classifier

摘要

The image features play an important role in object detection and recognition, the method of object detection based on Histogram of Oriented Gradient(HOG) for its good detection result receives extensive attention to many researchers. This article describes the approaches to Histogram of Oriented Gradient and support vector machine, focusing on studying the HOG feature and application, detailing the process of HOG feature extracted and the design of classifiers in pedestrian detection. In this paper, pedestrian sample library was built and the Histogram of Oriented Gradient feature of the samples was extracted. The pedestrian classifier was obtained by using support vector machine and the performance of the classifier was tested.