A Pedestrian Detection Method Based on MB_LBP Features and Intersection Kernel SVM

Abstract

Listen

Translate article

Pedestrian detection is a hot research topic in pattern recognition and computer vision. We combine MB_LBP (Multiscale Block Local Binary Patterns) feature and Histogram Intersection Kernel SVM and apply them to pedestrian detection. MB_LBP features, which make up for the lack of LBP (Local Binary Patterns) features in robustness, is a kind of effective texture description operator. Histogram Intersection Kernel Support Vector Machine has the advantage of fast classification and high accuracy in object recognition. It can be used for further enhancing the system's real-time performance. The experiments show that the proposed approach has higher precision than the classical algorithm HOG+LinearSVM and the HOG_LBP Features Fusion tested on the established benchmarking datasets—INRIA.