Pedestrian detection based on gradient and texture feature integration

Feature-based pedestrian detection method is currently the mainstream direction to solve the problem of pedestrian detection. In this kind of method, whether the appropriate feature can be extracted is the key to the comprehensive performance of the whole pedestrian detection system. It is believed that the appearance of a pedestrian can be better captured by the combination of edge/local shape feature and texture feature. In this field, the current method is to simply concatenate HOG (histogram of oriented gradient) features and LBP (local binary pattern) features extracted from an image to produce a new feature with large dimension. This kind of method achieves better performance at the cost of increasing the number of features. In this paper, Choquet integral based on the signed fuzzy measure is introduced to fuse HOG and LBP descriptors in parallel that is expected to improve accuracy without increasing feature dimensions. The parameters needed in the whole fusion process are optimized by a training algorithm based on genetic algorithm. This architecture has three advantages. Firstly, because the fusion of HOG and LBP features is parallel, the dimensions of the new features are not increased. Secondly, the speed of feature fusion is fast, thus reducing the time of pedestrian detection. Thirdly, the new features after fusion have the advantages of HOG and LBP features, which is helpful to improve the detection accuracy. The series of experimentation with the architecture proposed in this paper reaches promising and satisfactory results.

Download Full-text

Haar-Like Multi-Granularity Texture Features for Pedestrian Detection

International Journal of Image and Graphics ◽

10.1142/s0219467817500231 ◽

2017 ◽

Vol 17 (04) ◽

pp. 1750023 ◽

Cited By ~ 2

Author(s):

Ruzhong Cheng ◽

Yongjun Zhang ◽

Guoping Wang ◽

Yong Zhao ◽

Rahmatulloev Khusravsho

Keyword(s):

Human Body ◽

Pedestrian Detection ◽

Texture Feature ◽

Texture Features ◽

Surveillance Systems ◽

Robust Detection ◽

Driver Support Systems ◽

Haar Feature ◽

Feature Dimension ◽

Self Driving Cars

Pedestrian detection has been a significant problem for decades and remains a hot topic in computer vision. Pedestrian detection is one of the key algorithms for self-driving cars and some other functions in robotics, including driver support systems, road surveillance systems. In this paper, based on the characteristics of the human body and the Haar feature, the Haar-like multi-granularity local texture feature, i.e., multi-granularity Haar-like LBP (mgh-LBP), is proposed for pedestrian detection. The mgh-LBP feature combines four characteristics of the human body and their backgrounds to construct the Haar-like features, which can better describe human body texture and edge information. Compared with other texture features, including the rotation-invariant LBP feature, uniform LBP feature and basic-LBP feature, the proposed method greatly reduces the feature dimension and computational complexity, and obtains a higher pedestrian detection rate and robust detection performance.

Download Full-text