A robust pedestrian detector based on heterogeneous feature fusion

Pedestrian detection exhibits important application value in driver assistance systems, The detection performance often suffers from the various appearances of pedestrians, the illumination changes and complex background. Aiming at solving these challenges, in this paper, first, a new color moments feature is presented to describe the local similarity structure of pedestrians, which reduces the influence of complicated background. A combination coefficient method is introduced to effectively fuse three heterogeneous features, COLOR, HOG, and LBP, which makes better use of each feature. Then, pedestrians of various poses and views are divided into subclasses with S-Isomap and K-means algorithm. A classifier is trained for each subclass. Finally, with respect to the output values of different subclass classifiers, an equally weighted sum based multi-pose-view ensemble detector is proposed. Experiment results on public datasets demonstrate that the proposed feature combination method significantly improves the description capabilities of pedestrian features. Compared with the existing methods, the proposed detector combining the feature and multi-pose-view ensemble detector boosts the detection accuracy effectively.