A Pedestrian Re-Identification Method Based on Multi-Feature Fusion

Single feature of pedestrian is difficult to accurately describe the target using traditional algorithms. A new reidentification algorithm combing global features and local features with different distance metric function is introduced. First, weighted color histogram feature for whole pedestrian is extracted and combined with Bhattacharyya distance to roughly recognize targets. Then pedestrians’ maximum stable color area (MSCR) of torso and legs and histograms of oriented gradients (HOG) are combined with weights for fusion feature. Finally, all obtained features are combined and compared with pedestrian original characteristics for final recognition. Pearson correlation coefficient is used as similarity measurement to finely recognize pedestrian. Experimental results show that proposed method can achieve high identification accuracy.