Designing and optimizing the method for pedestrian detection based on Adaboost algorithm

The designing based on Adaboost algorithm not only achieved the nighttime pedestrian detection module of auxiliary driving system, but also realized the system optimization on issues of low detection speed and precision. In this design, variable step length and partition scanning track methods are used to improve the detection speed and variance normalization approach was applied to eliminate the influences to the detection result caused by light factors, moreover, multi-scale fusion technology was utilized to make analysis to the detected rectangular box, in this case, the redundant portion of detection result can be removed, and thus improved the detection rate and reduced the false alarm rate.