Pedestrian Detection Optimization Based on Random Filtering

The large number of surveillance cameras available nowadays in strategic points of major cities provides a safe environment. However, the huge amount of data provided by the cameras prevents the manual processing requiring the application of automated methods. Among such methods, pedestrian detection plays an important role in reducing the amount of data by locating only the regions of interest for further processing regarding activities being performed by agents in the scene. However, pedestrian detection methods currently available are unable to process such large amount of data in real time. Therefore, optimization techniques have to be employed to allow real-time detection even when large volumes of data need to be processed. Focusing on optimization, this work proposes a novel approach that performs a random filtering in the image to discard a large number of detection windows quickly allowing a reduction in the computational cost. Keywords-pedestrian detection optimization; random filtering; visual surveillance;