Renjun Lin

TL;DR: An improved registration method is proposed to eliminate the camera platform motion, thus techniques in stationary vision surveillance such as frame subtraction can perform well on the registered video to detect moving vehicles rapidly and robustly and is promising for real-time application.

TL;DR: In this paper, a vehicle-mounted pedestrian detection method and system is presented, in which the video image in front of the automobile is acquired by the installed CCD camera on a running automobile, and the pedestrian in the image is detected in real time by a trained classifier, thereby realizing the realtime detection and supporting the safety protection of pedestrians on the intelligent traffic background.

TL;DR: In this paper, a virus evolutionary genetic algorithm based selective ensemble pedestrian detection method was proposed, which includes the following steps: training individual classifiers different from each other by an original training set; initializing a main population of a genetic algorithm and an extra virus population; performing an iterative process of the genetic algorithm; and generating a final combined classifier by an optimal individual obtained by the iterative step.

TL;DR: In this article, the authors proposed a clustering method and a device which consists of selecting clustering centers in a data set and determining a characteristic point set for each clustering center.