Pedestrian Detection for Transformer Substation Based on Gaussian Mixture Model and YOLO
Qiwei Peng, Luo Wang, Gongyi Hong, Min Feng, Xia Yuan, Lei Yu, Xiaolong Hao, Xu Wang, Mingxuan Li · 2016
Safety is a core requirement of the transformer substation where is dangerous due to high voltage. It requires to detect pedestrians efficiently based on the surveillance video near the substation to ensure the safety of the pedestrian and the device. This paper presents a new method of pedestrian detection for transformer substation based on Gaussian Mixture Model (GMM) and YOLO. We use GMM to model the background and detect the pedestrians preliminarily, at the same time, YOLO, a kind of detection method based on convolution neural network (CNN), is also applied for pedestrian detection. Through combining the two results with different weights, the network outputs a better detection result. Our extensive experiments show that the work reaches 20% higher than the single method.