Internet of moving target detection method based on nonparametric background model

In traffic surveillance system, mobile target detection and identification is the key technology in traffic surveillance system. In this paper, one detection method based on non-parametric background model is adopted on the basis of the summary of previous background modeling. In the model, a series of sampling values are used to estimate and observe probability model of pixel points; and then, the probability model is used for binarization detection of mobile targets. In the end, we have brought favorable detection effects by noise suppression treatment. As for identification of mobile targets, several features are proposed in this paper and neural network is used for identification training. Experiment results show that classification of pedestrian and vehicle targets according to these features has a high rate of identification.