%0 Journal Article
%T Study of the Video Monitoring System Image Recognition Solutions Based on Mathematic models
%A Peilong Xu
%J International Journal of Computer Science Issues
%D 2013
%I IJCSI Press
%X objective: Through establishment a set of image recognition system based on mathematic models, to develop a auto alarm solution for the video monitoring system. Methods: compare the images the video monitoring system collected according to the time sequences. Then after binaryzation and wave filtering, the images were converted into numerical values using autocorrelation function, and the alarm threshold value was confirmed by experiences. Results: Through experiments, the change ratios of the two images before and after image processing were inversely proportional to the autocorrelation function. When the function value is less than 0.8, it indicates that there is an object volumes larger than 1m3 has invaded into 15m distances, and when the function value is less than 0.6, it indicates that there is an object volumes larger than 1m3 has invaded into 30m distances. Conclusion: Through calculation of autocorrelation functions, auto alarm for the images collected by video monitoring system could be effectively realized.
%K Video monitoring system
%K Autocorrelation function
%K Image processing Mathematic models.
%K IJCSI
%U http://www.ijcsi.org/papers/IJCSI-10-1-2-511-515.pdf