Moving Target Analysis Method Based On Frog Vision Characteristics

Moving target analysis technology is a challenging issue due to the lack ofmature and efficient computational model and method. Computer is far less than somebiological visual system with visual specialty in terms of the ability of perceiving,understanding and analyzing moving target. Therefore, it is significant to exploreefficient moving target analysis technology through learning from biological visualinformation processing mechanism. This article studies moving target analysismethod based on frog visual characteristics. The main work is as follows:According to the discussion on several typical receptive field model, either theclassic or non-classical model, their destination are both to describe the physiologicalmechanism of a biological visual information processing system, so as to be betterapplied to the actual situation. From this point of view, an asymmetric anisotropicreceptive field model based on frog visual characteristics is proposed. And combingwith the spatiotemporal dynamics of frog eye receptive field, the model is used for thefilter analysis of sequence images. The experiment result shows that the method has awonderful performance for blurring the background and highlighting moving target,which meets the frog eye visual characteristics.Considering that frog eye is sensitive to the moving target, and can identifies andtracks target rapidly and continuously according to the contour of moving target, acorresponding contour tracking method is put forward. The main idea follows as:firstly, the initial contour curve with the time-domain difference between threesuccessive frames is gained; then a spatial-temporal energy model is establishedaccording to the spatial-temporal correlation of video sequences, and thecorresponding level set expression is got, and finally the target contour is approachedby the evolution of the level set function. Experimental results show that target shapecan keep stable as the change of contour because the energy curve contains motionand region information.