Research On Detection Method Of Moving Target Based On Visual Characteristics Of Frog Eyes

The wireless video sensor network can analyze the collected monitoring information in real time and remotely transmit the valid signal to the control center,which improves the intelligence level of the monitoring system.Detection of moving targets is a key technology in monitoring networks,effective detection algorithms can not only provide reliable information for later intelligent system analysis and identification of targets,but also can reduce the pressure of the camera sensor nodes to transmit information.Through research,it has been found that some organisms with visual specialty have a simple and efficient visual system to realize the ability to perceive,understand and analyze moving objects.Therefore,it is of great significance to explore high-efficiency moving target detection technology by referring to the biological visual information processing mechanism that has sports-visual specialty.This paper focuses on the detection of moving targets,a bionic algorithm based on the visual characteristics of the frog eye is designed and implemented,and the task of detecting moving targets in a complex application scenario is completed.The main research work of the thesis is as follows:First of all,this paper analyzes the difficulties and deficiencies of traditional theories in solving related problems.By studying the visual information processing and cognitive mechanism of the frog eye which has sports visual specialty,new ideas and new methods for detecting moving targets are explored.Secondly,through the analysis of the role of different detectors in the frog eye neurons,video image processing methods are used to achieve the functions of visual nerve cells including motion detectors and edge detectors.Among them,in order to simulate the motion detector,This paper studies four methods of moving target detection,and introduces,summarizes and analyzes the principles,advantages and disadvantages of the four algorithms.By Using RPCA to restore the robustness of the background in dynamic scenes,it is combined with the frame difference method to improve the detection of significant regions of moving objects.Experimental results show that compared with the traditional methods,the improved algorithm in this paper improves the accuracy of detection in dynamic scenes.Thirdly,in order to simulate the function of the edge detector in frog-eye neurons,this paper explores the common methods of edge detection in image processing.By analyzing the deficiencies of the traditional canny edge detection,the traditional Canny edge detection is improved by Otsu threshold adaptively to select the best segmentation threshold.Through comparative analysis of experimental data,the improved algorithm can adaptively select high and low thresholds to improve the ability of traditional canny adaptation.Experimental results show that this method improves the problem of false edges and edge breakpoints in edge detection.Finally,based on the visual information processing method of frog eyes,this paper designs a detection algorithm based on the visual characteristics of frog eyes.The algorithm effectively combines the target's motion information and edge information,and realizes the multi-dimensional features of the dynamic target.And the detection of moving targets is more accurate.Experimental results show that the algorithm has high reliability and robustness in complex dynamic scenes,and the accuracy of detection has also been improved.