Research On Infrared Target Detection Algorithm Of Criminal Investigation Image Based On Biological Factor Mechanism

With the development of the society,Chinese criminal investigation technology is constantly improving,which ensures and maintains social stability.Facing various cases,timely getting suspicious traces on the scene are important prerequisites for detecting cases and arresting criminal suspects.China’s criminal investigation monitoring technology is gradually developing in the direction of non-contact,digital and intelligent.Infrared imaging technology is one of the most mature monitoring technology,which has surpassed the human visual barrier and can help people ‘see’ the surface thermal radiation information of objects.Through the residual information on the scene,it will not only help to analyze the situation of the case,but also provide strong evidence support.However,in the practical process,due to highly uncertain scene situation,the difficulty of extracting the effective target information is increased,and the reliability and accuracy of the result are affected.Therefore,according to the infrared criminal investigation images in different situations,around the characteristics of infrared trace images collected in different time and environment,taking biological immunology as the theoretical source and basis,this paper designs targeted target extraction algorithms to provide help for the detection of criminal cases.At present,we only take the criminal investigation handprint image as an example in the simulation experiment.The main contents of the study include:(1)In order to solve the problem of unclear target contour and fuzzy edge band in infrared criminal investigation image,a contour model target extraction algorithm based on apoptosis gene regulation mechanism was proposed.Based on the principle behind the activation and inhibition of complex proteins involved in apoptosis regulation,combined with the initial contour curve,image region information and edge gradient information,the contour model algorithm is guided and improved.(2)In order to solve the problem of low contrast between target and background area and unclear feature information of target in infrared criminal investigation image,a restoration algorithm of infrared fingerprint image based on apoptosis mechanism is proposed.The algorithm uses time series images to construct the restoration model;takes five fingers and the center point of the palm area as the initial feature points.And then collects the gray information of the four neighborhood of the feature points to construct the relationship model between the target feature point and the four neighborhood.Finally,the initial can be restored.(3)Aiming at the problem that the background of infrared criminal investigation image is complex and there are a large number of target like interference,an infrared target trace extraction algorithm based on immune extreme region is proposed.The algorithm uses Gaussian filtering and two-dimensional gamma convolution to correct the lightness component of the image,and uses the maximum entropy method for the initial segmentation of the image.According to the maturation process of T cells in biological immunity,it abstracts the extraction process of immune extremum region,obtains the feature region,and extracts the target region.In conclusion,referring to the mechanism of apoptosis and lymphocytic T cell maturation in biological immunology,this paper designs infrared criminal detection image target extraction algorithms based on biological factor mechanism.Through comparative experiments,the results show that our algorithms are effective and accurate.