Research On Infrared Dim And Small Target Detection Algorithm Based On Human Visual Contrast Mechanism

In recent years,with the continuous progress of technology,infrared imaging has been widely used in face recognition,target detection and medical diagnosis and other fields.However,due to the strong noise such as cloud clutter in the complex background,the long imaging range of the target in the practical application scene and the serious attenuation of the signal in the atmosphere,the infrared target has the characteristics of weak signal,small imaging area,missing shape information,inconspicuous texture features and low signal-to-noise ratio,at the same time,the presence of artificial light sources and buildings in complex background also leads to high false alarm rate of target detection algorithm,which brings great challenges to the performance of infrared dim and small target detection algorithm.In practical application,the target detection algorithm is usually required to ensure high accuracy,low false alarm rate and good real-time performance,therefore,how to balance each evaluation index of the algorithm to achieve high performance is one of the difficulties in the research of target detection algorithm.In order to solve the problem of infrared dim and small target detection in complex background,an algorithm based on human visual system contrast mechanism is proposed.Firstly,the infrared image is preprocessed,and the isolated noise points in the infrared image are removed by median filtering.Then,the processed image is subjected to difference of Gaussians filtering to suppress large-area highlight areas in the image.Finally,the improved local contrast method algorithm is used to remove the highlighted edge region,eliminate the high suspect target,and the detection of infrared dim and small targets in complex background is realized.Experiments show that compared with the traditional LCM algorithm,Top-Hat algorithm,TDLMS algorithm and Infrared Patch-Image Model algorithm,the algorithm has more advantages in false alarm rate,correct detection rate,detection time,etc.It has the characteristics of high detection rate,low false alarm rate,good robustness and short running time.