Research On Infrared Image Enhancement Method Based On Image Layering

With the rapid development of infrared imaging technology,this technology has now been widely used in the field of national defense and military and civilian industry,etc.However,due to the hardware defects of the infrared imaging system itself and the disturbance from the natural environment,the quality of the infrared images obtained is poor and its visual effect is not good,so it is necessary to enhance the low-quality infrared images.The infrared image enhancement method based on the hierarchical processing combines the advantages of various processing methods and it can effectively improve the visual effect of the image,therefore,more and more researchers have begun to study the hierarchical processing enhancement methods.The edge-preserving filtering algorithms are used to decompose the original image into background component and detail component,and perform enhancement and weighted fusion processing on the two layers of components to obtain the final enhanced image.But the hierarchical processing enhancement method ignores the visual characteristics of the image,therefore,the visual characteristics and the hierarchical processing idea are combined to study infrared image enhancement methods.The main research work done in this paper is as follows:First,the characteristics and characteristics of infrared image and its histograms are introduced,then analyzing the background and noise of infrared image,thus the importance of studying infrared image enhancement methods is pointed out.After studying the traditional infrared image enhancement algorithms,the hierarchical processing enhancement method combining multiple processing methods is studied,and aiming at the flaws of this method,combining with the visual characteristics,an hierarchical processing enhancement method combining single-scale Retinex and guided image filter is proposed.First,the single-scale Retinex based on the structure extraction algorithm is used to decompose the original image into the incident image and reflected image.Then,the background and detail components of the reflected image are obtained through the two weighted guided filtering algorithms.The edge information of the image is effectively preserved and the generation of "halo" artifacts is suppressed by the improved algorithms.Finally,the corresponding enhancement and weighted fusion processing are performed on each image component to obtain a high-quality infrared image with high contrast and rich texture details.Based on the research of algorithm theories,the MATLAB simulation platform is used to carry out simulation experiments on the hierarchical processing enhancement methods proposed in this study.And compared with the selected three commonly used infrared image enhancement algorithms: contrast limited adaptive histogram equalization,bilateral filtering and guided image filtering,objectively and subjectively.The experimental results show that the performance of the hierarchical processing enhancement methods proposed in this study are better compared with the other three algorithms,the enhanced infrared images have higher contrast,clearer details and textures,and theirs overall visual effects are better.