Study Of Infrared Image Contrast Enhancement Algorithm

With the rapid development of infrared technology, infrared thermal imagingtechnology has been widely used in more and more fields. However, because ofinfrared thermal imaging characteristics of imaging mechanism, it makes the infraredimages fuzzier and lower contrast than ordinary TV images and visible radiationimages. These characteristics are very harmful to the subsequent senior operations,such as target detection, target recognition, feature extraction, target tracking and so on.Therefore, for the image processing system of target detection, target recognition,feature extraction, target tracking, the infrared image enhancement is a necessarypretreatment technology. Thus the contrast enhancement of infrared image research isvery necessary.According to the widespread characteristics of infrared images, the paper focuseson how to improve the overall visual effect of infrared images and how to improve thecontrast of the infrared images, etc. This article mainly made the following severalaspects of the research work:Firstly, the paper discussed the working principle of thermal infrared imager,researched the imaging mechanism of infrared images and specifically analyzed noise-formed characteristics and histogram of the infrared images. Then, based on thegeneral features of infrared images, several more classic traditional imageenhancement algorithm are discussed and analyzed, and to test them respectively.Afterwards, the features of them are summarized through the experimental results.In order to better improve the contrast of infrared image, several more valuablegray transform algorithms for infrared images were discussed, and summarized theirrelative merits through a large number of experimental comparisons and analysis ofthem. On that basis, the paper put forward a new piecewise liner drawing algorithm forinfrared images contrast enhancement, which can get two adaptive threshold value based specific infrared image gray information. The result of the experiments showthat the target’s edge and outline of image processed by the algorithm is improved thanformer algorithm processed, and the overall visual effect is better than formers.Finally, focus on the problems, such as gray level in merger and detail lost, alongwith the two peaks or multimodal distribution present in many infrared images, thethesis proposed a new contrast enhancement algorithm of infrared images—doublethreshold sub-regional enhancement algorithm of infrared images based on. The resultof the experiment testify proposed algorithm not only realized the redistribution ofgray level of the infrared images, but also effectively widened the target detail of thegray dynamic display scope. Theories and simulation experiments proved that theproposed algorithms in this paper turned out to be concise and self-adaptable, and itcan be used for handling hardware real-time processing and bring about a betterenhancement to the infrared images.