The Study Of The Method Of Remote Sensing Image Enhancement

Image is one of the important means of transmission of information and media,which contains very important information. Since the1960s, the field of imageprocess has been in the vibrant development of increasingly. Its wide range ofapplications such as remote sensing, medicine and defense and industrial sectors havebrought great economic and social benefits.Image enhancement is a very important part of image processing, it is also veryimportant in other fields. The purpose of image enhancement algorithm is obtaininggood vision and detail information from the poor images. The application of imageenhancement is targeted, and a general enhancement algorithm dose not exists. Theevaluation of the enhanced results includes the person's subjective evaluation and theobjective evaluation based on the parameters.In this paper, the development of image enhancement and some new algorithmsin recent years in the world were introduced firstly. The basic theories and methods ofimage enhancement were introduced in the second chapter, which focus on thehistogram processing techniques, image smoothing and sharpening techniques andconclude with a brief analysis of some applications of fuzzy techniques in imageenhancement. Secondly, we introduced DRSHE and BHEPL in the third chapter. Thelogarithmic image processing model also has been introduced in the third chapter. Andtheir own characteristics, the process of the algorithm implementation and the issuesin the enhancement have carried out detailed analysis and discussion. Finally, to solvethe problem of the enhanced image visual effect is poorer and detail information notobvious, a new adaptive method of image enhancement based on relative entropy waspresented. In this method, we calculate every pixel's relative entropy between datemodel and uniform distribution model in the image firstly. And then we can set athreshold to judge the pixel's relative entropy whether more than the threshold. If it isthat, the pixel will be enhanced by the algorithm which base on the neighborhood information of pixel. And other pixels will be enhanced by the algorithm of contrastenhancement using Bi-Histogram equalization with a plateau limit. Experiments showthat the enhanced image by our algorithm has a good visual effect and its detailinformation is clearer.