Image Dynamic Range Broadening Method Research

Image is an important means of human perception world access to information, it has the advantages of image intuitive, large amount of information. However, when narrow partial slant dark or bright regions exist in the scene, the image dynamic range we obtained is relatively narrow, and the details of the scene can’t display properly. so it is significant to apply the wide dynamic technology to expand the dynamic range of image. In this paper, we study the method of image dynamic range expansion.To expand the dynamic range of single image, this paper introduces the Micro domain decomposition method for logarithmic enhancement model firstly.Based on the logarithmic enhancement model, we select the diagram 3?3 pixel areas to calculate average gray dispatching, after three times decomposition of refactorings, we get the wide dynamic range images, making the dark area contrast enhancement and distinct.In order to solves the bright region handling distortion problems caused by the micro domain decomposition method, this paper adopts the histogram equalization algorithm algorithm, which is based on the global histogram equalization. By selecting the appropriate threshold, the gray area which needs to be processed is processed by the histogram equalization algorithm. Although the algorithm is simple, the details of dark areas need to enhence while retaining the details of the brighter area.According to the fusion algorithm of multi exposure images, which is based on the YUV color space and the RGB space of the Laplacian pyramid decomposition, multiple images are decomposed by Laplasse Pyramid, and then the fusion strategy is used to reconstruct the image, which helps to get bright, dark area of the details of the clear, wide dynamic color image.The algorithm using fusion strategy is based on the contrast and color of two measure factor weighted graph. The difference between the two algorithms of YUV color space and the RGB space is that the YUV color space only decompose and reconstruct the Y channel of the pyramid, the U、V space only takes simple access to the most value, while in the RGB space, the R、G、B three channels are required to be decomposed and reconstructed in Pyramid.The algorithm is more simple and needs less time by using the YUV space.Simulation in the Matlab environment, the above three methods of image dynamic range broadening, the experimental simulation, using subjective and objective evaluation standard conducted in-depth comparative analysis on the performance of the algorithm. Three kinds of dynamic range extension algorithm can get wide dynamic range image, the image fusion algorithm based on Laplacian pyramid decomposition effect is best, but the algorithm is relatively complex; Micro domain decomposition method based on logarithmic image processing model is more suitable for processing of underexposed images; Area histogram equalization algorithm for underexposed and overexposed image can have good treatment effect, but it loses some of its gray level information. All three of these algorithms can effectively extend the dynamic range of image, improve the quality of image, the research conclusion for further application of the wide dynamic image technology, has practical guiding significance and reference value.