Image Quality Assessment And Image Enhancement Technology

Images often show lower contrast and thinner color in bad condition. The lack of image information caused by the low visibility makes it difficult for the images used in other applications such as video surveillance, object detection, classification, and recognition. The paper studies on the image enhancement technique and the measure of performance of bad condition.Logarithmic transform domain coefficient histograms algorithms can remain more details while enhance the contrast of the image. It has a good ability of color rendering in color image enhancement when combined with HSI color space. The algorithms also performances stably in different bad condition. This paper analyses three new methods for which transform histograms utilized for contrast enhancement of images: logarithmic transform histogram equalization, logarithmic transform histogram shift, logarithmic transform histogram shape. The contrast entropy is used to measure the performance and of chooses the best parameters and transform these methods.Structure-oriented noise estimation uses a low-complexity automated homogeneity measure to distinguish noise and structures so that it can get a more exact estimation. Adaptive neighborhood used in denoise can effectively restrain additive noise without blurring the edge and detail information of images. A new denoise algorithm based on structure-oriented noise estimation and adaptive neighborhood is proposed in this paper which can provide a good denoising effect. Then the performance of enhancement combined with denoise is measured by the algorithm based on human visual system.