Image Enhancement Algorithm Based On Brightness Changes

In life,for the limitation of light source condition,the imaging equipment may produce images that are too bright or too dark for people's eyes.Though these images are not suitable for people's eyes,the visual effect of these images can be improved by computational methods.This thesis has investigated the optical propagation process and a number of imaging algorithms.By integrating the Retinex theory and the response function of a camera,we propose a new algorithm for the enhancement of the luminance of images.When building the response function,according to the characteristics of the histogram,this algorithm selects the most representative feature points,connects a series of images in different manners and produces a set of three-dimensional luminance response functions.Through a set of strict evaluation including the subjective analysis and the objective quality index,it is verified that the proposed algorithm has a good performance in adjusting the luminance of the images.The proposed algorithm has a better performance than the traditional RGB luminance adjustment method in a number of indexes,i.e.,SSMI,mean gradient,root mean square error and error index of brightness relative order.When compared with the HSL luminance adjustment method,this algorithm also achieves better performance in the abovementioned?indexes.When employing the subjective analysis,the images produced by this algorithm are more suitable for people's eyes than the images produced by other image enhancement methods.