Research On Enhancement Algorithm For Non-uniform Illumination Image

With the continuous development of multimedia information, the digital image processing technology is widely used in many fields of family life and social work.In the process of acquiring images, the illumination of the local area is reduced, the color degradation and the contrast is weakened, and the image analysis and recognition is affected by the non-uniform illumination or dark environment.Therefore, it is necessary to enhance the image quality, so as to make it full use in the field of machine vision.This paper studies the relevant methods of image enhancement in depth, and some common algorithms are proposed and improved.The main works and results of this paper are shown as follows:1) The limitation of the traditional Retinex enhancement algorithm is analyzed, and the corresponding improving method is put forward.Because the illumination component of the non-uniform illumination image in the actual scene is not the gentle change,the Retinex algorithm based on the center surround does not consider the special nature of the edges, so easily lead to the gradient region of the halo effect.The improved method using recursive filters for illumination estimation, eliminated the influence of uneven illumination, also effectively inhibited the halo artifact in the enhancement results.In addition, in order to avoid color distortion phenomenon, the choice of color space to meet the human visual characteristics to deal with so as to enhance the color fidelity.2)A novel bio-inspired algorithm for color image enhancement is introduced and improved in detail.First of all, we have pointed out that the original method has the edge effect and lack of self adaptability, and proposed a new method to calculate the average intensity of the neighborhood and the brightness adjustment method, which can overcome the defects of the original method.Finally to describe the concrete process of the algorithm, is given more groups of the experimental results of contrast before and after improvement.3)An image enhancement method based on physical model and its MFCimplementation are presented.The image enhancement and physical model are combined with the idea of image, and the unknown parameters in the model are estimated accurately, and the edge-preserving filter is used to obtain the transmission of the media, and the intensity of the atmosphere is determined by combining Gaussian filter and histogram.Finally, the control parameters are set up by the characteristics of the image, and the adaptation ability of the algorithm is improved.In order to make the method proposed in this paper can be more convenient and intuitive to users, the use of MFC to build a platform with enhanced processing capabilities of human-computer interaction.In this paper, we give the realization and improvement steps of the three kinds of image enhancement theory and technology.Through experiment simulation platform show the method in this paper, the comparison results with existing methods,show that the proposed in this paper and the improved methods of enhancement effect compared with the existing methods have certain advantages.