Research Of Image Enhancement Algorithm Based On Human Vision Characteristics

In the processes of image-forming and digitalizing, due to noise disturbance,equipment limitation and other effect factors, inevitably it leads to inferior image quality, which can be reflected by lower contrast, fuzzy details and so on. The main objective of image enhancement is adopting variety of image processing technology to increase the contrast and definition,sharpen the interesting and important information, so that processed images are more suitable for higher-level processing and analysis of human and machine.The conventional image enhancement technology mainly starts from the engineering technology field. While enhancing images some shortcomings arise, such as the noise can not be effectively eliminated, the information of processed images have a larger alteration and so on. As the image information is obtained by human vision, if the characteristics of human vision system is taken into consideration, it will greatly improve the quality of the image enhancement. The thesis combines with human vision characteristics to improve current image enhancement technology, to research a comprehensive image enhancement technology. It aims to obtain a better enhancement effect and simultaneously eliminate the noise effectively without changing the original image a lot. The main work is as follow:(1)Illustrate the related human vision characteristics, analyze the performance of lateral inhibition in spatial and frequency domain, research the lateral inhibition by combining with adaptive filter, improve and propose a new grey image enhancement algorithm based on lateral inhibition and adaptable filter.(2)Compare and analyze the common color spaces, take advantage of the lateral inhibition to sharpen the contrast, combine with HSV color space transformation, improve and propose a color image enhancement algorithm based on lateral inhibition and color space transformation.(3)The thesis focuses on exploring how to effectively combine the human vision characteristics with the image enhancement technology. Two improving algorithms are proposed, the MATLAB simulation results show that the first algorithm can not only enhance the gray image but also eliminate the noise with a better vision effect; the second algorithm is not only able to effectively improve the color of the image but also sharpen the color image's edge and details.(4)Some innovation based on human vision characteristics to improve image enhancement algorithm is explored.