| The digital image processing technique has become focuses of academics, such as computer science, information science, biology and medical science etc. In the military, industry etc. the realm acquired extensively applied. The technique of image enhancement is one of the important research contents of digital image processing technique, whose purpose is to person or machine analysis and comprehends easily. This thesis in detail analyzed the digital image processing technique. In allusion to images with frog, the thesis carried on more meticulous researches.Aim at images with frog to appear a hoariness effect, lower contrast and the thinner color characteristics, this thesis on the foundation analysis of the color constancy theories, analyses the Retinex and local Retinex theory, gave two methods to improve Retinex and Multi_Scale Retinex arithmetic in order to get a better result.Aim at soupy images' bigger noisy, this thesis in detail analyzed the method of eliminating noisy, based wavelet theory. The application of wavelet transform in image de_noise is studied by thresholding operation in wavelet domain of image. And this thesis presents a fast and simple method for estimating the variance of noise in soupy images. And the method that can be used to estimate the threshold value was given. It can reduce noise more effectively than classic method.Contraposing colorful soupy images, this thesis studied the image enhancement method in different color spaces. Designed the method that carried out to enhance the intensity, pull stretch the saturation and smooth the hue in the HSI color space; Carried out to enhance lightness, keep the chromatic aberration invariable in the YUV color space.To evaluate the quality of image enhancement, this thesis studied the evaluating method of image's contrast and entropy. The experimentation indicates that the improved MSR method designed by this thesis, obtained better effect in contrast and entropy. |
PDF Full Text Request