Enhancement Methods Research On The Degraded Color Images In FOG

Fog is a common natural phenomenon, it make the images acquired with the contrast of low、the fuzzy scene etc. by the visual system of outdoor due to scattered particles in the atmosphere so it can produce tremendous impact in the aerospace field, such as outdoor surveillance、target identification、target tracking、navigation、land transport、sea transport and air transport and so on. Therefore, it is very important practical significance for us to study enhanced methods research on the degraded color images in fog.The paper firstly analyze objectively the physical causes of fog imaging, it start studying enhancement algorithm research on the degraded color images in fog in-depth from the perspective of image processing. Then according to the depth of the fog color image of the scene changes in the characteristics, It make the detail of the fog degradation color image enhancement by the classical image enhancement methods including linear transformation, nonlinear transformation, histogram equalization and adaptive histogram equalization and so on.Utilizing hue invariance principle and conversion between color models, the grayscale image enhancement methods are extend to the degraded color image in fog and get good effect.Owing to the fog color image with low contrast and color characteristics of a light white, the single-scale Retinex enhancement algorithm (SSR) and multi-scale Retinex enhancement algorithm (MSR) are introduced according to the Retinex theory based on color constancy, and through improvement based on them, a recovery factor with color multi-scale Retinex enhancement algorithm (MSRCR) is proposed and it is proved to better restore the original features of the scene through several experiments.As fog images generally contain large noise, the Wavelet enhancement algorithm and the algorithm combined with the theory of wavelet transform and the Retinex theory of color image enhancement are proposed based on the characteristics of wavelet filter and Retinex theory, through several processing of the degraded color images in fog, the latter has relatively good enhancement effect.Through research on several degraded images in fog, two original color images from them and processed color image of them are evaluated finally by entropy、average gradient and variance, it is shown that the design of the wavelet transform and the Retinex theory of color image enhancement algorithm can achieve relatively better treatment effect.