An Image Compensation Method Based On γ-Correction To Illumination Component

This thesis makes a discussion on illumination compensation. The purpose of this study is to eliminate inappropriate lighting influence, enhance structural information of object. The methods based on gray-level mapping, for instance, histogram Equalization, logarithm transform,γ-Correction, are defective since the resultant images are unnatural in grey level transition and short of three-dimensional sense. This thesis deals with illumination and reflection components respectively rather than the gray level directly. The illumination component is separated from the image by image smoothing based on assumption that the illumination component is smooth in the space; then the reflection component, expressed by base point, enhances linearly; finally aγ-correction is applied only to the illumination component. Experimental results show that the proposed method performs better in image compensation.