Efficient Segmentation Using Gamma Correction with Complement Image of Chinese Rubbing Image

For Chinese rubbing image under the complex background, to against its characteristic of low contrast and large noise, we develop a gamma correction enhancement algorithm in which the grayscale space is being conducted before the Otsu’s binarization, experiments on multiple pictures show the superiority of the algorithm. At first, the global contrast is enhanced based on the gamma correct algorithm for the complement image of the Chinese rubbing image. After that, we have implemented optimum global thresholding using Otsu’s method for image segmentation. The experimental results show that our algorithm could correct the background noise of the image and enhance the stroke in the low contrast Chinese rubbing image, and there is no need to denoise in advance. The performance of the algorithm is simple, fast, and produces very good segmentation.