A graying algorithm based on maximum preserving perception features of images whole structure

Although color images have gradually become the mainstream in image representation, gray images are still drawing more attention because of their small data amount and simple operation. Existing graying algorithms are difficult to ensure that the whole structure of images is not damaged and contrast difference information is consistent after dimensionality reduction. They cannot convey high quality gray images with human visual information. An optimization algorithm is proposed in this paper based on improving the objective energy function to retain these visual characteristics utmostly. The gray variance constraint is constructed to utmostly retain overall perceptual features of different gradients of the original color image regions, so as to ensure the stability of the global structure of the image. Let the difference between gray and color contrast be approximated to construct chromatic aberration of pixel pairs constraint term, and the contrast information is retained maximally. The final energy function is obtained by the linear combination of these two constraint terms. The iterative method is used to obtain the optimal channel weight coefficients which satisfy the minimum energy function and then get the result image. Finally, the experimental results show that the algorithm in this paper is basically consistent with the human visual perception of image contrast changes, and it can maintain the details and the global structure better.