The degraded images can be improved by image quality enhancement techniques considering color, contrast, and various other parameters related to images. Based on color constancy, the previous image enhancement algorithms, such as white patch assumption (WPA) and gray world assumption (GWA) algorithms, have several limitations. The color correction of resulting image has an antagonistic effect if the local region of an input image is biased by an individual color. Also, the correction result is degraded when the image does not have any white patches. Furthermore, the resulting image has low saturation, which degrades the correction result if images have monotonic color. To improve on these limitations, this paper proposed a color image enhancement algorithm based on the weighted multi-scale compensation coefficients using GWA algorithm. The multi-scale Gaussian filter is used for computing average values of local and global degraded color and calculating correction coefficients for size, pixel, and channel of multi-scale filtered images independently based on the brightness of the image. Then, the weights are determined for weight-sum of multi-scale correction coefficients by analyzing local color distribution of the image. Finally, the degraded color is improved by utilizing correction coefficients, which are integrated in original image, and degraded color saturation is improved using the proposed weights. The experimental results have shown that, compared with previous algorithms, the proposed algorithm improved color and contrast of various degraded image and produced better correction results.