基于 Welsh算法的灰度图像彩色化的研究

摘要

Welsh algorithm achieves good effect in grayscale image colourisation process, but there are the problems such as too many cycles in pixel matching process, too slow of the speed, and the image has noise after colourisation.Based on the analysis of the inadequacies of the Welsh algorithm, we adopt the approach of interval point picking in colour image, thus reduce the times of cycle.At the same time, aiming at the noise problem in the image to be colourised, we use guidance filtering to optimise the migrated image and remove the noise interference.It is demonstrated by the experiments that the image processed by guidance filtering can produce better results, the effect of the image is clearer and more natural.%Welsh算法在将灰度图像彩色化的过程中取得了较好的效果，但是却存在着像素匹配过程中循环过多、速度过慢、彩色化后的图像存在噪声的问题。分析Welsh算法的不足之处，采取在彩色图像中间隔取点的办法，减少循环次数。同时针对彩色化后的图像中存在噪声的问题，使用指导性滤波对迁移后的图像进行优化，去掉噪声的干扰。通过实验表明，经过指导性滤波处理的图像可以产生出更好的效果，使图像的效果更清晰和自然。