Improved Image Coloring Algorithm based on LapRLS
Coloring is the process of adding colors to a monochrome image or video with the help of a computer.Study on the process of using traditional image segmentation and tracking involves image sequence,cost of human resources,in view of this,in recent years,many scholars have made a lot of researches on this aspect.A simple algorithm is proposed to solve this problem by training the model through the method of supervised learning to reduce the prediction error.In addition,kernel principal component analysis algorithm is used to reduce the computational complexity.The advantage of this method lies in two points.One: collect all pigment points so as to avoid the complexity of choosing the point, two: clever combination of other algorithms to simplify computation complexity.Finally,the experimental results of our proposed algorithm is compared with other algorithms.Experiments show that the coloring effect and calculation complexity of this algorithm is better than algorithms before.
Color monochrome image supervised learning kernel based principal component complexity
Xuedong Liu Jingjing Jia Jie Yang
Key Laboratory of Fiber Optic Sensing Technology and Information Processing(Wuhan University of Technology),Ministry of Education,School of Information Engineering,Wuhan,Hubei Province,China
国际会议
重庆
英文
2252-2255
2017-03-25（万方平台首次上网日期，不代表论文的发表时间）