Large-scale image colorization based on divide-and-conquer support vector machines

doi:10.1007/s11227-015-1414-z

	Large-scale image colorization based on divide-and-conquer support vector machines
	Chen, Bo-Wei 1; He, Xinyu 1; Ji, Wen 2; Rho, Seungmin 3; Kung, Sun-Yuan 1
	2016-08-01
发表期刊	JOURNAL OF SUPERCOMPUTING
ISSN	0920-8542
卷号	72 期号:8 页码:2942-2961
摘要	This study presents a system that can automatically colorize grayscale images in large quantities. To enable big data training, divide-and-conquer support vector machines (SVMs) also proposed as classifiers are frequently used in this study. The system is composed of two components-image classification and local-descriptor classification. The former firstly analyzes an input by using a classifier, so that the system can determine which class should serve as the knowledge base. After the class is decided, the latter stage subsequently uses this knowledge base as the reference to colorize the input. Experimental results showed that the accuracy of classification in image classification could reach 90.50 %. Moreover, in the local-descriptor classification, the majority of pixels were successfully assigned correct colors. During the efficiency test, the proposed divide-and-conquer SVM enhanced computational speed while maintaining the accuracy. Such findings demonstrate the effectiveness of the proposed method and the feasibility of our idea.
关键词	Divide-and-conquer support vector machine Image colorization Big data
DOI	10.1007/s11227-015-1414-z
收录类别	SCI
语种	英语
资助项目	Ministry of Science and Technology, the Republic of China[103-2917-I-564-058] ; Beijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology, Chinese Academy of Sciences[2015-4] ; Department of Electrical Engineering, Princeton University
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Hardware & Architecture ; Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS记录号	WOS:000381986200003
出版者	SPRINGER
引用统计	被引频次：4[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/8110
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Chen, Bo-Wei
作者单位	1.Princeton Univ, Dept Elect Engn, Princeton, NJ 08544 USA 2.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 3.Sungkyul Univ, Dept Multimedia, Anyang, South Korea
推荐引用方式 GB/T 7714	Chen, Bo-Wei,He, Xinyu,Ji, Wen,et al. Large-scale image colorization based on divide-and-conquer support vector machines[J]. JOURNAL OF SUPERCOMPUTING,2016,72(8):2942-2961.
APA	Chen, Bo-Wei,He, Xinyu,Ji, Wen,Rho, Seungmin,&Kung, Sun-Yuan.(2016).Large-scale image colorization based on divide-and-conquer support vector machines.JOURNAL OF SUPERCOMPUTING,72(8),2942-2961.
MLA	Chen, Bo-Wei,et al."Large-scale image colorization based on divide-and-conquer support vector machines".JOURNAL OF SUPERCOMPUTING 72.8(2016):2942-2961.