基于结构风险最小化和全最小二乘法的色彩校正

丁二锐; 王义峰; 曾平; 丁阳

doi:10.3724/SP.J.1146.2006.01200

基于结构风险最小化和全最小二乘法的色彩校正

doi: 10.3724/SP.J.1146.2006.01200

丁二锐^{①;王义峰},
王义峰,
曾平,
丁阳

基金项目:

国家部级基金 (413160501)和西安电子科技大学研究生创新基金(05008)资助课题

计量
- 文章访问数: 3211
- HTML全文浏览量: 82
- PDF下载量: 1195
- 被引次数: 0
出版历程
- 收稿日期: 2006-08-14
- 修回日期: 2007-04-09
- 刊出日期: 2008-03-19

Color Calibration Based on Structural Risk Minimization and Total Least Squares

摘要

摘要: 该文提出了一种局部回归色彩校正算法。该算法从结构风险最小化原理出发,提出以全最小二乘法残差为经验风险,选取校正颜色点的K近邻域,采用局部回归进行色彩校正。实验结果表明该文算法在精度和鲁棒性方面均优于多重回归校正以及分区回归校正,其平均误差、最大误差,误差标准差分别下降了46%(27%),57%(21%),42%(20%)。
- 色彩校正; 结构风险最小化; 全最小二乘法; 局部回归
Abstract: A locally regressive algorithm for color calibration is proposed. Starting from the principle of structural risk minimization, the algorithm regards the residual of total least squares as the empirical risk and chooses the K-nearest neighborhood of a calibration color point to implement local regression for color calibration. Experimental results indicate that the proposed algorithm is superior, in both precision and robustness, to multiple regression and multiple regression based on subspaces and that its average error, maximum error and error standard deviation decrease by 46%(27%), 57%(21%) and 42%(20%) respectively.

HTML全文

参考文献(1)

Kang H R. Printer-related color processing techniques.Proceedings of SPIE, San Jose, USA, 1995, Vol.2413:410-419.[2]黄庆梅, 赵达尊. 一种彩色打印机标定的预测查找方法. 北京理工大学学报, 2003, 23(5): 601-603.Huang Qing-mei and Zhao Da-zun. A predictionapproachingmethod for the calibration of color printers.Journal of Beijing Institute of Technology, 2003, 23(5):601-603.[3]廖宁放, 杨卫平, 曾华等. 用多隐层BP网实现的CRT色度变换.中国图象图形学报, 2000, 5(6): 470-472.Liao Ning-fang, Yang Wei-ping, and Zeng Hua, et al.. CRTcolor conversion by a multi-layer BP neural network. Journalof Image and Graphics, 2000, 5(6): 470-472.[4]修晓杰, 王小华, 朱文华. 打印机颜色管理模型方法研究. 宁波大学学报(理工版), 2004, 17(2): 207-210.Xiu Xiao-jie, Wang Xiao-hua, and Zhu Wen-hua. Method ofbuilding printer color management model. Journal of NingboUniversity (Natural Science Engineering Edition), 2004,17(2): 207-210.[5]朱元泓, 孔令罔, K Muenger, 张旭亮. 扫描输入数据的颜色校正. 武汉大学学报(信息科学版), 2004, 29(9): 836-839.Zhu Yuan-hong, Kong Ling-wang, Kurt Muenger, and ZhangXu-liang. Color correction of scanning input data. Geomaticsand Information of Wuhan University, 2004, 29(9): 836-839.[6]Zhu Yuanhong, Kong Lingwang, Kurt Muenger, and ZhangXuliang. Color correction of scanning input data based onLMS and RGB subspaces. Chinese Journal of ScientificInstrument, 2004, 25(4): 284-287.[7]曾平, 谯婷婷, 顾伟松. 3-CCD 扫描输入的分区校正. 西安电子科技大学学报, 2006, 33(1): 19-23.Zeng Ping, Qiao Ting-ting, and Gu Wei-song. Region-divisioncalibration for 3-CCD scanner input. Journal of XidianUniversity, 2006, 33(1): 19-23.[8]Mou Tong-sheng and Shen Hui-liang. Colorimetriccharacterization of imaging device by total color differenceminimization[J].Journal of Zhejiang University SCIENCE A.2006, 7(6):1041-1045[9]Mendes L and Carvalho P. Adaptive polynomial regressionfor colorimetric scanner calibration using genetic algorithms.Proceedings of the IEEE International Workshop onIntelligent Signal Processing, Faro, Portugal, 2005: 22-27.[10]付凌晖, 王惠文. 多项式回归的建模方法比较研究. 数理统计与管理, 2004, 23(1): 48-52.Fu Ling-hui and Wang Hui-wen. A comparative research ofpolynomial regression modeling methods. Application ofStatistics and Management, 2004, 23(1): 48-52.[11]Vapnik N Vladimir. Statistical learning theory. New York:Wiley, 1998: 219-268.[12]Xia Minghua, Saber E, Sharma G, and Tekalp A Murat.End-to-end color printer calibration by total least squaresregression[J].IEEE Transactions on Image Processing.1999,8(5):700-716[13]孙金玮, 刘昕, 孙圣和. 基于总体最小二乘法的多功能信号重构方法研究. 电子学报, 2004, 32(3): 391-394.Sun Jin-wei, Liu Xin, and Sun Sheng-he. TLS algorithmbasedstudy on multi-functional sensor data reconstruction.Acta Electronica Sinica, 2004, 32(3): 391-394.[14]International Color Consortium. ISO 15076-1:2005 Imagetechnology colour management-architecture, profile format,and data structure- Part 1: Based on ICC. 1: 2004-04.[15]Gupta R Maya and Gray M Robert. Color conversion usingmaximum entropy estimation. IEEE International Conferenceon Image Processing, Thessaloniki, 2001, Vol.1: 118-121.

施引文献

资源附件(0)

访问统计