Journal of Guangdong University of Technology ›› 2011, Vol. 28 ›› Issue (4): 48-50.

• Comprehensive Studies • Previous Articles     Next Articles

The Extraction of Texture Based on the Co-occurrence Matrix -of LAB Color-Difference

  

  1. Faculty of Automation, Guangdong University of Technology, Guangzhou 510090, China
  • Online:2011-12-25 Published:2011-12-25

Abstract: To reduce the influence of illumination,mean filter was used to weaken the influence of the noise on the image. Then,the method of studying the image was presented, based on the cooccurrence matrix of LAB colordifference——a new conception which was proposed on the basis of gray level cooccurrence matrix. Both color and texture features were extracted from the cooccurrence matrix of LAB colordifference. The experimental results show that the method fits closely with human perception.

Key words: mean filter; color; texture; machine vision; contactless measurement

[1] 郭德军.基于灰度共生矩阵的纹理图像分类研究[J].林业机械与木工设备,2005(7):21-23.

[2] 冯建辉,杨玉静.基于灰度共生矩阵提取纹理特征图像的研究[J].北京测绘,2007(3):19-22.

[3] 白雪冰,王克奇.基于灰度共生矩阵的木材纹理分类方法的研究,2005,37(12):1667-1670.

[4] 刘广海,王强,聂桂军.基于彩色子块共生矩阵的图像检索[J].南京工业职业技术学院学报,2009,9(2):47-50.

[5] 刘海艳,刘巍.基于彩色局部模式共生矩阵的图像检索[J].科学时代,2011(5):51-53.

[6] 林福宗.Color-measurement[M].北京:清华大学智能技术与系统国家重点实验室,2001.

[7] 陈丽雪,陈昭炯.基于Lab空间的图像检索算法[J].计算机工程,2008,34(13):224-226.

[8] 林开颜,吴军辉,徐立鸿.彩色图像分割方法综述[J].中国图象图形学报,2005,10(1):1-10.

[9] Howarth P, Ruger S. Evaluation of texture features for contentbased image retrieval[C]∥LNCS 3115. International Conference on Image and Video Retrieval(CIVR ’04). Dublin,    Ireland: Springer 2004:326-334
[10] Smith J R, Chang S F. Tools and techniques for color image retrieval[C]∥Storage and Retrieval for Image and Video Database IV, San Jose, CA:SPIE Proceedings,1996,2670:426-437.
[1] LI Lin, GUO Da-Chang, YIN Chao-Jie-. The Extraction of Texture Based on the Co-occurrence Matrix -of LAB Color-Difference [J]. Journal of Guangdong University of Technology, 2011, 28(4): 45-47.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!