基于灰度差分的舌象图像分类方法
Tongue Image Classification Based on Gray-scale Difference

DOI: 10.12677/CSA.2020.102020, PP. 190-199

Keywords: 舌诊，纹理分析，舌象分类，灰度差分法
Tongue Diagnosis, Texture Analysis, Tongue Classification, Gray Difference

Full-Text   Cite this paper   Add to My Lib

Abstract:

传统的舌诊常常由于医师的个人主观以及光照、亮度等外部因素的干扰而造成诊断结果的不稳定。为了探究舌象特征与舌诊结果的映射关系，用客观的观测指标对舌象进行定性、定量的研究，把中医临床医学经验和传统的图像处理技术相结合已经成为一种趋势。舌象的纹理特征可以反映出舌象薄厚、腐腻、老嫩的程度，而非舌象传统的颜色和形状等特征所能体现。因此，本文提出了一种基于灰度差分的分类方法。该方法首先通过色偏检测，对舌象的图像进行颜色校正；然后借助GrabCut算法对舌象进行图像分割，从而保留舌象中舌体的有效部分；最后，结合灰度差分统计法对舌体的纹理特征进行特征提取和分析，并对不同类型的舌象图像实现分类。实验结果表明，本文的方法相较于其他的纹理及颜色分类方法在分类舌象老嫩的情况上更具优势。
The traditional tongue diagnosis is often unstable because of the physician’s personal subjectivity and external factors such as illumination and brightness. In order to explore the mapping relationship between the characteristics of tongue images and the results of tongue diagnosis, it has become a trend to study tongue images qualitatively and quantitatively with objective observation indexes, and to combine the clinical experience of traditional Chinese medicine with traditional image processing technology. The texture of the tongue image can reflect the thickness and the degree of rot, but not the traditional color and shape of the tongue image. Therefore, a classification method based on gray difference is proposed. The method firstly corrects the tongue image by color deviation detection. Then, GrabCut algorithm is used to segment the tongue image so as to preserve the effective part of the tongue body in the tongue image. Finally, the texture features of tongue are extracted and analyzed with the method of gray difference statistics, and different types of tongue images are classified. The experimental results show that the proposed method is superior to the other methods of texture and color classification in classification accuracy.

References

[1]	王爱民, 赵忠旭, 沈兰荪. 中医舌象自动分析中舌色、苔色分类方法的研究[J]. 北京生物医学工程, 2000, 19(3): 136.
[2]	张季, 钱峻, 王瑞娟, 等. 中医数字化舌象客观特征分析研究现状分析[J]. 世界科学技术-中医药现代化, 2014(11): 2420-2424.
[3]	Chiu, C.C. (1996) The Development of a Computerized Tongue Diagnosis System. Biomedical Engineering: Application, Basis and Communications, 8, 342.
[4]	罗大中, 梁嵘, 王召平, 等. 舌侧的颜色特征研究[C]//中国中西医结合学会诊断专业委员会会. 中国中西医结合学会诊断专业委员会2009年会论文集. 2009: 136-140.
[5]	杨新宇, 梁嵘, 王召平, 等. 基于色度学的舌色分类研究现状与分析[J]. 北京中医药大学学报, 2012, 35(8): 539-542.
[6]	许家佗, 张志枫, 任宏福, 等. 一种舌体胖瘦分析的图像诊断方法[J]. 中国中西医结合影像学杂志, 2009, 7(6): 407-410.
[7]	钟少丹, 韦玉科, 谢铮桂. 齿痕舌像自动分割的方法[J]. 计算机技术与发展, 2009, 19(1): 245-247.
[8]	许家佗, 孙炀, 张志枫, 等. 基于差分统计方法的舌象纹理特征的分析与识别[J]. 上海中医药大学学报, 2003, 17(3): 55-58.
[9]	黄贤武, 王加俊, 李家华. 数字图象处理与压缩编码技术[M]. 成都: 电子科技大学出版社, 2000: 185.
[10]	阮秋琦. 数字图像处理学[M]. 北京: 电子工业出版社, 2007.
[11]	徐晓昭, 蔡轶珩, 刘长江, 等. 基于图像分析的偏色检测及颜色校正方法[J]. 测控技术, 2008, 27(5): 10-12+21.
[12]	Rother, C. (2004) GrabCut: Interactive Foreground Extraction Using Iterated Graph Cuts. Proceedings of SIGGRAPH’04, 309-314. https://doi.org/10.1145/1186562.1015720
[13]	Gasparini, F. and Schettini, R. (2003) Color Correction for Digital Photographs. 12th International Conference on Image Analysis and Processing, 2003. Proceedings, Mantova, Italy, 17-19 September 2003, 646-651.
[14]	Gasparini, F. and Schettini, R. (2004) Color Balancing of Digital Photos Using Simple Image Statistics. Pattern Recognition, 37, 1201-1217. https://doi.org/10.1016/j.patcog.2003.12.007
[15]	卫保国, 沈兰荪, 蔡轶珩, 等. 中医舌苔腐腻分析算法的研究[J]. 电子学报, 2003, 31(S1): 2083-2086.

Full-Text