文献类型：期刊文献

中文题名：基于Gabor小波变换的舌苔腐腻识别

英文题名：RECOGNITION OF GREASY OR CURDY TONGUE COATING BASED ON GABOR WAVELET TRANSFORMATION

作者：瞿婷婷[1];夏春明[1];王忆勤[2];朱穆朗玛[2]

机构：[1]华东理工大学机械与动力工程学院,上海200237;[2]上海中医药大学基础医学院,上海201203

年份：2016

卷号：33

期号：10

起止页码：162

中文期刊名：计算机应用与软件

外文期刊名：Computer Applications and Software

收录：CSTPCD;;CSCD:【CSCD_E2015_2016】；

基金：国家自然科学基金项目(81173199);上海市科委技术标准专项项目(14DZ0500400)

语种：中文

中文关键词：舌苔腐腻;Gabor小波变换;纹理分析;弱化边缘

外文关键词：Greasy or curdy tongue coating;;Gabor wavelet transform;;Texture analysis;;Edge weakening

摘要：舌诊是中医"望诊"中的重要内容之一,通过对舌苔的观察能了解人体生理变化,从而进行病势判断。当前基于计算机的舌苔腐腻识别研究鲜有报道。提出一种基于Gabor小波变换的舌苔腐腻识别方法。考虑舌象的完整性以及腐腻舌苔的纹理不同,从纹理角度出发,将整体的舌象进行Gabor小波变换;并针对Gabor小波变换的边缘检测效应使纹理描述弱化这一问题,提出一种弱化边缘的方法,在此基础上提取均值和标准差作为纹理特征对正常苔、腐苔和腻苔进行分类识别。实验结果表明,用改进的特征分类平均准确率达到91%,验证了该方法对舌苔腐腻的识别是有效的。
Tongue diagnosis is one of the most important parts in'diagnosing by inspection'of TCM,through observing tongue coating it is able to understand the changes in human physiology and thereby to analyse the illness. Currently the reports in regard to the computer-based greasy and curdy tongue coating recognition are few. In this paper we propose a Gabor wavelet transformation-based greasy or curdy tongue coating recognition method. Considering the integrity of tongue signs and the different textures in greasy or curdy tongue coating,proceeding from the perspective of texture we apply Gabor wavelet transformation on whole tongue signs,and propose an edge weakening approach aimed at the problem that the edge detection effect of Gabor wavelet transformation weakens the texture description. On this basis we extracted the mean and standard deviation as the texture features to carry out classified recognition on normal,greasy and curdy tongue coatings.Experimental results showed that the average accuracy of the improved feature classification reached 91%,this verified the method to be effective in recognising greasy or curdy tongue coating.