摘要
裂纹舌是2型糖尿病较典型舌象特征之一。为准确提取舌象图像裂纹特征,提出基于局部灰度阈值的裂纹舌检测方法。首先,对图像进行拉普拉斯增强操作以提高舌中区域对比度,并结合中值滤波法去除增强后的噪声;接着根据区域一致性原理判断图像中是否存在裂纹;最后利用局部灰度平方差分离裂纹区域与背景区域。实验结果表明该方法可以有效地分离舌裂纹与背景,算法简单,实用性强,为裂纹舌客观化研究提供了较为可靠的技术支持。
Fissured tongue is one of the more typical features in qi and yin deficiency and blood stasis syndrome of Type 2 Diabetes Tongue. To detection and analysis fissure features of tongue image more accurately, this paper proposes a fissure detection method of tongue image which is Based on local gray threshold. First, paper use Laplacian image enhancement operations to improve the contrast of the middle region of tongue, and use median filtering method to remove random noise caused by Laplacian.Second, paper use regional consistency theory to test image have or not have fissure features. Finally, paper computes the local gray square difference to separate the fissure region and the background region. The experimental results show that this method can effectively detect the tongue fissure, and the algorithm is simple and practical, which provides reliable technical support for fissure tongue objective research.
出处
《电脑知识与技术（过刊）》
2017年第10X期163-165,共3页
Computer Knowledge and Technology
基金
2016年度安徽中医药大学青年科学研究基金项目(2016qn004)
关键词
裂纹舌
局部灰度阈值
裂纹检测
fissured tongue
local gray threshold
fissure detection
