基于区域局部特征的中医舌像检索

Tongue Image Retrieval Based on Local Feature

根据中医理论，舌象是五脏六腑的投影，舌体的不同区域可以反映出人体的相关疾病信息。本文以中医舌像为研究对象，针对全局底层图像特征很难描述舌像不同区域特征的缺点，提出了基于区域局部特征的中医舌像检索方法以提高中医舌像的识别率。由于SIFT局部描述子不仅具有良好的尺度和亮度不变性，同时对仿射形变，视角改变和噪声等也有一定的鲁棒性，因此系统选择了SIFT描述子来描述图像特征。同时针对SIFT特征在检索过程中匹配计算代价大的缺点，提出了基于奇异值分解的SIFT特征点集合匹配方法，实验结果表明该方法在检索效果和检索时间消耗方面取得了一个很好的平衡。

According to Traditional Chinese Medical Theory, the tongue information is the projection of body organ. This paper uses tongue images as study object, as the shortcomings which global feature cannot describe efficiently the information of the tongue images, this paper proposes a traditiomal Chinese medical tongue images retrieval method based on local feature to improve the recognition rate of tongue images. Because of its good stability and discrimination we choose SIFT descriptor to describe tongue images. SIFT descriptor not only has very well tolerant to scale changes, illumination variations, and image rotations, but also are robust to affine distortion, change of viewpoints and additive noise. Whereas, in order to overcome the shortcoming in terms of the large number of SIFT points and high computational costs for matching. Accordingly, we present a SVD- based method to compute the similarity of two SIFT feature points sets. Experimental results show that SVD-based SIFT feature points sets matching method gets a better tradeoff between the effect of detection and detection time costs.