Research On Image Feature Extraction Improved Algorithm For Images Of Spleen Deficiency Tongue Image Recognition

Tongue diagnosis is always being one important way of the traditional Chinese medicine inspection,till nowadays still has great medical reference value.A large number of studies have shown that the teeth-printed tongue as a kind of abnormal tongue with spleen deficiency symptom,the appearance of the body characteristics,often is very different with the normal tongue body,and it is these differences indicating a disease.The traditional tongue diagnosis results,most of them are dependent on subjective judgments of experts in Chinese medicine and the diagnosis results will be disturbed by the external environment,resulting in the diagnosis results are often not repeatable.In this paper,research on image feature extraction improved algorithm for images of spleen deficiency tongue image recognition is established to solve above problems.This paper is based on the teeth-printed tongue as the breakthrough point,through putting forward some improved algorithms and analysis methods to segment the right images of the teethprinted tongue and its related characteristics are analyzed,including the using of based on threshold,region and edge detection algorithm for the teeth-printed tongue images segmentation,using gray projection method to segment the teeth-printed tongue images,contrast enhancement,using the improved control marker watershed algorithm to segment the teeth-printed tongue images;combing the B sample wavelet transform and Canny operator to extract the teeth-printed tongue images edge information,using optimizing Snakes algorithm for image segmentation;tongue shape and histogram,color features are analyzed,the degree of severity of the teeth-printed tongue and spleen deficiency symptom is classified by using roundness parameters and color features.The experimental results show that the improved algorithm can achieve better segmentation for the teeth-printed tongue images,based on the tongue segmentation,combination of image features,and the tongue theories,putting forward a few common group of the teeth-printed tongue characteristic parameters,analyzing and quantitative extracting the circular degree parameter and color features represent the teeth-printed tongue,by using the characteristic parameters and color features,the severity of the teeth-printed tongue and spleen deficiency symptom can be effectively represented and diagnosed respectively.