Dissertation > Excellent graduate degree dissertation topics show

Research on Tongue Color Analysis Methods Based on Semi-Supervised Learning

Author: XiaoHongTao
Tutor: WangKuanQuan
School: Harbin Institute of Technology
Course: Applied Computer Technology
Keywords: Medical biometric identification Tongue Diagnosis Pixel classification Semi-supervised learning
CLC: TP391.41
Type: Master's thesis
Year: 2007
Downloads: 116
Quote: 0
Read: Download Dissertation

Abstract


Tongue diagnosis is one of the TCM diagnostic methods in medicine clinical value. In recent years, with the rapid development of computer science and technology, the development of traditional Chinese medicine tongue diagnosis learning toward computerized direction has become an inevitable trend. This article is trying to promote the development of computerized Tongue, focusing tongue image color training and classification problems. The main contribution of this paper is that: the analysis of the inadequacies of the existing Tongue color analysis method; design a pixel classification algorithm based on semi-supervised learning to solve the moss color pixel-based tongue color classification system based on the pixel structure; the establishment of a quality color distribution model; RKNN algorithm, a global optimization problem is transformed into a dynamic local issues, and apply the moss color quality color classification, to solve the the moss color quality color massive pixel classification time complexity problem; the color ratio eigenvectors and calculates the tongue image, its application in the overall color of tongue image classification and automatic diagnosis of pancreatitis. First, the proposed system has summed up the inadequacies of the existing Tongue tongue color analysis method, reasons and in accordance with these deficiencies, the correct choice of the Tongue image pixels as the object of study of the classification algorithm. Then a new algorithm based semi-supervised learning medical biometric identification, the algorithm's performance is better than the performance of supervised learning and unsupervised learning. Secondly, this paper to select and after training methods to select the the moss color quality color distribution model to establish a data set of 12 kinds of the moss color quality color distribution model to solve the tongue color classification system based on pixel color model problem, greatly improving the quality of the training sample. The paper further raised RKNN algorithm, application classification of the moss color quality color, a global optimization problem is transformed into a dynamic local algorithm to make it suitable for large-scale classification of pixels in the image of the Tongue calculated. Finally, according to the proportion of 12-dimensional color feature vector of the image of the tongue, the overall color of the tongue image classification. Finally, analysis of training samples and experimental results, and, under the guidance of the Chinese medicine experts to study the feasibility of automatic diagnosis of pancreatitis and its automatic diagnosis effect, and achieved satisfactory results.

Related Dissertations

  1. Tongue Feature Extraction and Research of Fusion Classification,TP391.41
  2. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  3. Two Kinds of Active Learning Methods,TP181
  4. Research on Expression Recognition Methods Based on Semi-Supervised Learning and Decision Level Fusion,TP391.41
  5. Research on Classification Method of Tongue Substance Color and Tongue Coating Color Based on SVM,TP391.41
  6. Semi Supervised Clustering Algorithm and Its Application and Research,TP301.6
  7. Analysis of Radial Arterial Ultrasonic Blood Flow Signal and Its Application in Medical Diagnosis,TP29-AI
  8. Traditional Chinese Medicine Based Texture Analysis of Thickness of Tongue Coating,TP391.41
  9. Fuzzy Clustering and the Application on Image Segmentation,TP391.41
  10. Semi-supervised learning based fault diagnosis,TP277
  11. Semi-supervised Manifold Learning and Its Application,TP18
  12. Clustering uncertain data,TP311.13
  13. Graph-based semi-supervised learning model and classifier design,TP181
  14. Feature-based semi- supervised learning space transformation,TP181
  15. Semi-supervised regression based on selective ensemble algorithm and its application,TP181
  16. The Application of the Tongue Diagnosis in Chronic HBV,R259
  17. A Semi-Supervised Learning Algorithm Based on Lie Group and Application,TP181
  18. Tongue Diagnosis of lung cancer diagnostic value of,R273
  19. Research about the Relationship between Diagnosis & Therapy of Spinal-related Pathogen and Tongue & Pulse Diagnosis,R274.9
  20. Spectrum Method in Tongue Inspection,R241

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile