Loading...

Application of Semi-Supervised Learning in Image Processing

Mianjy, Poorya | 2010

593 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 41829 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Rabiee, Hamidreza
  7. Abstract:
  8. In recent years, the emergence of semi-supervised learning methods has broadened the scope of machine learning, especially for pattern classification. Besides obviating the need for experts to label the data, efficient use of unlabeled data causes a significant improvement in supervised learning methods in many applications. With the advent of statistical learning theory in the late 80's, and the emergence of the concept of regularization, kernel learning has always been in deep concentration. In recent years, semi-supervised kernel learning, which is a combination of the two above-mentioned viewpoints, has been considered greatly.
    Large number of dimensions of the input data along with the considerable costs associated with gathering labeled data, makes the applications of image processing generally ill posed. Therefore, semi-supervised learning methods in this area of research, has been a center of attention in recent years.
    A major challenge in semi-supervised kernel learning methods is its single resolution kernel matrix representation, using spectral decomposition. Like the Fourier transform that eliminates time information of data, representation of a classifier by spectral decomposition of kernel matrix eliminates spatial information of the classifier. That means, we cannot judge about smoothness of the classifier with spatial separation on neighborhood graph. The proposed method offers a solution to resolve this issue. We have evaluated our proposed method on Gender Classification, face pose detection, image segmentation and diagnosis of breast cancer. The results showed superiority of the proposed method compared to the other existing methods in this area
  9. Keywords:
  10. Semi-Supervised Learning ; Kernel Learning ; Patterns Classification ; Statistical Learning Method

 Digital Object List

 Bookmark

No TOC