Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 47079 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s):
- Abstract:
- Classification is one the contrivesial problems in machine vision and pattern recongnition. Traditional feature extraction methods which are based on low level feature extraction do not have high classification accuracy, thus they do not have the ability to represent images in feature space in discriminative way. In this thesis we have proposed a grid base method and used hidden Markov model (HMM) to include topological and spatial information in feature vectors. Then the classifiers created based on HMM feature extraction are combind. Combination of classifiers is based on designing a convex goal function. The goal of this optimization is to determine the wight of each classifier for final decision making. The basic features are selected from five types of features such as SIFT, Gist, HSV, Gabor function and Centrist. These features are complementary for each other and can cover their incorrect decisions making. The proposed method improved the accuracy of classification. It showed that combining the features with different classifiers had better performance than fusing multiview features with dimensional reduction in the aspect of classification accuracy. In addition the experimental results show that the basic features used in feature extraction phase with HMM, follow the HMM ruls. The proposed method has been evaluated on two standard datasets and experimental results showed its superiority in comparison with most recent works
- Keywords:
- Scene Classification ; Hidden Markov Model ; Optimization ; Combining Classifiers ; Semantic Features
Digital Object List
-
محتواي کتاب
- view
Bookmark