Performance Comparison of Sift, Sift+sparse Coding Method and PCA_K-Means Clustering |
Author(s): |
| K Naga Divya Bhargavi , DMS SVH College of Engineering; Dr. Ch. Santhi Rani, DMS SVH College of Engineering |
Keywords: |
| CBIR, K-Means Clustering, Principal Component Analysis, SIFT, Sparse Coding |
Abstract |
|
Nowadays, the requirement of digital images has been grown tremendously. Respectively, different types of databases are maintained by different sources like military, industries and medical etc. In order to retrieve the required images from those databases, we have to use different classification and retrieving techniques. For Classification Principal Component Analysis (PCA) is better estimated method and for retrieval Content Based Image Retrieval (CBIR) is the popular method used. In which the texture features are extracted using Principal Component Analysis (PCA) with K-nearest neighbourhood applied. The results are better observed using precision and recall. The performance is better estimated with the percentage performance of precision and recall having nearly 96.9% and 67.07% respectively. |
Other Details |
|
Paper ID: IJSRDV4I110587 Published in: Volume : 4, Issue : 11 Publication Date: 01/02/2017 Page(s): 616-619 |
Article Preview |
|
|
|
|
