Abstract | ||
---|---|---|
To solve the curse of dimension caused by the high dimensional characteristics in image retrieval, an image retrieval method based on manifold learning was proposed. In the model, the low dimensional features were firstly extracted from images with the manifold learning algorithms. Then the query results were acquired according to the low dimensional features. To overcome defects of Euclidean distance in the high dimensional space, we designed the manifold learning method based the histogram intersection distance. Experiments showed that the presented method greatly improved the retrieval precision and recall. The manifold learning algorithm based on the histogram intersection distance was much superior to the method based on Euclidean distance. And, the execution efficiency was also greatly progressed with the low dimensional features to retrieve images. |
Year | Venue | Keywords |
---|---|---|
2015 | 2015 12TH INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS AND KNOWLEDGE DISCOVERY (FSKD) | Image retrieval, manifold learning, color histogram, histogram intersection distance |
Field | DocType | Citations |
Computer science,Image retrieval,Manifold alignment,Artificial intelligence,Image histogram,Nonlinear dimensionality reduction,Computer vision,Pattern recognition,Euclidean distance,Histogram matching,Content-based image retrieval,Machine learning,Visual Word | Conference | 0 |
PageRank | References | Authors |
0.34 | 12 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jin Shi | 1 | 0 | 0.34 |
Lukui Shi | 2 | 11 | 4.33 |
Xiaoteng Gong | 3 | 0 | 0.34 |
Shengli Shi | 4 | 0 | 0.34 |