Name
Abstract | ||
|---|---|---|
Automatic image annotation based on traditional association rules exists the problem of \"sharp boundary\", which makes classification more fuzzy and inaccurate. Moreover, with the development of multimedia technology, the storage of image information is expanded quickly, massive image data will produce many redundant association rules, which will greatly decrease the accuracy and efficiency of image classification. In order to solve these two problems, this paper proposed an automatic image annotation model based on fuzzy association rules and decision tree. This approach firstly obtains association rules which represent the correlations between image features and high-level semantic concepts of training images. Then, the decision tree is added to reduce the unnecessary rules. As a result, algorithm complexity and time cost are reduced to a large degree. Experiments are based on two datasets Corel5k and IAPR-TC12. Experiments show the proposed method performs well in comparison with other automatic annotation method. |
Year | DOI | Venue |
|---|---|---|
2015 | 10.1145/2808492.2808504 | ICIMCS |
Field | DocType | Citations |
Decision tree,Data mining,Automatic image annotation,Annotation,Pattern recognition,Fuzzy classification,Feature (computer vision),Computer science,Fuzzy logic,Association rule learning,Artificial intelligence,Contextual image classification | Conference | 0 |
PageRank | References | Authors |
0.34 | 11 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Zhixin Li | 1 | 111 | 24.43 |
| Lingzhi Li | 2 | 0 | 0.34 |
| Kaobi Yan | 3 | 0 | 0.34 |
| Canlong Zhang | 4 | 6 | 0.80 |