Abstract | ||
---|---|---|
The key of salient object detection is to extract the most attractive area in the scene. This paper fully explores the hierarchical cognitive mechanism of visual information, combines color contrast and depth contrast, and puts forward a salient object detection algorithm for the RGB-D image. Three saliency maps were prepared, namely, initial saliency map, middle saliency map and advanced saliency map. The three maps were then fused into a final saliency map. The proposed method was compared with six popular salient object detection methods on three RGB-D image datasets. The comparison shows that our algorithm achieved the best results in accuracy, recall rate and F-value. The research findings shed important new light on salient object detection in RGB-D images. |
Year | DOI | Venue |
---|---|---|
2020 | 10.18280/ts.370104 | TRAITEMENT DU SIGNAL |
Keywords | DocType | Volume |
cognitive mechanism,salient object detection,RGB-D image,saliency map | Journal | 37 |
Issue | ISSN | Citations |
1 | 0765-0019 | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Xiao Tang | 1 | 0 | 0.34 |
Ting Zeng | 2 | 0 | 0.34 |
Benxiang Ding | 3 | 0 | 0.34 |
Yang Tan | 4 | 0 | 0.34 |