IEICE TRANSACTIONS on Information
Salient Region Detection with Multi-Feature Fusion and Edge Constraint
Summary :
In this paper, we propose a salient region detection method with multi-feature fusion and edge constraint. First, an image feature extraction and fusion network based on dense connection structure and multi-channel convolution channel is designed. Then, a multi-scale atrous convolution block is applied to enlarge reception field. Finally, to increase accuracy, a combined loss function including classified loss and edge loss is built for multi-task training. Experimental results verify the effectiveness of the proposed method.
- Publication
- IEICE TRANSACTIONS on Information Vol.E103-D No.4 pp.910-913
- Publication Date
- 2020/04/01
- Publicized
- 2020/01/17
- Online ISSN
- 1745-1361
- DOI
- 10.1587/transinf.2019EDL8181
- Type of Manuscript
- LETTER
- Category
- Pattern Recognition
Authors
Cheng XU
Nanjing University of Aeronautics and Astronautics
Wei HAN
Nanjing University of Aeronautics and Astronautics
Dongzhen WANG
Nanjing University of Aeronautics and Astronautics
Daqing HUANG
Nanjing University of Aeronautics and Astronautics
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Cheng XU, Wei HAN, Dongzhen WANG, Daqing HUANG, "Salient Region Detection with Multi-Feature Fusion and Edge Constraint" in IEICE TRANSACTIONS on Information,
vol. E103-D, no. 4, pp. 910-913, April 2020, doi: 10.1587/transinf.2019EDL8181.
Abstract: In this paper, we propose a salient region detection method with multi-feature fusion and edge constraint. First, an image feature extraction and fusion network based on dense connection structure and multi-channel convolution channel is designed. Then, a multi-scale atrous convolution block is applied to enlarge reception field. Finally, to increase accuracy, a combined loss function including classified loss and edge loss is built for multi-task training. Experimental results verify the effectiveness of the proposed method.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2019EDL8181/_p
Copy
@ARTICLE{e103-d_4_910,
author={Cheng XU, Wei HAN, Dongzhen WANG, Daqing HUANG, },
journal={IEICE TRANSACTIONS on Information},
title={Salient Region Detection with Multi-Feature Fusion and Edge Constraint},
year={2020},
volume={E103-D},
number={4},
pages={910-913},
abstract={In this paper, we propose a salient region detection method with multi-feature fusion and edge constraint. First, an image feature extraction and fusion network based on dense connection structure and multi-channel convolution channel is designed. Then, a multi-scale atrous convolution block is applied to enlarge reception field. Finally, to increase accuracy, a combined loss function including classified loss and edge loss is built for multi-task training. Experimental results verify the effectiveness of the proposed method.},
keywords={},
doi={10.1587/transinf.2019EDL8181},
ISSN={1745-1361},
month={April},}
Copy
TY - JOUR
TI - Salient Region Detection with Multi-Feature Fusion and Edge Constraint
T2 - IEICE TRANSACTIONS on Information
SP - 910
EP - 913
AU - Cheng XU
AU - Wei HAN
AU - Dongzhen WANG
AU - Daqing HUANG
PY - 2020
DO - 10.1587/transinf.2019EDL8181
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E103-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 2020
AB - In this paper, we propose a salient region detection method with multi-feature fusion and edge constraint. First, an image feature extraction and fusion network based on dense connection structure and multi-channel convolution channel is designed. Then, a multi-scale atrous convolution block is applied to enlarge reception field. Finally, to increase accuracy, a combined loss function including classified loss and edge loss is built for multi-task training. Experimental results verify the effectiveness of the proposed method.
ER -
English
