Salient object detection method based on multi-scale feature-fusion guided by edge information

被引：0

作者：

Wang X. ^{[1
,2
]}

Li M. ^{[1
,2
]}

Wang L. ^{[1
,2
]}

Liu F. ^{[1
,2
]}

Wang W. ^{[1
,2
]}

机构：

[1] State Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, Tianjin

[2] MOEMS Education Ministry Key Laboratory, Tianjin University, Tianjin

来源：

Hongwai yu Jiguang Gongcheng/Infrared and Laser Engineering | 2023年 / 52卷 / 01期

关键词：

boundary loss function; edge-information guidance; multi-scale feature-fusion; salient object detection; spatial attention module;

D O I：

10.3788/IRLA20220344

中图分类号：

学科分类号：

摘要：

In this paper, an Edge-information Guided Multi-scale Feature-fusion Network (EGMFNet) is proposed to solve the problems of unclear boundary and incomplete structure of saliency map extracted by deep learning saliency target detection method based on FCN and U-shaped network architecture. EGMFNet uses Residual muti-Channel Fusion Block (RCFBlock) and uses a nested U-shaped network architecture as the backbone model. At the same time, an Edge-information Guided Global Spatial Attention Module (EGSAM) is introduced at the lower level of the network to enhance spatial features and edge features. In addition, image boundary loss is introduced into the loss function, which is used to improve the quality of saliency map and keep clearer boundaries in the learning process. Experiments on four benchmark data sets show that the F values of the proposed method are increased by 1.5%, 2.7%, 1.8% and 1.6% compared with typical methods, which verifies the effectiveness of EGMFNet network model. © 2023 Chinese Society of Astronautics. All rights reserved.

引用

共 16 条

[1]

Sun Zhaolei, Hui Bin, Qin Mofan, Et al., Object detection method based on saliency measure for infrared radiation image, Infrared and Laser Engineering, 44, 9, pp. 2633-2637, (2015)

[2]

Huang Mengke, Liu Zhi, Ye Linwei, Et al., Saliency detection via multi-level integration and multi-scale fusion neural networks, Neurocomputing, 364, 9, pp. 310-321, (2019)

[3]

Li Tengpeng, Song Huihui, Zhang Kaihua, Et al., Recurrent reverse attention guided residual learning for saliency object detection, Neurocomputing, 389, 3, pp. 170-178, (2020)

[4]

Jia Fengwei, Wang Xuan, Guan Jian, Et al., Bi-connect net for salient object detection, Neurocomputing, 384, 1, pp. 142-155, (2020)

[5]

Ullah Inam, Jian Muwei, Hussain Sumaira, Et al., Global context-aware multi-scale features aggregative network for salient object detection, Neurocomputing, 455, 1, pp. 139-153, (2021)

[6]

Jiang Guoqing, Wan Lanjun, Detection of dim and small infrared targets based on the most appropriate contrast saliency analysis, Infrared and Laser Engineering, 50, 4, (2021)

[7]

Li Ning, Huang Jincai, Feng Yanghe, Construction of multichannel fusion salient object detection network based on gating mechanism and pooling network, Multimedia Tools and Applications, 81, pp. 12111-12126, (2021)

[8]

Li Zun, Lang Congyan, Liew Junhao, Et al., Cross-layer feature pyramid network for salient object detection, IEEE Transactions on Image Processing, 30, pp. 4587-4598, (2021)

[9]

Chen Tianyou, Hu Xiaoguang, Xiao Jin, Et al., BPFINet: Boundary-aware progressive feature integration network for salient object detection, Neurocomputing, 451, 8, pp. 152-166, (2021)

[10]

Yao Zhaojian, Wang Luping, ERBANet: Enhancing region and boundary awareness for salient object detection, Neurocomputing, 448, pp. 152-167, (2021)

← 1 2 →