Opto-Electronic Engineering, Volume. 49, Issue 12, 220029(2022)
Cascade pooling self-attention research for remote sensing image retrieval
Figures & Tables(12)
Fig. 4. Sample images. (a) Sample images of the UC Merced data set; (b) Sample images of the AID data set; (c) Sample images of the NWPU-RESISC45 data set
Table 1. mAP value of different attention modules
View tableView in ArticleTable 1. mAP value of different attention modules
网络结构 UC Merced/% AID/% NWPU-RESISC45/% ResNet50 91.17 87.35 60.07 ResNet50_DANet 92.13 88.13 73.02 ResNet50_CSM 91.50 92.38 77.12 ResNet50_SSM 92.60 92.91 77.13 ResNet50_CBAM 93.72 93.33 78.19 ResNet50_DSM 92.67 93.48 78.28
Table 2. mAP value of different pooling methods
View tableView in ArticleTable 2. mAP value of different pooling methods
网络结构 UC Merced/% AID/% NWPU-RESISC45/% ResNet50 91.17 87.35 60.07 ResNet50_CSM 91.50 92.38 77.12 ResNet50_SSM 92.60 92.91 77.13 ResNet50_DSM 92.67 93.48 78.28 ResNet50_CPCSM 93.35 93.23 78.17 ResNet50_CPSSM 93.62 93.15 77.19 ResNet50_CPSM 94.12 93.76 79.00
Table 3. Comparison of different dimensionality reduction methods
View tableView in ArticleTable 3. Comparison of different dimensionality reduction methods
方法 mAP/% 时间/ms 维度 - 94.12 24.06 100352 PCA 95.85 23.11 20 PCA 93.56 23.46 64 LDA 98.07 29.67 20 KSH 97.43 23.32 20 KSH 98.23 23.44 64
Table 4. Comparison of different methods on UC Merced data set
View tableView in ArticleTable 4. Comparison of different methods on UC Merced data set
参考文献 方法 mAP/% Ye等[ 28 ]Pool5_W 94.86 Ye等[ 29 ]Pool5 95.62 叶发茂等[ 30 ]Pool5_ACO 96.71 Roy等[ 31 ]MiLaN 91.60 Wang等[ 17 ]W-CAN 95.10 Zhang等[ 7 ]- 97.89 Song等[ 32 ]DHCNN 98.02 Liu等[ 33 ]FAH 98.30 Cheng等[ 8 ]- 97.77 Zhou等[ 9 ]Gabor-CA-ResNet 97.50 本文 ResNet50_CPSM_KSH 98.23
Table 5. Comparison of different methods on AID data set
View tableView in ArticleTable 5. Comparison of different methods on AID data set
参考文献 方法 mAP/% Liu等[ 33 ]FAH 91.75 Roy等[ 31 ]MiLaN 92.60 Song等[ 32 ]DHCNN 94.27 Cheng等[ 8 ]- 93.84 Zhou等[ 9 ]Gabor-CA-ResNet 94.34 本文 ResNet50_CPSM_KSH 94.96
Table 6. Comparison of different methods on NWPU-RESISC45 data set
View tableView in ArticleTable 6. Comparison of different methods on NWPU-RESISC45 data set
参考文献 方法 mAP/% Tang等[ 34 ]RFM 25.63 Yang等[ 24 ]SBOW 37.02 Demir等[ 35 ]Hash 34.49 Marmanis等[ 36 ]DN7 60.54 Marmanis等[ 36 ]DN8 59.47 Demir等[ 34 ]DBOW 82.15 Imbriaco等[ 37 ]QE-S 85.70 Imbriaco等[ 37 ]V-DELF 84.00 Liu等[ 33 ]FAH 70.41 Hou等[ 38 ]- 83.07 Wang等[ 39 ]JSST 80.39 Fan等[ 40 ]N-pair-mc Loss 93.06 Fan等[ 40 ]Triplet Loss 93.82 本文 ResNet50_CPSM_KSH 94.53
Tools
Get Citation
Copy Citation Text
Gang Wu, Yun Ge, Jun Chu, Famao Ye. Cascade pooling self-attention research for remote sensing image retrieval[J]. Opto-Electronic Engineering, 2022, 49(12): 220029
Paper Information
Category: Article
Received: Mar. 30, 2022
Accepted: --
Published Online: Jan. 17, 2023
The Author Email: Ge Yun (geyun@nchu.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20