{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T10:18:12Z","timestamp":1779099492707,"version":"3.51.4"},"reference-count":26,"publisher":"MDPI AG","issue":"13","license":[{"start":{"date-parts":[[2021,7,1]],"date-time":"2021-07-01T00:00:00Z","timestamp":1625097600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100005320","name":"Xidian University","doi-asserted-by":"publisher","award":["JB210206"],"award-info":[{"award-number":["JB210206"]}],"id":[{"id":"10.13039\/501100005320","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100005320","name":"Xidian University","doi-asserted-by":"publisher","award":["JC2111"],"award-info":[{"award-number":["JC2111"]}],"id":[{"id":"10.13039\/501100005320","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Target recognition is one of the most challenging tasks in synthetic aperture radar (SAR) image processing since it is highly affected by a series of pre-processing techniques which usually require sophisticated manipulation for different data and consume huge calculation resources. To alleviate this limitation, numerous deep-learning based target recognition methods are proposed, particularly combined with convolutional neural network (CNN) due to its strong capability of data abstraction and end-to-end structure. In this case, although complex pre-processing can be avoided, the inner mechanism of CNN is still unclear. Such a \u201cblack box\u201d only tells a result but not what CNN learned from the input data, thus it is difficult for researchers to further analyze the causes of errors. Layer-wise relevance propagation (LRP) is a prevalent pixel-level rearrangement algorithm to visualize neural networks\u2019 inner mechanism. LRP is usually applied in sparse auto-encoder with only fully-connected layers rather than CNN, but such network structure usually obtains much lower recognition accuracy than CNN. In this paper, we propose a novel LRP algorithm particularly designed for understanding CNN\u2019s performance on SAR image target recognition. We provide a concise form of the correlation between output of a layer and weights of the next layer in CNNs. The proposed method can provide positive and negative contributions in input SAR images for CNN\u2019s classification, viewed as a clear visual understanding of CNN\u2019s recognition mechanism. Numerous experimental results demonstrate the proposed method outperforms common LRP.<\/jats:p>","DOI":"10.3390\/s21134536","type":"journal-article","created":{"date-parts":[[2021,7,1]],"date-time":"2021-07-01T12:03:27Z","timestamp":1625141007000},"page":"4536","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":33,"title":["CNN-LRP: Understanding Convolutional Neural Networks Performance for Target Recognition in SAR Images"],"prefix":"10.3390","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5056-6596","authenticated-orcid":false,"given":"Bo","family":"Zang","sequence":"first","affiliation":[{"name":"School of Electronic Engineering, Xidian University, Xi\u2019an 710071, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7667-1298","authenticated-orcid":false,"given":"Linlin","family":"Ding","sequence":"additional","affiliation":[{"name":"School of Electronic Engineering, Xidian University, Xi\u2019an 710071, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0383-4794","authenticated-orcid":false,"given":"Zhenpeng","family":"Feng","sequence":"additional","affiliation":[{"name":"School of Electronic Engineering, Xidian University, Xi\u2019an 710071, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7962-3344","authenticated-orcid":false,"given":"Mingzhe","family":"Zhu","sequence":"additional","affiliation":[{"name":"School of Electronic Engineering, Xidian University, Xi\u2019an 710071, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tao","family":"Lei","sequence":"additional","affiliation":[{"name":"School of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi\u2019an 710021, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mengdao","family":"Xing","sequence":"additional","affiliation":[{"name":"School of Electronic Engineering, Xidian University, Xi\u2019an 710071, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xianda","family":"Zhou","sequence":"additional","affiliation":[{"name":"Beijing Aerospace Automatic Control Institute, Beijing 100070, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2021,7,1]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Wang, Y.P., Zhang, Y.B., Qu, H.Q., and Tian, Q. (2018, January 13\u201315). Target Detection and Recognition Based on Convolutional Neural Network for SAR Image. Proceedings of the 2018 11th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), Beijing, China.","DOI":"10.1109\/CISP-BMEI.2018.8633151"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1882","DOI":"10.1109\/LGRS.2018.2865608","article-title":"Multiple Feature Aggregation Using Convolutional Neural Networks for SAR Image-Based Automatic Target Recognition","volume":"15","author":"Chao","year":"2018","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Zhang, Y., Song, Y., Wang, Y.P., and Qu, H.Q. (2018, January 13\u201315). A fast training method for SAR large scale samples based on CNN for targets recognition. Proceedings of the 2018 11th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), Beijing, China.","DOI":"10.1109\/CISP-BMEI.2018.8633175"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Zhu, M., Zhou, X., and Zang, B. (2018). Micro-Doppler Feature Extraction of Inverse Synthetic Aperture Imaging Laser Radar Using Singular-Spectrum Analysis. Sensors, 18.","DOI":"10.3390\/s18103303"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Zang, B., Zhou, X., and Zhu, M. (2019). Application of S-Transform Random Consistency in Inverse Synthetic Aperture Imaging Laser Radar Imaging. Appl. Sci., 9.","DOI":"10.3390\/app9112313"},{"key":"ref_6","first-page":"1","article-title":"Feature-Enhanced Speckle Reduction via Low-Rank and Space-Angle Continuity for Circular SAR Target Recognition","volume":"99","author":"Chen","year":"2020","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Wu, T.D., Yen, Y., and Wang, J.H. (2020, January 26\u201328). Automatic Target Recognition in SAR Images Based on a Combination of CNN and SVM. Proceedings of the 2020 International Workshop on Electromagnetics: Applications and Student Innovation Competition (iWEM), Penghu, Taiwan.","DOI":"10.1109\/iWEM49354.2020.9237422"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"7282","DOI":"10.1109\/TGRS.2018.2849967","article-title":"SAR ATR of Ground V ehicles Based on LM-BN-CNN","volume":"56","author":"Zhou","year":"2018","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TGRS.2020.3038405","article-title":"FEC: A Feature Fusion Framework for SAR Target Recognition Based on Electromagnetic Scattering Features and Deep CNN Features","volume":"99","author":"Zhang","year":"2020","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"701","DOI":"10.1109\/TCSVT.2020.2987346","article-title":"SPAM-Net: A CNN-Based SAR Target Recognition Network With Pose Angle Marginalization Learning","volume":"31","author":"Oh","year":"2021","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Dong, Y.P., Su, H., and Wu, B.Y. (2019, January 15\u201320). Efficient Decision-based Black-box Adversarial Attacks on Face Recognition. Proceedings of the 2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00790"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Girshick, R., Donahue, J., and Darrell, T. (2014, January 23\u201328). Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Columbus, OH, USA.","DOI":"10.1109\/CVPR.2014.81"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"103457","DOI":"10.1016\/j.artint.2021.103457","article-title":"GLocalX - From Local to Global Explanations of Black Box AI Models","volume":"294","author":"Setzu","year":"2021","journal-title":"Artif. Intell."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Xiong, K., Zhao, G., Wang, Y., and Shi, G. (2021). SPB-Net: A Deep Network for SAR Imaging and Despeckling With DownSampled Data. IEEE Trans. Geosci. Remote Sens.","DOI":"10.1109\/TGRS.2020.3034102"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Zhou, B., Khosla, K., Lapedriza, A., Oliva, A., and Torralba, A. (2016, January 27\u201330). Learning Deep Features for Discriminative Localization. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.319"},{"key":"ref_16","unstructured":"Ramprasaath, R.S., Michael, C., and Abhishek, D. (2017, January 22\u201329). Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization. Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy."},{"key":"ref_17","unstructured":"Fu, H.G., Hu, Q.Y., Dong, X.H., Guo, Y.I., Gao, Y.H., and Li, B. (2020, January 7\u201310). Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNs. Proceedings of the 2020 31th British Machine Vision Conference (BMVC), Manchester, UK."},{"key":"ref_18","unstructured":"Saurabh, D., and Harish, G.R. (2020, January 1\u20135). Ablation-CAM: Visual Explanations for Deep Convolutional Network via Gradient-free Localization. Proceedings of the 2020 IEEE Winter Conference on Applications of Computer Vision (WACV), Snowmass, CO, USA."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Feng, Z., Zhu, M., Stankovi\u0107, L., and Ji, H. (2021). Self-Matching CAM: A Novel Accurate Visual Explanation of CNNs for SAR Image Interpretation. Remote Sens., 13.","DOI":"10.3390\/rs13091772"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Bach, S., Binder, A., Montavon, G., Klauschen, F., M\u00fcller, K.R., and Samek, W. (2014). On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation. PLoS ONE, 10.","DOI":"10.1371\/journal.pone.0130140"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Amin, M.G., and Erol, B. (2018, January 23\u201327). Understanding deep neural networks performance for radar-based human motion recognition. Proceedings of the 2018 IEEE Radar Conference (RadarConf18), Oklahoma City, OK, USA.","DOI":"10.1109\/RADAR.2018.8378780"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Jokanovic, B., Amin, M., and Ahmad, F. (2016, January 2\u20136). Radar fall motion detection using deep learning. Proceedings of the 2016 IEEE Radar Conference (RadarConf16), Philadelphia, PA, USA.","DOI":"10.1109\/RADAR.2016.7485147"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Samek, W., Montavon, G., Vedaldi, A., Hansen, L., and M\u00fcller, K.R. (2019). Layer-Wise Relevance Propagation: An Overview. Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, Springer.","DOI":"10.1007\/978-3-030-28954-6"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.dsp.2017.10.011","article-title":"Methods for Interpreting and Understanding Deep Neural Networks","volume":"73","author":"Montavon","year":"2018","journal-title":"Digit. Signal Process."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Ma, X., and Wu, P. (2019). Multitemporal SAR Image Despeckling Based on a Scattering Covariance Matrix of Image Patch. Sensors, 19.","DOI":"10.3390\/s19143057"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Paulson, C., and Zelnio, E. (2014, January 19\u201323). The effect of SAR imaging conditions on classification algorithms. Proceedings of the 2014 IEEE Radar Conference (RadarConf16), Cincinnati, OH, USA.","DOI":"10.1109\/RADAR.2014.6875720"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/21\/13\/4536\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T06:24:51Z","timestamp":1760163891000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/21\/13\/4536"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,7,1]]},"references-count":26,"journal-issue":{"issue":"13","published-online":{"date-parts":[[2021,7]]}},"alternative-id":["s21134536"],"URL":"https:\/\/doi.org\/10.3390\/s21134536","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,7,1]]}}}