{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,7]],"date-time":"2026-03-07T00:47:14Z","timestamp":1772844434130,"version":"3.50.1"},"reference-count":37,"publisher":"Springer Science and Business Media LLC","issue":"20","license":[{"start":{"date-parts":[[2025,5,16]],"date-time":"2025-05-16T00:00:00Z","timestamp":1747353600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,5,16]],"date-time":"2025-05-16T00:00:00Z","timestamp":1747353600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100018693","name":"HORIZON EUROPE Framework Programme","doi-asserted-by":"publisher","award":["101120657"],"award-info":[{"award-number":["101120657"]}],"id":[{"id":"10.13039\/100018693","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Neural Comput &amp; Applic"],"published-print":{"date-parts":[[2025,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>The rapid development and widespread adoption of deep neural networks (DNNs) are attributed to their remarkable performance and practical efficacy across numerous applications. Nevertheless, the sophistication of the inference process often comes at the expense of explainability, becoming an increasingly critical concern. Explainable AI endeavors to shed light on the inner workings of neural networks, thereby enhancing trust in their predictive capabilities and facilitating more reliable and informed decision-making. Visual explanations for the behavior of DNNs contribute to understanding their functioning. However, selecting the right visualization method for a particular task and evaluating its performance remains a challenge. This paper offers a comprehensive analysis of post hoc methods for generating visual explanations, concentrating on class activation mapping (CAM) techniques. It introduces two novel approaches: high-resolution relevance propagation CAM\u00a0(HiResRP-CAM) and EigenLayer-CAM. These methods are evaluated both qualitatively and quantitatively across various experimental setups within the context of convolutional neural networks (CNNs) for image classification tasks. The evaluation encompasses diverse network architectures, datasets, and metrics, including certain proposed enhancements. The paper discusses experimental results, including instances of model failure due to adversarial noise, and explores factors to deepen understanding of model behavior, characteristics of explainability methods and current strategies for assessing explanation quality. In this way, we provide a perspective on the challenges and limitations inherent in interpreting CNNs and on the relevant aspects for choosing a specific method.<\/jats:p>","DOI":"10.1007\/s00521-025-11282-7","type":"journal-article","created":{"date-parts":[[2025,5,16]],"date-time":"2025-05-16T18:31:14Z","timestamp":1747420274000},"page":"14935-14970","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Evaluation and analysis of visual methods for CNN explainability: a novel approach and experimental study"],"prefix":"10.1007","volume":"37","author":[{"given":"Andrei","family":"Dug\u0103e\u0219escu","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7249-1871","authenticated-orcid":false,"given":"Adina Magda","family":"Florea","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2025,5,16]]},"reference":[{"key":"11282_CR1","doi-asserted-by":"publisher","first-page":"52138","DOI":"10.1109\/access.2018.2870052","volume":"6","author":"A Adadi","year":"2018","unstructured":"Adadi A, Berrada M (2018) Peeking inside the black-box: a survey on explainable artificial intelligence (XAI). IEEE Access 6:52138\u201352160. https:\/\/doi.org\/10.1109\/access.2018.2870052","journal-title":"IEEE Access"},{"issue":"1","key":"11282_CR2","doi-asserted-by":"publisher","first-page":"18","DOI":"10.3390\/e23010018","volume":"23","author":"P Linardatos","year":"2020","unstructured":"Linardatos P, Papastefanopoulos V, Kotsiantis S (2020) Explainable AI: a review of machine learning interpretability methods. Entropy 23(1):18. https:\/\/doi.org\/10.3390\/e23010018","journal-title":"Entropy"},{"issue":"5","key":"11282_CR3","doi-asserted-by":"publisher","first-page":"593","DOI":"10.3390\/electronics10050593","volume":"10","author":"J Zhou","year":"2021","unstructured":"Zhou J, Gandomi AH, Chen F, Holzinger A (2021) Evaluating the quality of machine learning explanations: a survey on methods and metrics. Electronics 10(5):593. https:\/\/doi.org\/10.3390\/electronics10050593","journal-title":"Electronics"},{"key":"11282_CR4","doi-asserted-by":"publisher","unstructured":"Zhou B, Khosla A, Lapedriza A, Oliva A, Torralba A (2016) Learning deep features for discriminative localization. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2921\u20132929. https:\/\/doi.org\/10.48550\/arXiv.1512.04150","DOI":"10.48550\/arXiv.1512.04150"},{"key":"11282_CR5","doi-asserted-by":"publisher","unstructured":"Selvaraju RR, Das A, Vedantam R, Cogswell M, Parikh D, Batra D (2017) Grad-CAM: why did you say that? Visual explanations from deep networks via gradient-based localization. In: Proceedings of the IEEE international conference on computer vision, pp 618\u2013626. https:\/\/doi.org\/10.48550\/arXiv.1611.07450","DOI":"10.48550\/arXiv.1611.07450"},{"key":"11282_CR6","doi-asserted-by":"publisher","unstructured":"Chattopadhyay A, Sarkar A, Howlader P, Balasubramanian VN (2018) Grad-CAM++: improved visual explanations for deep convolutional networks. In: 2018 IEEE winter conference on applications of computer vision (WACV), pp 839\u2013847. https:\/\/doi.org\/10.48550\/arXiv.1710.11063","DOI":"10.48550\/arXiv.1710.11063"},{"key":"11282_CR7","doi-asserted-by":"publisher","unstructured":"Wang H, Wang Z, Du M, Yang F, Zhang Z, Ding S, Mardziel P, Hu X (2020) Score-CAM: score-weighted visual explanations for convolutional neural networks. In: IEEE\/CVF conference on computer vision and pattern recognition workshops (CVPRW), pp 24\u201325. https:\/\/doi.org\/10.48550\/arXiv.1910.01279","DOI":"10.48550\/arXiv.1910.01279"},{"key":"11282_CR8","doi-asserted-by":"publisher","unstructured":"Poppi S, Cornia M, Baraldi L, Cucchiara R (2021) Revisiting the evaluation of class activation mapping for explainability: a novel metric and experimental analysis. arXiv:2104.10252 [cs]. https:\/\/doi.org\/10.48550\/arXiv.2104.10252","DOI":"10.48550\/arXiv.2104.10252"},{"key":"11282_CR9","doi-asserted-by":"publisher","first-page":"101805","DOI":"10.1016\/j.inffus.2023.101805","volume":"99","author":"S Ali","year":"2023","unstructured":"Ali S, Abuhmed T, El-Sappagh S, Muhammad K, Alonso-Moral JM, Confalonieri R, Guidotti R, Ser JD, D\u00edaz-Rodr\u00edguez N, Herrera F (2023) Explainable Artificial Intelligence (XAI): what we know and what is left to attain Trustworthy Artificial Intelligence. Inf Fusion 99:101805. https:\/\/doi.org\/10.1016\/j.inffus.2023.101805","journal-title":"Inf Fusion"},{"issue":"7","key":"11282_CR10","doi-asserted-by":"publisher","first-page":"e0130140","DOI":"10.1371\/journal.pone.0130140","volume":"10","author":"S Bach","year":"2015","unstructured":"Bach S, Binder A, Montavon G, Klauschen F, M\u00fcller K-R, Samek W (2015) On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PLoS ONE 10(7):e0130140. https:\/\/doi.org\/10.1371\/journal.pone.0130140","journal-title":"PLoS ONE"},{"key":"11282_CR11","doi-asserted-by":"publisher","unstructured":"Montavon G, Binder A, Lapuschkin S, Samek W, M\u00fcller K-R (2019) Layer-wise relevance propagation: an overview. In: Explainable AI: interpreting, explaining and visualizing deep learning, pp 193\u2013209. https:\/\/doi.org\/10.1007\/978-3-030-28954-6_10","DOI":"10.1007\/978-3-030-28954-6_10"},{"key":"11282_CR12","doi-asserted-by":"publisher","unstructured":"Gu J, Yang Y, Tresp V (2019) Understanding individual decisions of CNNs via contrastive backpropagation. In: Lecture notes in computer science, pp 119\u2013134. https:\/\/doi.org\/10.1007\/978-3-030-20893-6_8","DOI":"10.1007\/978-3-030-20893-6_8"},{"key":"11282_CR13","doi-asserted-by":"publisher","unstructured":"Collins E, Achanta R, S\u00fcsstrunk S (2018) Deep feature factorization for concept discovery. https:\/\/doi.org\/10.48550\/arXiv.1806.10206","DOI":"10.48550\/arXiv.1806.10206"},{"key":"11282_CR14","doi-asserted-by":"publisher","unstructured":"Fu R, Hu Q, Dong X, Guo Y, Gao Y, Li B (2020) Axiom-based Grad-CAM: towards accurate visualization and explanation of CNNs. In: Proceedings of the British machine vision conference. https:\/\/doi.org\/10.48550\/arXiv.2008.02312","DOI":"10.48550\/arXiv.2008.02312"},{"issue":"8","key":"11282_CR15","doi-asserted-by":"publisher","first-page":"5875","DOI":"10.1109\/tip.2021.3089943","volume":"30","author":"P-T Jiang","year":"2021","unstructured":"Jiang P-T, Zhang C-B, Hou Q, Cheng M-M, Wei Y (2021) LayerCAM: exploring hierarchical class activation maps for localization. IEEE Trans Image Process 30(8):5875\u20135888. https:\/\/doi.org\/10.1109\/tip.2021.3089943","journal-title":"IEEE Trans Image Process"},{"key":"11282_CR16","doi-asserted-by":"publisher","unstructured":"Srinivas S, Fleuret F (2019) Full-gradient representation for neural network visualization. arXiv:1905.00780 [cs, stat]. https:\/\/doi.org\/10.48550\/arXiv.1905.00780","DOI":"10.48550\/arXiv.1905.00780"},{"key":"11282_CR17","doi-asserted-by":"publisher","unstructured":"Naidu R, Ghosh A, Maurya Y, Kundu SS (2020) IS-CAM: integrated score-CAM for axiomatic-based explanations. https:\/\/doi.org\/10.48550\/arXiv.2010.03023","DOI":"10.48550\/arXiv.2010.03023"},{"key":"11282_CR18","doi-asserted-by":"publisher","unstructured":"Wang H, Naidu R, Michael J, Kundu SS (2020) SS-CAM: smoothed score-CAM for sharper visual feature localization. arXiv:2006.14255 [cs]. https:\/\/doi.org\/10.48550\/arXiv.2006.14255","DOI":"10.48550\/arXiv.2006.14255"},{"key":"11282_CR19","doi-asserted-by":"publisher","unstructured":"Desai S, Ramaswamy HG (2020) Ablation-CAM: visual explanations for deep convolutional network via gradient-free localization. In: Proceedings of the IEEE winter conference on applications of computer vision, pp 972\u2013980. https:\/\/doi.org\/10.1109\/WACV45572.2020.9093360","DOI":"10.1109\/WACV45572.2020.9093360"},{"key":"11282_CR20","doi-asserted-by":"publisher","unstructured":"Muhammad MB, Yeasin M (2020) Eigen-CAM: class activation map using principal components. In: International joint conference on neural networks (IJCNN), pp 1\u20137. https:\/\/doi.org\/10.1109\/IJCNN48605.2020.9206626","DOI":"10.1109\/IJCNN48605.2020.9206626"},{"key":"11282_CR21","unstructured":"Gildenblat J (2021) PyTorch library for CAM methods. GitHub. https:\/\/github.com\/jacobgil\/pytorch-grad-cam"},{"key":"11282_CR22","doi-asserted-by":"publisher","unstructured":"Lee JR, Kim S, Park I, Eo T, Hwang D (2021) Relevance-CAM: your model already knows where to look. In: IEEE\/CVF conference on computer vision and pattern recognition (CVPR), pp 14939\u201314948. https:\/\/doi.org\/10.1109\/CVPR46437.2021.01470","DOI":"10.1109\/CVPR46437.2021.01470"},{"key":"11282_CR23","doi-asserted-by":"publisher","unstructured":"Qiu C, Jin F, Zhang Y (2023) Empowering CAM-based methods with capability to generate fine-grained and high-faithfulness explanations. https:\/\/doi.org\/10.48550\/arXiv.2303.09171","DOI":"10.48550\/arXiv.2303.09171"},{"key":"11282_CR24","doi-asserted-by":"publisher","unstructured":"Zeiler MD, Fergus R (2013) Visualizing and understanding convolutional networks. https:\/\/doi.org\/10.48550\/arXiv.1311.2901","DOI":"10.48550\/arXiv.1311.2901"},{"key":"11282_CR25","doi-asserted-by":"publisher","unstructured":"Springenberg JT, Dosovitskiy A, Brox T, Riedmiller M (2015) Striving for simplicity: the all convolutional net. In: 3rd International conference on learning representations, San Diego, CA, USA. https:\/\/doi.org\/10.48550\/arXiv.1412.6806","DOI":"10.48550\/arXiv.1412.6806"},{"key":"11282_CR26","doi-asserted-by":"publisher","unstructured":"Ribeiro MT, Singh S, Guestrin C (2016) Why should I trust you? Explaining the predictions of any classifier. In: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. ACM, pp 1135\u20131144. https:\/\/doi.org\/10.48550\/arXiv.1602.04938","DOI":"10.48550\/arXiv.1602.04938"},{"key":"11282_CR27","doi-asserted-by":"publisher","unstructured":"Lundberg S, Lee S-I (2017) A unified approach to interpreting model predictions. arXiv:1705.07874 [cs, stat]. https:\/\/doi.org\/10.48550\/arXiv.1705.07874","DOI":"10.48550\/arXiv.1705.07874"},{"key":"11282_CR28","unstructured":"Shrikumar A, Greenside P, Kundaje A. Learning important features through propagating activation differences. In: Proceedings of the 34th international conference on machine learning, international convention centre, Sydney, Australia, vol 70, pp 3145\u20133153. http:\/\/proceedings.mlr.press\/v70\/shrikumar17a.html"},{"key":"11282_CR29","doi-asserted-by":"publisher","DOI":"10.1145\/3563691","author":"R Ibrahim","year":"2023","unstructured":"Ibrahim R, Shafiq MO (2023) Explainable convolutional neural networks: a taxonomy, review, and future directions. ACM Comput Surv. https:\/\/doi.org\/10.1145\/3563691","journal-title":"ACM Comput Surv"},{"key":"11282_CR30","doi-asserted-by":"publisher","unstructured":"Li X-H, Shi Y, Li H, Bai W, Song Y, Cao CC, Chen L (2020) Quantitative evaluations on saliency methods: an experimental study. https:\/\/doi.org\/10.48550\/arXiv.2012.15616","DOI":"10.48550\/arXiv.2012.15616"},{"key":"11282_CR31","unstructured":"Athanasoula I, Apostolopoulos I, Groumpos P (2023) Evaluation of GradCAM for explaining Deep Learning\u2019s decisions on various medical imaging datasets"},{"key":"11282_CR32","doi-asserted-by":"publisher","first-page":"119809","DOI":"10.1016\/j.eswa.2023.119809","volume":"222","author":"G Batchuluun","year":"2023","unstructured":"Batchuluun G, Choi J, Park KR (2023) CAM-CAN: class activation map-based categorical adversarial network. Expert Syst Appl 222:119809. https:\/\/doi.org\/10.1016\/j.eswa.2023.119809","journal-title":"Expert Syst Appl"},{"key":"11282_CR33","doi-asserted-by":"crossref","unstructured":"Deng J, Dong W, Socher R, Li L-J, Li K, Fei-Fei L (2009) Imagenet: a large-scale hierarchical image database. In: IEEE conference on computer vision and pattern recognition, pp 248\u2013255. https:\/\/image-net.org\/","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"11282_CR34","unstructured":"Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M, Berg AC, Fei-Fei. ImageNet large scale visual recognition challenge (ILSVRC). https:\/\/www.image-net.org\/challenges\/LSVRC\/"},{"key":"11282_CR35","unstructured":"Howard J (2019) Imagenette: a smaller subset of 10 easily classified classes from Imagenet. GitHub. https:\/\/github.com\/fastai\/imagenette"},{"key":"11282_CR36","doi-asserted-by":"publisher","unstructured":"Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L, Desmaison A, Kopf A, Yang E, DeVito Z, Raison M, Tejani A, Chilamkurthy S, Steiner B, Fang L, Bai J, Chintala S (2019) PyTorch: an imperative style, high-performance deep learning library. https:\/\/doi.org\/10.48550\/arXiv.1912.01703","DOI":"10.48550\/arXiv.1912.01703"},{"key":"11282_CR37","doi-asserted-by":"publisher","unstructured":"Moosavi-Dezfooli S-M, Fawzi A, Frossard P (2016) DeepFool: a simple and accurate method to fool deep neural networks. https:\/\/doi.org\/10.48550\/arXiv.1511.04599","DOI":"10.48550\/arXiv.1511.04599"}],"container-title":["Neural Computing and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-025-11282-7.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00521-025-11282-7\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-025-11282-7.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,7]],"date-time":"2025-07-07T11:06:21Z","timestamp":1751886381000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00521-025-11282-7"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,16]]},"references-count":37,"journal-issue":{"issue":"20","published-print":{"date-parts":[[2025,7]]}},"alternative-id":["11282"],"URL":"https:\/\/doi.org\/10.1007\/s00521-025-11282-7","relation":{},"ISSN":["0941-0643","1433-3058"],"issn-type":[{"value":"0941-0643","type":"print"},{"value":"1433-3058","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5,16]]},"assertion":[{"value":"8 August 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"5 April 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 May 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"Not applicable.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}}]}}