{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T05:51:56Z","timestamp":1777528316342,"version":"3.51.4"},"reference-count":36,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,12,12]],"date-time":"2023-12-12T00:00:00Z","timestamp":1702339200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,12,12]],"date-time":"2023-12-12T00:00:00Z","timestamp":1702339200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100007109","name":"Universit\u00e0 degli Studi di Ferrara","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100007109","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Appl Intell"],"published-print":{"date-parts":[[2024,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Nowadays, deep learning is a key technology for many applications in the industrial area such as anomaly detection. The role of Machine Learning (ML) in this field relies on the ability of training a network to learn to inspect images to determine the presence or not of anomalies. Frequently, in Industry 4.0 w.r.t. the anomaly detection task, the images to be analyzed are not optimal, since they contain edges or areas, that are not of interest which could lead the network astray. Thus, this study aims at identifying a systematic way to train a neural network to make it able to focus only on the area of interest. The study is based on the definition of a loss to be applied in the training phase of the network that, using masks, gives higher weight to the anomalies identified within the area of interest. The idea is to add an<jats:italic>Overlap Coefficient<\/jats:italic>to the standard cross-entropy. In this way, the more the identified anomaly is outside the<jats:italic>Area of Interest<\/jats:italic>(AOI) the greater is the loss. We call the resulting loss<jats:italic>Cross-Entropy Overlap Distance<\/jats:italic>(CEOD). The advantage of adding the masks in the training phase is that the network is forced to learn and recognize defects only in the area circumscribed by the mask. The added benefit is that, during inference, these masks will no longer be needed. Therefore, there is no difference, in terms of execution times, between a standard Convolutional Neural Network (CNN) and a network trained with this loss. In some applications, the masks themselves are determined at run-time through a trained segmentation network, as we have done for instance in the \"Machine learning for visual inspection and quality control\" project, funded by the MISE Competence Center Bi-REX.<\/jats:p>","DOI":"10.1007\/s10489-023-05177-0","type":"journal-article","created":{"date-parts":[[2023,12,12]],"date-time":"2023-12-12T06:02:39Z","timestamp":1702360959000},"page":"414-427","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":7,"title":["Exploiting CNN\u2019s visual explanations to drive anomaly detection"],"prefix":"10.1007","volume":"54","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7656-7204","authenticated-orcid":false,"given":"Michele","family":"Fraccaroli","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alice","family":"Bizzarri","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Paolo","family":"Casellati","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Evelina","family":"Lamma","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2023,12,12]]},"reference":[{"issue":"1","key":"5177_CR1","doi-asserted-by":"publisher","first-page":"35","DOI":"10.1007\/s00170-021-06592-8","volume":"113","author":"X Zheng","year":"2021","unstructured":"Zheng X, Zheng S, Kong Y, Chen J (2021) Recent advances in surface defect inspection of industrial products using deep learning techniques. The International Journal of Advanced Manufacturing Technology. 113(1):35\u201358. https:\/\/doi.org\/10.1007\/s00170-021-06592-8","journal-title":"The International Journal of Advanced Manufacturing Technology."},{"issue":"7","key":"5177_CR2","doi-asserted-by":"publisher","first-page":"1829","DOI":"10.3390\/s20071829","volume":"20","author":"G Liu","year":"2020","unstructured":"Liu G, Yang N, Guo L, Guo S, Chen Z (2020) A one-stage approach for surface anomaly detection with background suppression strategies. Sensors. 20(7):1829","journal-title":"Sensors."},{"key":"5177_CR3","unstructured":"MarkouM, Singh S (2004) Novelty detection: a reviewpart 2: statistical approaches"},{"key":"5177_CR4","doi-asserted-by":"crossref","unstructured":"Hodge V (2004) Austin J: A survey of outlier detection methodologies. Artif Intell Rev","DOI":"10.1007\/s10462-004-4304-y"},{"key":"5177_CR5","doi-asserted-by":"crossref","unstructured":"Weimer\u00a0D, Shpitalni M, Scholz-Reiter B (2016) Design of deep convolutional neural network architectures for automated feature extraction in industrial inspection. CIRP","DOI":"10.1016\/j.cirp.2016.04.072"},{"key":"5177_CR6","doi-asserted-by":"crossref","unstructured":"Staar B, L\u00fctjen M, Freitag M (2019) Anomaly detection with convolutional neural networks for industrial surface inspection. Procedia CIRP","DOI":"10.1016\/j.procir.2019.02.123"},{"key":"5177_CR7","doi-asserted-by":"crossref","unstructured":"Hoffer E, Ailon N (2015) Deep metric learning using triplet network. International Workshop on Similarity-Based Pattern Recognition","DOI":"10.1007\/978-3-319-24261-3_7"},{"key":"5177_CR8","doi-asserted-by":"crossref","unstructured":"Akcay S, Atapour -Abarghouei A, Breckon TP (2018) Ganomaly: Semi-supervised anomaly detection via adversarial training. Asian conference on computer vision. Springer, pp 622\u2013637","DOI":"10.1007\/978-3-030-20893-6_39"},{"key":"5177_CR9","unstructured":"Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial nets. Adv Neural Inf Process Syst 27"},{"issue":"1","key":"5177_CR10","first-page":"1","volume":"2","author":"A Jinwon","year":"2015","unstructured":"Jinwon A, Sungzoon C (2015) Variational autoencoder based anomaly detection using reconstruction probability. Special Lect IE 2(1):1\u201318","journal-title":"Special Lect IE"},{"key":"5177_CR11","unstructured":"Kingma D, Welling M (2014) Auto-encoding variational bayes. International Conference on Learning Representation"},{"key":"5177_CR12","doi-asserted-by":"publisher","first-page":"7773481","DOI":"10.1155\/2023\/7773481","volume":"2023","author":"N Ferrari","year":"2023","unstructured":"Ferrari N, Fraccaroli M, Lamma E (2023) Grd-net: Generative-reconstructive-discriminative anomaly detection with region of interest attention module. Int J Intell Syst 2023:7773481. https:\/\/doi.org\/10.1155\/2023\/7773481","journal-title":"Int J Intell Syst"},{"issue":"9","key":"5177_CR13","doi-asserted-by":"publisher","first-page":"2095","DOI":"10.3390\/ma14092095","volume":"14","author":"IY Moon","year":"2021","unstructured":"Moon IY, Lee HW, Kim S-J, Oh Y-S, Jung J, Kang S-H (2021) Analysis of the region of interest according to cnn structure in hierarchical pattern surface inspection using cam. Materials. 14(9):2095","journal-title":"Materials."},{"key":"5177_CR14","unstructured":"Cohen N, Hoshen Y (2020) Sub-image anomaly detection with deep pyramid correspondences. arXiv:2005.02357"},{"key":"5177_CR15","doi-asserted-by":"crossref","unstructured":"Yi J, Yoon S (2020) Patch svdd: Patch-level svdd for anomaly detection and segmentation. In: Proceedings of the Asian Conference on Computer Vision","DOI":"10.1007\/978-3-030-69544-6_23"},{"key":"5177_CR16","doi-asserted-by":"crossref","unstructured":"Kimura D, Chaudhury S, Narita M, Munawar A, Tachibana R (2020) Adversarial discriminative attention for robust anomaly detection. In: Proceedings of the IEEE\/CVF winter conference on applications of computer vision. pp 2172\u20132181","DOI":"10.1109\/WACV45572.2020.9093428"},{"key":"5177_CR17","doi-asserted-by":"crossref","unstructured":"Venkataramanan S, Peng K-C, Singh RV, Mahalanobis A (2020) Attention guided anomaly localization in images. In: European conference on computer vision. Springer, pp 485\u2013503","DOI":"10.1007\/978-3-030-58520-4_29"},{"key":"5177_CR18","unstructured":"Song J, Kong K, Park Y-I, Kim S-G, Kang S-J (2021) Anoseg: Anomaly segmentation network using self-supervised learning. arXiv:2110.03396"},{"key":"5177_CR19","doi-asserted-by":"crossref","unstructured":"Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D (2017) Grad-cam: Visual explanations from deep networks via gradient-based localization. In: Proceedings of the IEEE international conference on computer vision. pp. 618\u2013626","DOI":"10.1109\/ICCV.2017.74"},{"issue":"17","key":"5177_CR20","doi-asserted-by":"publisher","first-page":"14881","DOI":"10.1007\/s00521-022-07264-8","volume":"34","author":"F Xu","year":"2022","unstructured":"Xu F, Wang H, Sun X, Fu X (2022) Refined marine object detector with attention-based spatial pyramid pooling networks and bidirectional feature fusion strategy. Neural Comput Appl 34(17):14881\u201314894","journal-title":"Neural Comput Appl"},{"key":"5177_CR21","doi-asserted-by":"publisher","first-page":"3637","DOI":"10.1007\/s00521-020-05217-7","volume":"33","author":"F Xu","year":"2021","unstructured":"Xu F, Wang H, Peng J, Fu X (2021) Scale-aware feature pyramid architecture for marine object detection. Neural Comput Appl 33:3637\u20133653","journal-title":"Neural Comput Appl"},{"issue":"10","key":"5177_CR22","doi-asserted-by":"publisher","first-page":"10484","DOI":"10.1109\/TVT.2020.3009162","volume":"69","author":"H Wang","year":"2020","unstructured":"Wang H, Peng J, Zhao Y, Fu X (2020) Multi-path deep cnns for fine-grained car recognition. IEEE Trans Veh Technol 69(10):10484\u201310493","journal-title":"IEEE Trans Veh Technol"},{"key":"5177_CR23","doi-asserted-by":"crossref","unstructured":"Zhou A, Ai B, Qu P, Shao W (2021) Defect detection for highly reflective rotary surfaces: An overview. Meas Sci Technol 32(6):062001","DOI":"10.1088\/1361-6501\/abd579"},{"issue":"4","key":"5177_CR24","doi-asserted-by":"publisher","first-page":"1038","DOI":"10.1007\/s11263-020-01400-4","volume":"129","author":"P Bergmann","year":"2021","unstructured":"Bergmann P, Batzner K, Fauser M, Sattlegger D, Steger C (2021) The mvtec anomaly detection dataset: a comprehensive real-world dataset for unsupervised anomaly detection. Int J Comput Vision 129(4):1038\u20131059","journal-title":"Int J Comput Vision"},{"key":"5177_CR25","doi-asserted-by":"crossref","unstructured":"Bergmann P, Fauser M, Sattlegger D, Steger C (2019) Mvtec ad\u2013a comprehensive real-world dataset for unsupervised anomaly detection. In: Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition. pp. 9592\u20139600","DOI":"10.1109\/CVPR.2019.00982"},{"key":"5177_CR26","unstructured":"Vijaymeena MK, Kavitha K (2016) A survey on similarity measures in text mining"},{"key":"5177_CR27","doi-asserted-by":"crossref","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","DOI":"10.1109\/CVPR.2016.319"},{"key":"5177_CR28","doi-asserted-by":"crossref","unstructured":"He K, Gkioxari G, Doll\u00e1r P, Girshick R (2018) Mask R-CNN","DOI":"10.1109\/ICCV.2017.322"},{"issue":"4","key":"5177_CR29","doi-asserted-by":"publisher","first-page":"834","DOI":"10.1109\/TPAMI.2017.2699184","volume":"40","author":"L-C Chen","year":"2017","unstructured":"Chen L-C, Papandreou G, Kokkinos I, Murphy K, Yuille AL (2017) Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE Trans Pattern Anal Mach Intell 40(4):834\u2013848","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"5177_CR30","doi-asserted-by":"crossref","unstructured":"Ronneberger O, Fischer P, Brox T (2015) U-net: Convolutional networks for biomedical image segmentation. International conference on medical image computing and computer-assisted intervention. Springer, pp 234\u2013241","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"5177_CR31","doi-asserted-by":"crossref","unstructured":"Mannor S, Peleg D, Rubinstein R (2005) The cross entropy method for classification. In: Proceedings of the 22nd international conference on machine learning. pp 561\u2013568","DOI":"10.1145\/1102351.1102422"},{"key":"5177_CR32","unstructured":"Tan M, Le Q (2019) Efficientnet: Rethinking model scaling for convolutional neural networks. International Conference on Machine Learning. PMLR, pp 6105\u20136114"},{"key":"5177_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. 2009 IEEE conference on computer vision and pattern recognition. Ieee, pp 248\u2013255","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"5177_CR34","unstructured":"Bagherinezhad H, Horton M, Rastegari M, Farhadi A (2018) Label refinery: Improving imagenet classification through label progression. arXiv:1805.02641"},{"key":"5177_CR35","unstructured":"Yang S, Xiao W, Zhang M, Guo S, Zhao J, Shen F (2022) Image data augmentation for deep learning: A survey. arXiv:2204.08610"},{"issue":"1","key":"5177_CR36","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s40537-019-0197-0","volume":"6","author":"C Shorten","year":"2019","unstructured":"Shorten C, Khoshgoftaar TM (2019) A survey on image data augmentation for deep learning. J Big Data 6(1):1\u201348","journal-title":"J Big Data"}],"container-title":["Applied Intelligence"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10489-023-05177-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10489-023-05177-0\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10489-023-05177-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,11,5]],"date-time":"2024-11-05T20:06:50Z","timestamp":1730837210000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10489-023-05177-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,12,12]]},"references-count":36,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2024,1]]}},"alternative-id":["5177"],"URL":"https:\/\/doi.org\/10.1007\/s10489-023-05177-0","relation":{},"ISSN":["0924-669X","1573-7497"],"issn-type":[{"value":"0924-669X","type":"print"},{"value":"1573-7497","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,12,12]]},"assertion":[{"value":"12 November 2023","order":1,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"12 December 2023","order":2,"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 competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflicts of interest"}}]}}