Crossref DOIMemberCiteFactoropen accessissn
SJIF Impact Factor 8.659 | ISSN: 2582-3930
IJSREM Logo

International Journal of Scientific Research in Engineering and Management

(Gateway to Global Scholars - An International, Scholarly, Multi-Disciplinary, Open Access Journal)

Peer-Reviewed Journal Designed to Meet UGC's Quality Parameters

HomeCurrent IssuePast IssuesPay OnlineConferencesEditorial BoardAbout UsContact Us
IJSREM

Loading article...

Follow Us

  • Google Scholar
  • ResearchGate
  • Academia.edu
  • Mendeley
  • LinkedIn
  • Facebook
  • Instagram

Journal Policies

  • Peer Review Policy
  • Publication Ethics
  • Copyright Infringement
  • Refund & Cancellation Policy
  • Terms and Conditions
  • Privacy Policy

Journal Identifiers & Licenses

  • ISSN Verified: 2582-3930ISSN Verified: 2582-3930
  • Crossref DOI: 10.55041Crossref DOI: 10.55041
  • SJIF Ranking: 08.659SJIF Ranking: 08.659
  • Licensed under CC BY 4.0Licensed under CC BY 4.0
  • ISO Certification: 9001:2015ISO Certification: 9001:2015
  • Indexed & Cited In: Google ScholarIndexed & Cited In: Google Scholar

Contact Us

International Journal of Scientific Research in Engineering and Management (IJSREM)

#62/1, New No 7, 1st Cross, 2nd Main, Ganganagar, R T Nagar, Bangalore North, Bangalore, Karnataka, India – 560032
[email protected]+91 93911 67991
logos
logos
Working Hours

Mon – Sat: 9:00 AM – 6:00 PM

Sunday: 10:00 AM – 4:00 PM

+91 93911 67991

© 2026 IJSREM – International Journal of Scientific Research in Engineering and Management. This is an open-access journal. Articles are published under the Creative Commons Attribution (CC BY 4.0) License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DisclaimerPrivacy PolicyTerms and Conditions
ArticlePeer ReviewedOpen AccessPlagiarism Passed
Paper ID:IJSREM68702

Sentinel-X: A Category-Wise Audit of Grad-CAM Defect Localization on the VisA Benchmark

Stream/Area:

-

Subject:

-

Paper ID:

IJSREM68702

Volume:

10 |

11 |

Nov 2026

Pages:

1-8

Publication Frequency:

Monthly (12 issues per year)

DOI
Digital Object Identifier
10.55041/IJSREM68702

License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.

Share this article:

Authors

Niranjan

Niranjan

Artificial Intelligence And Data Science

Sri Krishna College Of Engineering and Technology

Download Author Certificate

Author Services & Downloads

Full ArticlesAcceptance Certificate

Abstract

An inspection system is far more useful when the operator can see why an image was flagged. We study Sentinel-X, a ResNet18 classifier for normal versus anomalous products, and ask how well its Grad-CAM explanations agree with real defect annotations on the VisA benchmark. As a classifier it performs well: on 1,629 test images it reaches an AUROC of 0.981 and an accuracy of 95.9%. The heatmaps are a different matter. On 180 anomalous test images (15 per category, all with readable, non-empty masks) the pooled pixel AUROC is 0.756 but the average precision is only 0.030, and at a fixed threshold of 0.5 the mean per-image IoU is 0.020 and the mean Dice coefficient 0.035. A threshold sweep peaks at 0.8 (IoU 0.029, Dice 0.050); because that value was picked on test data, we report it as exploratory. Overlap differs about 28-fold between categories and follows defect size: the thresholded map covers on average 23 times more area than the defect itself, and the categories with the largest defects overlap most. We conclude that Grad-CAM from this classifier shows roughly where the network looked, not where the defect is, and we list the protocol changes needed before such maps could support inspection decisions.

KEYWORDS

industrial inspection visual anomaly detection explainable artificial intelligence Grad-CAM ResNet18 VisA defect localization

Cite This Article

Niranjan, "Sentinel-X: A Category-Wise Audit of Grad-CAM Defect Localization on the VisA Benchmark," International Journal of Scientific Research in Engineering and Management, vol. 10, no. 11, pp. 1-8, Nov 2026. https://doi.org/10.55041/IJSREM68702

Full Article/Conclusion

Sentinel-X, a ResNet18 classifier, separates normal from anomalous VisA images well (AUROC 0.981, accuracy 95.9% on 1,629 test images), but its Grad-CAM maps localise defects poorly. On 180 anomalous test images with valid masks, pooled pixel AUROC is 0.756 while mean IoU and Dice at τ = 0.5 are only 0.020 and 0.035; overlap differs strongly between categories and follows defect size. Raising the threshold to 0.8 helps on average but hurts several categories, and it was selected on test data. The maps are a coarse guide to the model’s focus and should not be presented as defect segmentations. Future work should (i) choose thresholds on a validation split, per category if possible, and evaluate once on held-out data; (ii) include normal images in a fully specified protocol; (iii) compare against dedicated anomaly-localisation methods such as PatchCore and few-shot DINOv2-based detectors under identical splits and metrics; (iv) report confidence intervals over repeated runs; and (v) analyse overlap by defect size. These steps will show whether the patterns reported here are stable and whether an anomaly-map-based approach is needed for practical inspection.

References

[1]     Y. Zou, J. Jeong, L. Pemula, D. Zhang, and O. Dabeer, “SPot-the-Difference self-supervised pre-training for anomaly detection and segmentation,” in Proc. European Conf. Computer Vision (ECCV), 2022, pp. 392–408, doi: 10.1007/978-3-031-20056-4_23. [2]     K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770–778, doi: 10.1109/CVPR.2016.90. [3]     R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-CAM: Visual explanations from deep networks via gradient-based localization,” in Proc. IEEE Int. Conf. Computer Vision (ICCV), 2017, pp. 618–626, doi: 10.1109/ICCV.2017.74. [4]     P. Bergmann, M. Fauser, D. Sattlegger, and C. Steger, “MVTec AD – A comprehensive real-world dataset for unsupervised anomaly detection,” in Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), 2019, pp. 9584–9592, doi: 10.1109/CVPR.2019.00982. [5]     L. Heckler-Kram et al., “The MVTec AD 2 dataset: Advanced scenarios for unsupervised anomaly detection,” Int. J. Computer Vision, 2026, doi: 10.1007/s11263-026-02743-0. [6]     T. Defard, A. Setkov, A. Loesch, and R. Audigier, “PaDiM: A patch distribution modeling framework for anomaly detection and localization,” in Pattern Recognition. ICPR International Workshops and Challenges, 2021, doi: 10.1007/978-3-030-68799-1_35. [7]     K. Roth, L. Pemula, J. Zepeda, B. Schölkopf, T. Brox, and P. Gehler, “Towards total recall in industrial anomaly detection,” in Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), 2022, doi: 10.1109/CVPR52688.2022.01392. [8]     H. Deng and X. Li, “Anomaly detection via reverse distillation from one-class embedding,” in Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), 2022, doi: 10.1109/CVPR52688.2022.00951. [9]     K. Batzner, L. Heckler, and R. König, “EfficientAD: Accurate visual anomaly detection at millisecond-level latencies,” in Proc. IEEE/CVF Winter Conf. Applications of Computer Vision (WACV), 2024, doi: 10.1109/WACV57701.2024.00555. [10]   J. Jeong, Y. Zou, T. Kim, D. Zhang, A. Ravichandran, and O. Dabeer, “WinCLIP: Zero-/few-shot anomaly classification and segmentation,” in Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), 2023, doi: 10.1109/CVPR52729.2023.01878.

Editorial & Review Transparency

Publication Date

11/5/2026

Peer Review Type

Double-Blind Peer Review

Review Rounds: 2

Documents

Review Report (Confidential)
Copyright Form (Office)

Article Metrics & Impact

Article Views

0

Total views since publication

PDF Downloads

4

Total downloads since publication

Geographic Reach

5

Indexing, Archiving & Preservation

Google Scholar
Academia
ResearchGate
Crossref
SSRN
Semantic Scholar
Thomson Reuters
WorldCat
Mendeley
OpenAlex
EBSCO
Zenodo

Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.

Ethics, Compliance & Transparency

Plagiarism Screening

This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.

Conflict of Interest

The authors declare no conflict of interest related to this research work.

Funding Information

No funding information available for this research.

Similar & Recommended Articles

Similar Articles by Topic

Related research in the same subject area

Loading similar articles...