Image-based anomaly detection in low-light industrial environments with feature enhancement
Industrial anomaly detection and localization are essential for maintaining product quality and safety in manufacturing. However, these tasks become significantly more challenging in low-light environments, where poor illumination introduces noise and reduces visibility, leading to degraded performa...
Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Results in Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590123025003901 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823864283222507520 |
|---|---|
| author | Dinh-Cuong Hoang Phan Xuan Tan Anh-Nhat Nguyen Son-Anh Bui Ta Huu Anh Duong Tuan-Minh Huynh Duc-Manh Nguyen Viet-Anh Trinh Quang-Huy Ha Nguyen Dinh Bao Long Duc-Thanh Tran Xuan-Tung Dinh Van-Hiep Duong Tran Thi Thuy Trang |
| author_facet | Dinh-Cuong Hoang Phan Xuan Tan Anh-Nhat Nguyen Son-Anh Bui Ta Huu Anh Duong Tuan-Minh Huynh Duc-Manh Nguyen Viet-Anh Trinh Quang-Huy Ha Nguyen Dinh Bao Long Duc-Thanh Tran Xuan-Tung Dinh Van-Hiep Duong Tran Thi Thuy Trang |
| author_sort | Dinh-Cuong Hoang |
| collection | DOAJ |
| description | Industrial anomaly detection and localization are essential for maintaining product quality and safety in manufacturing. However, these tasks become significantly more challenging in low-light environments, where poor illumination introduces noise and reduces visibility, leading to degraded performance of traditional methods. Existing approaches often rely on computationally expensive low-light image enhancement techniques or are limited by their sensitivity to noise and inability to adapt to varying illumination conditions, resulting in suboptimal anomaly detection performance. In this paper, we present DarkAD, a novel end-to-end approach specifically designed to address the difficulties of anomaly detection under low-light conditions. Our framework features a Dark-Aware Feature Adapter (DAFA), which enhances feature extraction by integrating two key modules: Frequency-based Feature Enhancement (FFE) and Illumination-aware Feature Enhancement (IFE). The FFE module suppresses high-frequency noise and amplifies structural details, while the IFE module adaptively boosts features from both well-lit and dimly illuminated regions, allowing the model to focus on critical areas without relying on computationally expensive image enhancement techniques. Extensive experiments on multiple industrial object categories demonstrate that DarkAD significantly outperforms state-of-the-art methods. Specifically, it achieves a mean image-level anomaly detection accuracy (I-AUROC) of 0.899 and a localization performance (AUPRO) of 0.862, surpassing the next-best method by 16.2% and 16.5%, respectively. Additionally, the framework maintains a real-time inference speed of 60 frames per second (FPS), making it well-suited for deployment in industrial settings. The ablation study further highlights the synergistic contributions of FFE and IFE, with their combined effect driving the model's superior performance across diverse and complex scenarios. |
| format | Article |
| id | doaj-art-e40cd725f4804afdbc8d8b30c1929701 |
| institution | Kabale University |
| issn | 2590-1230 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Results in Engineering |
| spelling | doaj-art-e40cd725f4804afdbc8d8b30c19297012025-02-09T05:01:07ZengElsevierResults in Engineering2590-12302025-03-0125104309Image-based anomaly detection in low-light industrial environments with feature enhancementDinh-Cuong Hoang0Phan Xuan Tan1Anh-Nhat Nguyen2Son-Anh Bui3Ta Huu Anh Duong4Tuan-Minh Huynh5Duc-Manh Nguyen6Viet-Anh Trinh7Quang-Huy Ha8Nguyen Dinh Bao Long9Duc-Thanh Tran10Xuan-Tung Dinh11Van-Hiep Duong12Tran Thi Thuy Trang13Greenwich Vietnam, FPT University, Hanoi, 10000, Viet NamCollege of Engineering, Shibaura Institute of Technology, Tokyo, 135-8548, Japan; Corresponding author.ICT Department, FPT University, Hanoi, 10000, Viet NamGreenwich Vietnam, FPT University, Hanoi, 10000, Viet NamGreenwich Vietnam, FPT University, Hanoi, 10000, Viet NamGreenwich Vietnam, FPT University, Hanoi, 10000, Viet NamGreenwich Vietnam, FPT University, Hanoi, 10000, Viet NamGreenwich Vietnam, FPT University, Hanoi, 10000, Viet NamGreenwich Vietnam, FPT University, Hanoi, 10000, Viet NamGreenwich Vietnam, FPT University, Hanoi, 10000, Viet NamICT Department, FPT University, Hanoi, 10000, Viet NamICT Department, FPT University, Hanoi, 10000, Viet NamICT Department, FPT University, Hanoi, 10000, Viet NamICT Department, FPT University, Hanoi, 10000, Viet NamIndustrial anomaly detection and localization are essential for maintaining product quality and safety in manufacturing. However, these tasks become significantly more challenging in low-light environments, where poor illumination introduces noise and reduces visibility, leading to degraded performance of traditional methods. Existing approaches often rely on computationally expensive low-light image enhancement techniques or are limited by their sensitivity to noise and inability to adapt to varying illumination conditions, resulting in suboptimal anomaly detection performance. In this paper, we present DarkAD, a novel end-to-end approach specifically designed to address the difficulties of anomaly detection under low-light conditions. Our framework features a Dark-Aware Feature Adapter (DAFA), which enhances feature extraction by integrating two key modules: Frequency-based Feature Enhancement (FFE) and Illumination-aware Feature Enhancement (IFE). The FFE module suppresses high-frequency noise and amplifies structural details, while the IFE module adaptively boosts features from both well-lit and dimly illuminated regions, allowing the model to focus on critical areas without relying on computationally expensive image enhancement techniques. Extensive experiments on multiple industrial object categories demonstrate that DarkAD significantly outperforms state-of-the-art methods. Specifically, it achieves a mean image-level anomaly detection accuracy (I-AUROC) of 0.899 and a localization performance (AUPRO) of 0.862, surpassing the next-best method by 16.2% and 16.5%, respectively. Additionally, the framework maintains a real-time inference speed of 60 frames per second (FPS), making it well-suited for deployment in industrial settings. The ablation study further highlights the synergistic contributions of FFE and IFE, with their combined effect driving the model's superior performance across diverse and complex scenarios.http://www.sciencedirect.com/science/article/pii/S2590123025003901Industrial anomaly detectionDefect recognitionSurface inspectionSurface defect localizationVisual inspectionLow-light environments |
| spellingShingle | Dinh-Cuong Hoang Phan Xuan Tan Anh-Nhat Nguyen Son-Anh Bui Ta Huu Anh Duong Tuan-Minh Huynh Duc-Manh Nguyen Viet-Anh Trinh Quang-Huy Ha Nguyen Dinh Bao Long Duc-Thanh Tran Xuan-Tung Dinh Van-Hiep Duong Tran Thi Thuy Trang Image-based anomaly detection in low-light industrial environments with feature enhancement Results in Engineering Industrial anomaly detection Defect recognition Surface inspection Surface defect localization Visual inspection Low-light environments |
| title | Image-based anomaly detection in low-light industrial environments with feature enhancement |
| title_full | Image-based anomaly detection in low-light industrial environments with feature enhancement |
| title_fullStr | Image-based anomaly detection in low-light industrial environments with feature enhancement |
| title_full_unstemmed | Image-based anomaly detection in low-light industrial environments with feature enhancement |
| title_short | Image-based anomaly detection in low-light industrial environments with feature enhancement |
| title_sort | image based anomaly detection in low light industrial environments with feature enhancement |
| topic | Industrial anomaly detection Defect recognition Surface inspection Surface defect localization Visual inspection Low-light environments |
| url | http://www.sciencedirect.com/science/article/pii/S2590123025003901 |
| work_keys_str_mv | AT dinhcuonghoang imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT phanxuantan imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT anhnhatnguyen imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT sonanhbui imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT tahuuanhduong imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT tuanminhhuynh imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT ducmanhnguyen imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT vietanhtrinh imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT quanghuyha imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT nguyendinhbaolong imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT ducthanhtran imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT xuantungdinh imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT vanhiepduong imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement AT tranthithuytrang imagebasedanomalydetectioninlowlightindustrialenvironmentswithfeatureenhancement |