TY  - JOUR
AU  - Wang, Zhixi
AU  - He, Jinpeng
AU  - Chen, Huaixin
PY  - 2025
DA  - 2025/09/25
TI  - Lightweight Mura Defect Detection via Semantic Interscale Integration and Neighbor Fusion
JO  - Chinese Journal of Information Fusion
T2  - Chinese Journal of Information Fusion
JF  - Chinese Journal of Information Fusion
VL  - 2
IS  - 3
SP  - 237
EP  - 252
DO  - 10.62762/CJIF.2025.864944
UR  - https://www.icck.org/article/abs/CJIF.2025.864944
KW  - mura defect detection
KW  - lightweight network
KW  - interscale-fusion
KW  - neighbor fusion
AB  - Considering the large-area distribution, smooth brightness gradients, and blurred boundaries of Mura defects in real industrial scenarios, as well as the challenge of balancing accuracy and efficiency in existing methods, we propose a lightweight deep learning-based detection method for large-area Mura defects, termed SIFNet. The SIFNet adopts a classical encoder-decoder architecture with MobileNet-V2 as the backbone. Furthermore, we design a Graph-based Semantic Interscale-fusion Block (GSIB) that integrates the Semantic Fluid Aggregation Module (SFAM) and the Semantic Graph Inference Module (SGIM) to collaboratively extract high-level semantic features across multiple scales and establish abstract semantic representations for accurately localizing large-area Mura defects. Specifically, SFAM leverages a global attention mechanism to extract cross-spatial semantic flows, guiding the model to focus on potential brightness anomaly regions in the image and SGIM explicitly models the semantic relationships between multi-scale features using graph convolution, enhancing the model's ability to interpret regions with blurred boundaries and ambiguous structures. To further improve the model’s sensitivity to edges in regions with smooth brightness transitions, we introduce a NeighborFusion Edge Enhancement Module (NEEM). This module integrates depthwise separable convolutions with a spatial attention mechanism and introduces a CrossNorm-based feature alignment strategy to enhance spatial collaboration across feature layers. Additionally, an edge enhancement mechanism is employed to significantly improve the model’s ability to delineate blurred Mura defect boundaries, while keeping computational cost low and strengthening edge perception and representation. Extensive quantitative and qualitative experiments on three large-area Mura defect datasets constructed in this study demonstrate that SIFNet achieves excellent detection performance with only 3.92M parameters and 6.89 GFLOPs, striking an effective balance between accuracy and efficiency, and fully meeting the demands of industrial deployment.
SN  - 2998-3371
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{Wang2025Lightweigh,
  author = {Zhixi Wang and Jinpeng He and Huaixin Chen},
  title = {Lightweight Mura Defect Detection via Semantic Interscale Integration and Neighbor Fusion},
  journal = {Chinese Journal of Information Fusion},
  year = {2025},
  volume = {2},
  number = {3},
  pages = {237-252},
  doi = {10.62762/CJIF.2025.864944},
  url = {https://www.icck.org/article/abs/CJIF.2025.864944},
  abstract = {Considering the large-area distribution, smooth brightness gradients, and blurred boundaries of Mura defects in real industrial scenarios, as well as the challenge of balancing accuracy and efficiency in existing methods, we propose a lightweight deep learning-based detection method for large-area Mura defects, termed SIFNet. The SIFNet adopts a classical encoder-decoder architecture with MobileNet-V2 as the backbone. Furthermore, we design a Graph-based Semantic Interscale-fusion Block (GSIB) that integrates the Semantic Fluid Aggregation Module (SFAM) and the Semantic Graph Inference Module (SGIM) to collaboratively extract high-level semantic features across multiple scales and establish abstract semantic representations for accurately localizing large-area Mura defects. Specifically, SFAM leverages a global attention mechanism to extract cross-spatial semantic flows, guiding the model to focus on potential brightness anomaly regions in the image and SGIM explicitly models the semantic relationships between multi-scale features using graph convolution, enhancing the model's ability to interpret regions with blurred boundaries and ambiguous structures. To further improve the model’s sensitivity to edges in regions with smooth brightness transitions, we introduce a NeighborFusion Edge Enhancement Module (NEEM). This module integrates depthwise separable convolutions with a spatial attention mechanism and introduces a CrossNorm-based feature alignment strategy to enhance spatial collaboration across feature layers. Additionally, an edge enhancement mechanism is employed to significantly improve the model’s ability to delineate blurred Mura defect boundaries, while keeping computational cost low and strengthening edge perception and representation. Extensive quantitative and qualitative experiments on three large-area Mura defect datasets constructed in this study demonstrate that SIFNet achieves excellent detection performance with only 3.92M parameters and 6.89 GFLOPs, striking an effective balance between accuracy and efficiency, and fully meeting the demands of industrial deployment.},
  keywords = {mura defect detection, lightweight network, interscale-fusion, neighbor fusion},
  issn = {2998-3371},
  publisher = {Institute of Central Computation and Knowledge}
}