English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Light guide plates are essential optical components widely used in a diverserange of applications ranging from medical lighting fixtures to back-lit TVdisplays. An essential step in the manufacturing of light guide plates is thequality inspection of defects such as scratches, bright/dark spots, andimpurities. This is mainly done in industry through manual visual inspectionfor plate pattern irregularities, which is time-consuming and prone to humanerror and thus act as a significant barrier to high-throughput production.Advances in deep learning-driven computer vision has led to the exploration ofautomated visual quality inspection of light guide plates to improve inspectionconsistency, accuracy, and efficiency. However, given the cost constraints invisual inspection scenarios, the widespread adoption of deep learning-drivencomputer vision methods for inspecting light guide plates has been greatlylimited due to high computational requirements. In this study, we explore theutilization of machine-driven design exploration with computational and"best-practices" constraints as well as L$_1$ paired classification discrepancyloss to create LightDefectNet, a highly compact deep anti-aliased attentioncondenser neural network architecture tailored specifically for light guideplate surface defect detection in resource-constrained scenarios. Experimentsshow that LightDetectNet achieves a detection accuracy of $\sim$98.2% on theLGPSDD benchmark while having just 770K parameters ($\sim$33$\times$ and$\sim$6.9$\times$ lower than ResNet-50 and EfficientNet-B0, respectively) and$\sim$93M FLOPs ($\sim$88$\times$ and $\sim$8.4$\times$ lower than ResNet-50and EfficientNet-B0, respectively) and $\sim$8.8$\times$ faster inference speedthan EfficientNet-B0 on an embedded ARM processor.

LightDefectNet: A Highly Compact Deep Anti-Aliased Attention Condenser Neural Network Architecture for Light Guide Plate Surface Defect Detection