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

Surface defect detection plays an increasingly important role inmanufacturing industry to guarantee the product quality. Many deep learningmethods have been widely used in surface defect detection tasks, and have beenproven to perform well in defects classification and location. However, deeplearning-based detection methods often require plenty of data for training,which fail to apply to the real industrial scenarios since the distribution ofdefect categories is often imbalanced. In other words, common defect classeshave many samples but rare defect classes have extremely few samples, and it isdifficult for these methods to well detect rare defect classes. To solve theimbalanced distribution problem, in this paper we propose TL-SDD: a novelTransfer Learning-based method for Surface Defect Detection. First, we adopt atwo-phase training scheme to transfer the knowledge from common defect classesto rare defect classes. Second, we propose a novel Metric-based Surface DefectDetection (M-SDD) model. We design three modules for this model: (1) featureextraction module: containing feature fusion which combines high-level semanticinformation with low-level structural information. (2) feature reweightingmodule: transforming examples to a reweighting vector that indicates theimportance of features. (3) distance metric module: learning a metric space inwhich defects are classified by computing distances to representations of eachcategory. Finally, we validate the performance of our proposed method on a realdataset including surface defects of aluminum profiles. Compared to thebaseline methods, the performance of our proposed method has improved by up to11.98% for rare defect classes.

TL-SDD: A Transfer Learning-Based Method for Surface Defect Detection with Few Samples