Matches in SemOpenAlex for { <https://semopenalex.org/work/W3193024002> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3193024002 endingPage "115673" @default.
- W3193024002 startingPage "115673" @default.
- W3193024002 abstract "In the printed circuit board (PCB) industry, cosmetic defect detection is an essential process to ensure product quality. However, existing PCB cosmetic defect detection approaches have a high false alarm rate, which lead to expensive labor costs of manual confirmation. To solve this problem, some traditional machine learning-based approaches have been proposed, but they just utilize hand-crafted features to build classifiers and thus are rough and sub-optimal. Recently, due to its powerful capability in automatic feature extraction, convolutional neural network (CNN) has been widely used in PCB cosmetic defect detection. However, few of them pay attention to the imbalanced class distribution as well as the different misclassification costs of real and pseudo defects, both of which are common problems in the PCB industry. To this end, in this study, we propose a novel model called cost-sensitive residual convolutional neural network (CS-ResNet) by adding a cost-sensitive adjustment layer in the standard ResNet. Specifically, we assign larger weights to minority real defects based on the class-imbalance degree and then optimize CS-ResNet by minimizing the weighted cross-entropy loss function. We conducted a series of experiments by comparing CS-ResNet with the standard ResNet, state-of-the-art CNN-based approach Auto-VRS and traditional machine learning-based approach HOG-SVM on a real-world PCB cosmetic defect dataset. Experimental results show that CS-ResNet achieves the highest S e n s i t i v i t y (0.89), G - m e a n (0.91) and the lowest misclassification costs. • Cosmetic defect detection is an essential process in the PCB industry. • We propose a cost-sensitive ResNet (CS-ResNet) for PCB cosmetic defect detection. • In CS-ResNet, we add a cost-sensitive adjustment layer to optimize the CS-ResNet. • Extensive empirical studies validate the effectiveness of the proposed CS-ResNet." @default.
- W3193024002 created "2021-08-16" @default.
- W3193024002 creator A5024120345 @default.
- W3193024002 creator A5040372217 @default.
- W3193024002 creator A5045568216 @default.
- W3193024002 date "2021-12-01" @default.
- W3193024002 modified "2023-10-16" @default.
- W3193024002 title "CS-ResNet: Cost-sensitive residual convolutional neural network for PCB cosmetic defect detection" @default.
- W3193024002 cites W2071785112 @default.
- W3193024002 cites W2125927049 @default.
- W3193024002 cites W2194775991 @default.
- W3193024002 cites W2558580397 @default.
- W3193024002 cites W2562319768 @default.
- W3193024002 cites W2618530766 @default.
- W3193024002 cites W2747355315 @default.
- W3193024002 cites W2767106145 @default.
- W3193024002 cites W2789904726 @default.
- W3193024002 cites W2791479474 @default.
- W3193024002 cites W2794869810 @default.
- W3193024002 cites W2800452261 @default.
- W3193024002 cites W2896357994 @default.
- W3193024002 cites W2901161402 @default.
- W3193024002 cites W2902198254 @default.
- W3193024002 cites W2913997398 @default.
- W3193024002 cites W2954875452 @default.
- W3193024002 cites W2963596856 @default.
- W3193024002 cites W2963613787 @default.
- W3193024002 cites W2964050365 @default.
- W3193024002 cites W2979661171 @default.
- W3193024002 cites W3013949960 @default.
- W3193024002 cites W3035802597 @default.
- W3193024002 cites W3082097505 @default.
- W3193024002 cites W3090069523 @default.
- W3193024002 cites W3103904664 @default.
- W3193024002 cites W3111485235 @default.
- W3193024002 cites W3113225827 @default.
- W3193024002 cites W3121306240 @default.
- W3193024002 cites W3125210124 @default.
- W3193024002 cites W3129420947 @default.
- W3193024002 cites W3153829905 @default.
- W3193024002 cites W728297 @default.
- W3193024002 doi "https://doi.org/10.1016/j.eswa.2021.115673" @default.
- W3193024002 hasPublicationYear "2021" @default.
- W3193024002 type Work @default.
- W3193024002 sameAs 3193024002 @default.
- W3193024002 citedByCount "48" @default.
- W3193024002 countsByYear W31930240022021 @default.
- W3193024002 countsByYear W31930240022022 @default.
- W3193024002 countsByYear W31930240022023 @default.
- W3193024002 crossrefType "journal-article" @default.
- W3193024002 hasAuthorship W3193024002A5024120345 @default.
- W3193024002 hasAuthorship W3193024002A5040372217 @default.
- W3193024002 hasAuthorship W3193024002A5045568216 @default.
- W3193024002 hasConcept C108583219 @default.
- W3193024002 hasConcept C11413529 @default.
- W3193024002 hasConcept C119857082 @default.
- W3193024002 hasConcept C153180895 @default.
- W3193024002 hasConcept C154945302 @default.
- W3193024002 hasConcept C155512373 @default.
- W3193024002 hasConcept C167981619 @default.
- W3193024002 hasConcept C2776836416 @default.
- W3193024002 hasConcept C2944601119 @default.
- W3193024002 hasConcept C41008148 @default.
- W3193024002 hasConcept C81363708 @default.
- W3193024002 hasConceptScore W3193024002C108583219 @default.
- W3193024002 hasConceptScore W3193024002C11413529 @default.
- W3193024002 hasConceptScore W3193024002C119857082 @default.
- W3193024002 hasConceptScore W3193024002C153180895 @default.
- W3193024002 hasConceptScore W3193024002C154945302 @default.
- W3193024002 hasConceptScore W3193024002C155512373 @default.
- W3193024002 hasConceptScore W3193024002C167981619 @default.
- W3193024002 hasConceptScore W3193024002C2776836416 @default.
- W3193024002 hasConceptScore W3193024002C2944601119 @default.
- W3193024002 hasConceptScore W3193024002C41008148 @default.
- W3193024002 hasConceptScore W3193024002C81363708 @default.
- W3193024002 hasFunder F4320321001 @default.
- W3193024002 hasFunder F4320328119 @default.
- W3193024002 hasLocation W31930240021 @default.
- W3193024002 hasOpenAccess W3193024002 @default.
- W3193024002 hasPrimaryLocation W31930240021 @default.
- W3193024002 hasRelatedWork W2337926734 @default.
- W3193024002 hasRelatedWork W2563602643 @default.
- W3193024002 hasRelatedWork W2732542196 @default.
- W3193024002 hasRelatedWork W2738221750 @default.
- W3193024002 hasRelatedWork W3021430260 @default.
- W3193024002 hasRelatedWork W3089733734 @default.
- W3193024002 hasRelatedWork W3136076031 @default.
- W3193024002 hasRelatedWork W3156786002 @default.
- W3193024002 hasRelatedWork W3207322857 @default.
- W3193024002 hasRelatedWork W564581980 @default.
- W3193024002 hasVolume "185" @default.
- W3193024002 isParatext "false" @default.
- W3193024002 isRetracted "false" @default.
- W3193024002 magId "3193024002" @default.
- W3193024002 workType "article" @default.