Classification and Fast Few-Shot Learning of Steel Surface Defects with Randomized Network
Quality inspection is inevitable in the steel industry so there are already benchmark datasets for the visual inspection of steel surface defects. In our work, we show, contrary to previous recent articles, that a generic state-of-art deep neural network is capable of almost-perfect classification o...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/8/3967 |
| _version_ | 1797437033462890496 |
|---|---|
| author | Amr M. Nagy László Czúni |
| author_facet | Amr M. Nagy László Czúni |
| author_sort | Amr M. Nagy |
| collection | DOAJ |
| description | Quality inspection is inevitable in the steel industry so there are already benchmark datasets for the visual inspection of steel surface defects. In our work, we show, contrary to previous recent articles, that a generic state-of-art deep neural network is capable of almost-perfect classification of defects of two popular benchmark datasets. However, in real-life applications new types of errors can always appear, thus incremental learning, based on very few example shots, is challenging. In our article, we address the problems of the low number of available shots of new classes, the catastrophic forgetting of known information when tuning for new artifacts, and the long training time required for re-training or fine-tuning existing models. In the proposed new architecture we combine EfficientNet deep neural networks with randomized classifiers to aim for an efficient solution for these demanding problems. The classification outperforms all other known approaches, with an accuracy 100% or almost 100%, on the two datasets with the off-the-shelf network. The proposed few-shot learning approach shows considerably higher accuracy at a low number of shots than the different methods under testing, while its speed is significantly (at least 10 times) higher than its competitors. According to these results, the classification and few-shot learning of steel surface defects can be solved more efficiently than was possible before. |
| first_indexed | 2024-03-09T11:12:08Z |
| format | Article |
| id | doaj.art-89c596c291744402b389880a85469de8 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T11:12:08Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-89c596c291744402b389880a85469de82023-12-01T00:42:34ZengMDPI AGApplied Sciences2076-34172022-04-01128396710.3390/app12083967Classification and Fast Few-Shot Learning of Steel Surface Defects with Randomized NetworkAmr M. Nagy0László Czúni1Faculty of Information Technology, University of Pannonia, Egyetem u. 10, 8200 Veszprém, HungaryFaculty of Information Technology, University of Pannonia, Egyetem u. 10, 8200 Veszprém, HungaryQuality inspection is inevitable in the steel industry so there are already benchmark datasets for the visual inspection of steel surface defects. In our work, we show, contrary to previous recent articles, that a generic state-of-art deep neural network is capable of almost-perfect classification of defects of two popular benchmark datasets. However, in real-life applications new types of errors can always appear, thus incremental learning, based on very few example shots, is challenging. In our article, we address the problems of the low number of available shots of new classes, the catastrophic forgetting of known information when tuning for new artifacts, and the long training time required for re-training or fine-tuning existing models. In the proposed new architecture we combine EfficientNet deep neural networks with randomized classifiers to aim for an efficient solution for these demanding problems. The classification outperforms all other known approaches, with an accuracy 100% or almost 100%, on the two datasets with the off-the-shelf network. The proposed few-shot learning approach shows considerably higher accuracy at a low number of shots than the different methods under testing, while its speed is significantly (at least 10 times) higher than its competitors. According to these results, the classification and few-shot learning of steel surface defects can be solved more efficiently than was possible before.https://www.mdpi.com/2076-3417/12/8/3967steel surface defectsvisual inspectiondeep learningfew-shot learningEfficientNetrandomized neural network |
| spellingShingle | Amr M. Nagy László Czúni Classification and Fast Few-Shot Learning of Steel Surface Defects with Randomized Network Applied Sciences steel surface defects visual inspection deep learning few-shot learning EfficientNet randomized neural network |
| title | Classification and Fast Few-Shot Learning of Steel Surface Defects with Randomized Network |
| title_full | Classification and Fast Few-Shot Learning of Steel Surface Defects with Randomized Network |
| title_fullStr | Classification and Fast Few-Shot Learning of Steel Surface Defects with Randomized Network |
| title_full_unstemmed | Classification and Fast Few-Shot Learning of Steel Surface Defects with Randomized Network |
| title_short | Classification and Fast Few-Shot Learning of Steel Surface Defects with Randomized Network |
| title_sort | classification and fast few shot learning of steel surface defects with randomized network |
| topic | steel surface defects visual inspection deep learning few-shot learning EfficientNet randomized neural network |
| url | https://www.mdpi.com/2076-3417/12/8/3967 |
| work_keys_str_mv | AT amrmnagy classificationandfastfewshotlearningofsteelsurfacedefectswithrandomizednetwork AT laszloczuni classificationandfastfewshotlearningofsteelsurfacedefectswithrandomizednetwork |