Full metadata record
| DC pole | Hodnota | Jazyk |
|---|---|---|
| dc.contributor.author | Rádli, Richárd | |
| dc.contributor.author | Czúni, László | |
| dc.contributor.editor | Skala, Václav | |
| dc.date.accessioned | 2021-08-31T09:53:23Z | |
| dc.date.available | 2021-08-31T09:53:23Z | |
| dc.date.issued | 2021 | |
| dc.identifier.citation | WSCG 2021: full papers proceedings: 29. International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, p. 181-188. | en |
| dc.identifier.isbn | 978-80-86943-34-3 | |
| dc.identifier.issn | 2464-4617 | |
| dc.identifier.issn | 2464–4625(CD/DVD) | |
| dc.identifier.uri | http://hdl.handle.net/11025/45023 | |
| dc.format | 8 s. | cs |
| dc.format.mimetype | application/pdf | |
| dc.language.iso | en | en |
| dc.publisher | Václav Skala - UNION Agency | cs |
| dc.rights | © Václav Skala - UNION Agency | cs |
| dc.subject | neuronová síť autoencoderu | cs |
| dc.subject | konvoluční neuronová síť | cs |
| dc.subject | detekce defektů | cs |
| dc.subject | detekce anomálií bez dozoru | cs |
| dc.title | About the Application of Autoencoders For Visual Defect Detection | en |
| dc.type | konferenční příspěvek | cs |
| dc.type | conferenceObject | en |
| dc.rights.access | openAccess | en |
| dc.type.version | publishedVersion | en |
| dc.description.abstract-translated | Visual defect detection is a key technology in modern industrial manufacturing systems. There are many possibleappearances of product defects, including distortions in color, shape, contamination, missing or superfluous parts.For the detection of those, besides traditional image processing techniques, convolutional neural networks basedmethods have also appeared to avoid the usage of hand-crafted features and to build more efficient detectionmechanisms. In our article we deal with autoencoder convolutional networks (AEs) which do not require examplesof defects for training. Unfortunately, the manual and/or trial-and-error design of AEs is still required to achievegood performance, since there are many unknown parameters of AEs which can greatly influence the detectionabilities. For our study we have chosen a well performing AE known as structural similarity AE (SSIM-AE),where the loss function and the comparison of the output with the input is implemented via the SSIM instead ofthe often used L1 or L2 norms. Investigating the performance of SSIM-AE on different data-sets, we found that itsperformance can be improved with modified convolutional structures without modifying the size of latent space.We also show that finding a model with low reconstruction error during training does not mean good detectionabilities and denoising AEs can increase efficiency. | en |
| dc.subject.translated | autoencoder neural network | en |
| dc.subject.translated | convolutional neural network | en |
| dc.subject.translated | defect detection | en |
| dc.subject.translated | unsupervised anomaly detection | en |
| dc.identifier.doi | https://doi.org/10.24132/CSRN.2021.3101.20 | |
| dc.type.status | Peer-reviewed | en |
| Vyskytuje se v kolekcích: | WSCG 2021: Full Papers Proceedings | |
Soubory připojené k záznamu:
| Soubor | Popis | Velikost | Formát | |
|---|---|---|---|---|
| I79.pdf | Plný text | 3,43 MB | Adobe PDF | Zobrazit/otevřít |
Použijte tento identifikátor k citaci nebo jako odkaz na tento záznam:
http://hdl.handle.net/11025/45023Všechny záznamy v DSpace jsou chráněny autorskými právy, všechna práva vyhrazena.