| Presentation | 2023-03-02 nvestigation of Appearance Inspection Method Considering the Number of Corresponding Local Patches Katsuhisa Kitaguchi, Yohei Nishizaki, Mamoru Saito, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | There has been a great deal of research on appearance inspection using deep learning, which learns only from normal images. Recently, datasets incorporating not only structural anomalies such as scratches and stains, but also logical anomalies such as wrong number of parts and wrong combination of parts have been released, and there is a need to deal with such anomalies. In this study, we propose that PatchCore, a SOTA method for structural anomalies, can deal with logical anomalies by adding the consideration of consistency in the number of patches. Specifically, the histograms of the features of the test image and the good image are compared, and if there is a difference, it is considered as an anomaly. Experiments are conducted on a set of images with logical anomalies added to MVTec AD to verify the effectiveness of the proposed method. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Visual inspection / Anomaly Detection / Deep Learning |
| Paper # | PRMU2022-74,IBISML2022-81 |
| Date of Issue | 2023-02-23 (PRMU, IBISML) |
| Conference Information | |
| Committee | PRMU / IBISML / IPSJ-CVIM |
|---|---|
| Conference Date | 2023/3/2(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Future University Hakodate |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Seiichi Uchida(Kyushu Univ.) / Masashi Sugiyama(Univ. of Tokyo) |
| Vice Chair | Takuya Funatomi(NAIST) / Mitsuru Anpai(Denso IT Lab.) / Toshihiro Kamishima(AIST) / Koji Tsuda(Univ. of Tokyo) |
| Secretary | Takuya Funatomi(CyberAgent) / Mitsuru Anpai(Univ. of Tokyo) / Toshihiro Kamishima(NTT) / Koji Tsuda(Hokkaido Univ.) |
| Assistant | Nakamasa Inoue(Tokyo Inst. of Tech.) / Yasutomo Kawanishi(Riken) / Yoshinobu Kawahara(Osaka Univ.) / Taiji Suzuki(Tokyo Inst. of Tech.) |
| Paper Information | |
| Registration To | Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Information-Based Induction Sciences and Machine Learning / Special Interest Group on Computer Vision and Image Media |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | nvestigation of Appearance Inspection Method Considering the Number of Corresponding Local Patches |
| Sub Title (in English) | |
| Keyword(1) | Visual inspection |
| Keyword(2) | Anomaly Detection |
| Keyword(3) | Deep Learning |
| 1st Author's Name | Katsuhisa Kitaguchi |
| 1st Author's Affiliation | Osaka Research Institure of Industrial Science and Technology(ORIST) |
| 2nd Author's Name | Yohei Nishizaki |
| 2nd Author's Affiliation | Osaka Research Institure of Industrial Science and Technology(ORIST) |
| 3rd Author's Name | Mamoru Saito |
| 3rd Author's Affiliation | Osaka Research Institure of Industrial Science and Technology(ORIST) |
| Date | 2023-03-02 |
| Paper # | PRMU2022-74,IBISML2022-81 |
| Volume (vol) | vol.122 |
| Number (no) | PRMU-404,IBISML-405 |
| Page | pp.pp.88-92(PRMU), pp.88-92(IBISML), |
| #Pages | 5 |
| Date of Issue | 2023-02-23 (PRMU, IBISML) |