eprintid: 16415 rev_number: 2 eprint_status: archive userid: 1 dir: disk0/00/01/64/15 datestamp: 2023-12-19 03:22:56 lastmod: 2023-12-19 03:22:56 status_changed: 2023-12-19 03:06:12 type: article metadata_visibility: show creators_name: Anwer, A. creators_name: Ainouz, S. creators_name: Saad, M.N.M. creators_name: Ali, S.S.A. creators_name: Meriaudeau, F. title: SpecSeg Network for Specular Highlight Detection and Segmentation in Real-World Images ispublished: pub keywords: Large dataset; Pixels; Textures, Condition; False positive; Highlights detection; Images segmentations; Modern techniques; Multiple objects; Network-based; Non trivial problems; Real-world image; Specular highlight, Image segmentation, article; false positive result; image segmentation note: cited By 4 abstract: Specular highlights detection and removal in images is a fundamental yet non-trivial problem of interest. Most modern techniques proposed are inadequate at dealing with real-world images taken under uncontrolled conditions with the presence of complex textures, multiple objects, and bright colours, resulting in reduced accuracy and false positives. To detect specular pixels in a wide variety of real-world images independent of the number, colour, or type of illuminating source, we propose an efficient Specular Segmentation (SpecSeg) network based on the U-net architecture that is expeditious to train on nominal-sized datasets. The proposed network can detect pixels strongly affected by specular highlights with a high degree of precision, as shown by comparison with the state-of-the-art methods. The technique proposed is trained on publicly available datasets and tested using a large selection of real-world images with highly encouraging results. © 2022 by the authors. date: 2022 publisher: MDPI official_url: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85137597770&doi=10.3390%2fs22176552&partnerID=40&md5=8b08d60fc30f4e00a590165488e0bc82 id_number: 10.3390/s22176552 full_text_status: none publication: Sensors volume: 22 number: 17 refereed: TRUE issn: 14248220 citation: Anwer, A. and Ainouz, S. and Saad, M.N.M. and Ali, S.S.A. and Meriaudeau, F. (2022) SpecSeg Network for Specular Highlight Detection and Segmentation in Real-World Images. Sensors, 22 (17). ISSN 14248220