Segmentation of Rays in Wood Microscopy Images Using the U-Net Model

dc.contributor.authorErgun, Halime
dc.date.accessioned2024-02-23T14:30:01Z
dc.date.available2024-02-23T14:30:01Z
dc.date.issued2021
dc.departmentNEÜen_US
dc.description.abstractRays are an important anatomical feature in tree species identification. They are found in certain proportions in trees, which vary for each tree. In this study, the U-Net model is adopted for the first time to detect wood rays. A dataset is created with images taken from the wood database. The resolution of microscopic wood images in tangential section is 640x400. The input image for training is divided into 32x32 image blocks. Each pixel in the dataset is labeled as belonging to the ray or the background. Then, the dataset is increased by applying scale, rotation, salt-and-pepper noise, circular mean filter, and gauss filter. The U-Net network created for ray segmentation is trained using the Adam optimization algorithm. The experimental results show that the ray segmentation accuracy in testing is 96.3%.en_US
dc.identifier.doi10.15376/biores.16.1.721-728
dc.identifier.endpage728en_US
dc.identifier.issn1930-2126
dc.identifier.issue1en_US
dc.identifier.scopus2-s2.0-85121997524en_US
dc.identifier.scopusqualityQ3en_US
dc.identifier.startpage721en_US
dc.identifier.urihttps://doi.org/10.15376/biores.16.1.721-728
dc.identifier.urihttps://hdl.handle.net/20.500.12452/14953
dc.identifier.volume16en_US
dc.identifier.wosWOS:000616046900051en_US
dc.identifier.wosqualityQ2en_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.language.isoenen_US
dc.publisherNorth Carolina State Univ Dept Wood & Paper Scien_US
dc.relation.ispartofBioresourcesen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/openAccessen_US
dc.subjectImage Segmentationen_US
dc.subjectWood Raysen_US
dc.subjectU-Neten_US
dc.titleSegmentation of Rays in Wood Microscopy Images Using the U-Net Modelen_US
dc.typeArticleen_US

Dosyalar