Institutional Repository of Vilnius University
Žiedadulkių atpažinimas ir identifikavimas iš skaitmeninių holografijos vaizdų /
Abstract
dc:descriptionThis bachelor thesis explores the application of deterministic and artificial neural network models to recognize and identify pollen particles from digital holography images. During the research, various pollen recognition and classification methods and algorithms were examined in the literature, and aſter applying them to the available data, the results were obtained that, in the absence of a marked dataset of nonbiological particles, the task of pollen recognition was best performed by the method of applying an ellipse to the contour of the particle, which recognizes the detected particles based on their shape. Aſter identifying the bestperforming models used in the analysed literature, these models were finetuned and their results were compared. It was found that the model based on ResNeSt101 architecture was the best suited for the classification of pollen particles.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Stonkus, Justas,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD210648062&prefLang=en_US