Mobiilirakendus seente tuvastamiseks
Files
Date
2019
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Bakalaureusetöö eesmärk on luua mobiilirakendus Androidile, mis suudab pildilt tuvastada Eesti metsades kasvavaid seeni ning tõsta seeläbi inimeste seeneteadlikkust. Töös kirjeldatakse seente määramisega seotud probleeme, tuues välja rakenduse vajalikkus. Samuti tutvustatakse probleemi lahendamiseks kasutatavate tehisnärvivõrkude tööpõhimõtet ja uuritakse sarnasel põhimõttel töötavaid olemasolevaid lahendusi. Töös kirjeldatakse nii tehisnärvivõrgu mudeli kui ka mobiilirakenduse loomise protsesse ning selleks kasutatud tehnoloogiliste vahendeid ja tööriistu. Töö lõpus analüüsitakse saadud tulemusi ning arutletakse rakenduse edasiarendamise võimaluste üle.
The aim of this bachelor thesis is to develop an Android mobile application, which identifies mushrooms in Estonian forests based on its picture and thus improves people’s knowledge of mushrooms. In this thesis problems related to mushroom identification and necessity for an application are described. Also the working principles of artificial neural networks that are used for image recognition are explained and some already existing solutions described. Furthermore, the processes of developing a mobile application and an artificial neural network model are described, as well as used technologies and tools for both of them. In the end of the thesis the results are analyzed and further development opportunities are discussed.
The aim of this bachelor thesis is to develop an Android mobile application, which identifies mushrooms in Estonian forests based on its picture and thus improves people’s knowledge of mushrooms. In this thesis problems related to mushroom identification and necessity for an application are described. Also the working principles of artificial neural networks that are used for image recognition are explained and some already existing solutions described. Furthermore, the processes of developing a mobile application and an artificial neural network model are described, as well as used technologies and tools for both of them. In the end of the thesis the results are analyzed and further development opportunities are discussed.