Skip navigation
Please use this identifier to cite or link to this item: https://libeldoc.bsuir.by/handle/123456789/45849
Full metadata record
DC FieldValueLanguage
dc.contributor.authorMikhailov, V.-
dc.contributor.authorSobolevskii, V.-
dc.date.accessioned2021-11-08T07:50:03Z-
dc.date.available2021-11-08T07:50:03Z-
dc.date.issued2021-
dc.identifier.citationMazouka, D. Reindeer Recognition and Counting System Based on Aerial Images and Convolutional Neural Networks / Mikhailov V., Sobolevskii V. // Pattern Recognition and Information Processing (PRIP'2021) = Распознавание образов и обработка информации (2021) : Proceedings of the 15th International Conference, 21–24 Sept. 2021, Minsk, Belarus / United Institute of Informatics Problems of the National Academy of Sciences of Belarus. – Minsk, 2021. – P. 89–91.ru_RU
dc.identifier.urihttps://libeldoc.bsuir.by/handle/123456789/45849-
dc.description.abstractDescribed animal recognition and counting system based on convolutional neural networks with MRCNN architecture. Initial training of the network is performed using a basic array of MS COCO images, and additional training is performed using an array of aerial photographs of reindeer herds. A webinterface of the system has been developed. The error of counting reindeer in the image from the verification sample is about 13%.ru_RU
dc.language.isoenru_RU
dc.publisherUIIP NASBru_RU
dc.subjectматериалы конференцийru_RU
dc.subjectconference proceedingsru_RU
dc.subjectreindeerru_RU
dc.subjectrecognitionru_RU
dc.subjectconvolutional neural networksru_RU
dc.subjectaerial imageryru_RU
dc.titleReindeer Recognition and Counting System Based on Aerial Images and Convolutional Neural Networksru_RU
dc.typeСтатьяru_RU
Appears in Collections:Pattern Recognition and Information Processing (PRIP'2021) = Распознавание образов и обработка информации (2021)

Files in This Item:
File Description SizeFormat 
Mikhailov_Reindeer.pdf1.02 MBAdobe PDFView/Open
Show simple item record Google Scholar

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.