Підхід до автоматизації анотування зображень для навчання моделей штучного інтелекту

Autor:	O. Lisovychenko, A. Yakovlev
Rok vydání:	2020
Předmět:	Annotation Software Automatic image annotation Process (engineering) Computer science business.industry Key (cryptography) Artificial intelligence business computer.software_genre computer Natural language processing Task (project management)
Zdroj:	Адаптивні системи автоматичного управління. 1:32-40
ISSN:	2522-9575 1560-8956
DOI:	10.20535/1560-8956.36.2020.209755
Popis:	The article provides an analysis of image annotation process for artificialintelligence models within modern detection systems learning using modern tools for annotation. Software application requirements and parameters list has been formed for image annotation, which are sufficiently consistent with the annotation process. Provided charts displays key parameters for image annotation process in modern applications. Mass factor approach role importance reviewed in accordance with annotation task solving in modern recognition systems. Yoloanno application has been developed, whichincorporates all requirements for an annotation process: functional and timing, - and provides tools to solve the task, what was proven during the experiments. Obtained results could be used for image annotation task practical solution, as well as provided approaches could be used for new image annotation applications creation. Ref. 8, pic. 4
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::1898b73dbea5cacc91c85667b21bfebc https://doi.org/10.20535/1560-8956.36.2020.209755 Zobrazit plný text záznamu