每年出版 12 期
ISSN 打印: 0040-2508
ISSN 在线: 1943-6009
Indexed in
IMAGE CLASSIFICATION METHODS IN THE SPACE OF DESCRIPTIONS IN THE FORM OF A SET OF THE KEY POINT DESCRIPTORS
摘要
There are discussed the issues of development of the classification rules in the structural methods of image recognition in the case that descriptions of the objects are represented as the sets of the key point descriptors. It is suggested the classification technology based on the direct formation of the class centers for the visual objects in the space of the values of descriptors. At creation of the centers there are applied the binary data analysis and training using the Kohonen neural network based on the set of descriptors from the application image database. There are provided the computer experiment results confirming effectiveness of the suggested methods of visual object classification.
-
Gorokhovatskyi Oleksii, Gorokhovatskyi Volodymyr, Peredrii Olena, Analysis of Application of Cluster Descriptions in Space of Characteristic Image Features, Data, 3, 4, 2018. Crossref
-
Gorokhovatskyi Volodymyr, Gadetska Svitlana, Gorokhovatskyi Oleksii, Ponomarenko Roman, Statistical Methods for Analyzing and Processing Data Components When Recognizing Visual Objects in the Space of Key Point Descriptors, in Data Stream Mining & Processing, 1158, 2020. Crossref