ALGORITHMS FOR IMAGE CLASSIFICATION WITH A LARGE NUMBER OF OBJECT CATEGORIES |
| 4 | |
| 2013 |
| INFORMATION TECHNOLOGIES |
| scientific article | 004.932 | ||
| 225-230 | computer vision, machine learning, image classification, decision tree, random forest, semantic hierarchy |
| An image classification problem with a large number of object categories is considered. Classification algorithm modifications are proposed which are based on decision trees and their ensembles with the account of semantic hierarchies of objects. The results of numerical experiments presented show better prediction accuracy on a number of data sets. |
 |