Метод ідентифікації рухомих об’єктів на основі аналізу комбінованих відеопотоків
Obtaining accurate and complete information about the location of certain moving objects requires a large number of aerial surveying tools, because each of these tools has limitations in accuracy and visibility. The lack of accuracy is explained by the fact that shooting from a single angle provides...
Gespeichert in:
| Datum: | 2025 |
|---|---|
| Hauptverfasser: | , , |
| Format: | Artikel |
| Sprache: | Ukrainian |
| Veröffentlicht: |
V.M. Glushkov Institute of Cybernetics of NAS of Ukraine
2025
|
| Schlagworte: | |
| Online Zugang: | https://jais.net.ua/index.php/files/article/view/517 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Назва журналу: | Problems of Control and Informatics |
Institution
Problems of Control and Informatics| Zusammenfassung: | Obtaining accurate and complete information about the location of certain moving objects requires a large number of aerial surveying tools, because each of these tools has limitations in accuracy and visibility. The lack of accuracy is explained by the fact that shooting from a single angle provides limited information about the object. In this case, the angle may be unsuccessful, and the distance to the object is too great. One reason for the lack of completeness of information may be that the objects are distributed over an area that none of the available aerial photography tools can cover. Another reason may be the temporary disappearance of objects from the camera’s field of view due to obstacles. Common reasons for inaccurate or incomplete information include: masking of the object by the environment, insufficient contrast of the object with the background, and poor visibility due to insufficient lighting or fog. Therefore, a tool is needed for analyzing multi-camera survey data that allows you to increase the accuracy and completeness of the information obtained compared to traditional approaches using a single camera. In this study, a method for object identification based on the analysis of combined video streams is developed. The key element of the method is the fusion algorithm, the result of which is a set of identified objects. It is proven that the execution time of the algorithm linearly depends on the size of the input data. The presented method can be integrated into video surveillance information systems with many cameras for the classification and tracking of vehicles or other moving objects. The proposed method can find wide application in civilian areas: vehicle recognition in an urban environment, road traffic management, search and rescue operations, environmental monitoring, and security control of critical infrastructure. |
|---|