Analyzing Remote Sensing Radar Images using Machine Learning Based Object Recognition
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Object detection is challenging due to unpredictable object orientations, complex backgrounds, and a variety of object scales, and that why object detection in remote sensing photography is difficult. Conventional horizontal bounding box (CHBB) algorithms have difficulties in providing precise motion and localization data in these scenarios. This study presents an innovative approach to remote sensing image object recognition with the YOLO technique. By converting angle regression into a classification problem using a binary encoding methodology, the method eliminates border disconnectedness probem and guarantees quick prediction. Evaluations of the publicly available DOTA dataset and ablation tests confirm the effectiveness of the suggested approach. This approach surpassing leading techniques in detection accuracy and precise orientation prediction. In order to avoid border discontinuity problems and provide fast prediction, the method uses a binary encoding methodology to transform angle regression into a classification problem. The efficiency of the proposed method is confirmed by ablation tests and evaluations of the publicly accessible DOTA dataset. [5]
Fields of Study & Subjects
Advanced Neural Network Applications
Advanced SAR Imaging Techniques
Remote-Sensing Image Classification
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 28 March 2026
- Language
- EN