Contribution of Semantic segmentation to drone detection
conference paper
Drone detection in ground-based imaging remains particularly challenging when observed against cluttered backgrounds – such as urban areas or natural landscapes with vegetation and buildings – leading to increased false positives and missed detections for both classical computer vision and deep learning approaches. In contrast, detection over simple backgrounds like uniform sky is significantly more reliable. This paper investigates the contribution of semantic segmentation to improving drone detection performance by enabling context-aware processing.
TNO Identifier
1024361
Source title
12th International Symposium on Optronics in Defence & Security (OPTRO2026)
Pages
1-9
Files
To receive the publication files, please send an e-mail request to TNO Repository.