Damaged Airport Runway Extraction Based on Line and Corner Constraints

The incompleteness of damaged airport runway areas is a significant challenge to the existing airport runway extraction methods. In this paper, we propose a damaged airport runway extraction method that combines line with corner constraints. In the proposed method, we first extract lines and corners from the continuous edges of the region of interest. Then, based on line and corner constraints, we identify the rough location of each runway by calculating its central axis parameters and generate candidate runway areas by calculating and connecting the runway vertex coordinates. Further, the candidate runway areas are selected as final results that satisfy the runway length constraint. In 6 typical airport satellite images and 3 post-damage images, the average integrity of runway extraction is more than 95%, and the average quality of runway extraction is nearly 93%. The experimental results show that the proposed method performs well both on damaged and undamaged airport runway areas. Compared with the state-of-the-art method, the proposed method is more effective and efficient.