Facilitating Air Transportation of Agricultural Systems via Intelligent Runway Detection

Abstract

Intelligent runway detection technology is crucial for the development of low-carbon, smart agricultural systems pertaining to the air transportation of agricultural products. Accurate detection of the location and orientation of the runway can effectively assist in safe aircraft landings and avoid potential risks. However, existing runway detection methods struggle in foggy conditions due to light scattering, causing blurry images and obscuring runway details, resulting in poor detection performance. Towards this issue, this paper proposes an adaptive image-based runway boundary detection method by combining image processing and filter prediction to enhance images automatically. It leverages runway symmetry to enhance feature maps and global-local information fusion. A shape loss function based on the runway's parallel boundaries is also introduced. These developments finally endow the proposed method with robustness towards foggy conditions. Experimental results demonstrate the method's effectiveness, achieving an average IoU of 73.58$\%$ on internal datasets, surpassing other advanced methods.