Recognition of Low-Altitude Targets Using Improved Canny Edge Detection

Low-altitude aerial target detection is crucial for military, civilian aviation, and unmanned aerial systems. This study enhances Canny’s edge detection algorithm to accurately identify drones and helicopters. Initial contours are positioned at Canny-detected edges and continuously optimized via a snake active contour model, adjusting local thresholds. The improved Canny algorithm effectively extracts target contours, enabling accurate recognition and localization. It is robust and adaptable to different environments, overcoming limitations of the traditional Canny method. This research provides an effective framework for integrating active contours with Canny edge detection for reliable low-altitude target delineation. Context information and machine learning could further optimize edge detection for complex real-world conditions. Overall, this study delivers a robust edge-detection solution tailored to aerial target tracking and analytics across diverse operating environments.