A YOLO-based lightweight aerial small target detection method, system, device and storage medium

Abstract

The application discloses a kind of YOLO-based light-weight aerial small target detection method, system, equipment and storage medium based on, it is related to computer vision and unmanned aerial vehicle remote sensing detection technical field, method includes: obtaining aerial image and uniform size, constructs light-weight detection network;Using the C2f module based on partial convolution and efficient multi-scale attention improvement extracts multi-scale feature;Introduce the feature enhancement module of frequency domain and spatial domain cooperation carries out feature fusion;Task dynamic alignment detection head output detection result based on shared mechanism is used;The application can significantly improve inference speed and reduce power consumption while ensuring high recall of aerial small target, realize real-time, accurate, low-power stable detection of unmanned aerial vehicle airborne platform.

Images