Angled high-resolution remote sensing image target detection method based on improved CenterNet

The invention discloses an angled high-resolution remote sensing image target detection method based on an improved CenterNet. The method comprises using HRNet as a backbone network under a CenterNetframework to obtain an improved CenterNet framework; giving a plurality of remote sensing target images as training samples, inputting the improved CenterNet framework for training to obtain a remotesensing image target detection framework; cutting the to-be-detected remote sensing image into a plurality of unit images with the same size; inputting each unit image into a remote sensing image target detection framework for target detection; and determining a target detection frame of each unit image, performing edge splicing on each unit image according to the target detection frame of each unit image, and determining a detection target of the to-be-detected remote sensing image so as to improve the precision of detecting a corresponding target in the to-be-detected remote sensing image. 本发明公开了一种基于改进CenterNet的有角度高分