Aircraft targets in remote sensing images have the characteristics of small scale,uneven distribution,and complex back-ground,and the detection effect of existing methods is not promising.Aiming at this problem,a remote sensing image aircraft detec-tion model based on dense connection mechanism and spatial attention mechanism is constructed.First,the convolution kernel of the dense connection mechanism and the spatial attention module are used to form a feature extraction network for feature capture,and then the feature fusion network based on shallow features is used to obtain four-scale fusion feature maps for detection output.Models are trained and tested on mixed datasets and compared with current mainstream detection models.The results show that the detection model constructed in this paper is significantly better than the comparison model in terms of accuracy,at the same time shows good generalization ability in a variety of complex environments,and can meet the real-time requirements in the experimental environment.
NETL
NSTL
万方数据
