Experimental design of ground target rotation detection in aerial images based on attention mechanism
[Objects]Object detection algorithms have shown promising results in drone aerial photography detection and dynamic tracking.However,challenges arise during drone aerial photography owing to occlusion,rotation,and scale changes in captured target instances.The traditional YOLO object detection algorithm is not ideal for detecting rotating targets effectively.[Methods]To address this issue,this article introduces an experimental ground target rotation detection algorithm.It incorporates the CBAM attention mechanism into the S2ANet algorithm,offering improved solutions to mixed background occlusion,target instance rotation and tilt,and multi-scale image changes in aerial imagery.Prior to experimentation,preprocessing is conducted to expand the DOTA dataset.The experimental dataset's sample images have been increased from 2806 to 4209 through image rotation and translation techniques.To facilitate the training and processing of deep learning algorithms,all images are standardized to a resolution of 1024×1024.Further cropping and enhancing of the expanded samples result in the creation of the DOTA_split dataset,containing a total of 32,877 samples.Following experimental preprocessing,the overall sample size increases to 11.72 times the original,significantly enhancing the accuracy of neural network training.Subsequently,the S2ANet network is chosen as the rotation object detection model,with RetinaNet serving as the baseline network model.The backbone network,ResNet,is paired with the neck network,FPN,while the head detection network consists of two FCN subnetworks.Additionally,the S2ANet network model incorporates the feature alignment network and the orientation detection module.Furthermore,a CBAM attention mechanism module is integrated into the feature fusion section of the S2ANet network model to enhance the effective utilization of input feature space and channels.Through the embedding of a CBAM attention mechanism network into each layer of the fused feature map obtained via upsampling in the FPN layer,the quality of the feature map is enhanced,leading to improved detection speed and accuracy of the object detection algorithm.The experimental plan is designed for a Linux system environment with Python 3.8,Pytorch 1.7.0,and CUDA 10.2 configurations.The NVIDIA GeForce RTX 2080TI graphics card is employed,and all experimental procedures are executed on a single GPU.Finally,the entire experiment is validated and analyzed,including an in-depth examination of the experimental results.[Results&Conclusions]By conducting a horizontal comparison with the traditional object detection algorithm YOLOv5,the experimental findings demonstrate that the ground object rotation detection algorithm proposed in this paper,based on attention mechanisms,effectively enhances the handling of occlusion,rotation,and multi-scale changes in target objects captured in unmanned aerial vehicle images.Additionally,the detection frame rate is enhanced,with YOLOv5 achieving 16.7 frames per second(fps),while the improved S2ANet algorithm achieves 24 fps.Through vertical comparison of ablation experiments,the effectiveness of the CBAM attention mechanism module is verified.Compared with the original algorithm,the enhanced algorithm increases the overall detection performance by 1.3%,and the mAP value for detecting small targets in aerial photography is increased from 0.402 9 to 0.539 8.This represents a 34%performance increase,and the detection rate reaches 24 fps.
deep learningaerial remote sensingtilted object detectionattention mechanismexperimental design
NETL
NSTL
万方数据
