| 949 | 3 | 668 |
| 下载次数 | 被引频次 | 阅读次数 |
针对遥感图像目标排列紧密、背景复杂和小目标众多导致检测精度低的问题，提出了一种基于改进实时检测Transformer(Real-Time Detection Transformer, RT-DETR)的遥感图像检测算法。将Mosaic9数据增强应用到遥感数据中，丰富训练数据中场景和目标的组合，增强模型对不同环境下目标的识别能力。在主干网络中添加卷积块注意力模块(Convolutional Block Attention Module, CBAM),增强复杂背景下目标的关注度和图像特征提取能力，在模型中额外添加一个针对小目标的检测层，使小目标的细节特征更加突出，提升模型对小目标的检测能力。在DSTD舰船遥感数据集和NWPU VHR-10多类别遥感数据集上的实验结果显示，改进后的算法在交并比(Intersection over Union, IoU)阈值为0.5时，平均精度均值(mean Average Precision, mAP)分别达到了94.9%和94.5%,较原始RT-DETR算法分别提升了1%和1.3%,体现了改进算法在遥感图像检测上的有效性和通用性。
Abstract：To address issues of low detection accuracy in remote sensing images due to tightly arranged objects, complex backgrounds, and a high number of small objects, an improved detection algorithm based on the Real-Time Detection Transformer(RT-DETR) is proposed. This algorithm incorporates Mosaic9 data augmentation into the remote sensing data, enriching the combination of scenes and objects in training data and enhancing the model's ability to recognize objects under different environmental conditions. A Convolutional Block Attention Module(CBAM) is added to the backbone network to increase focus on objects against complex backgrounds and improve image feature extraction capabilities. Additionally, a specialized detection layer for small objects is integrated into the model, highlighting the detail features of small objects and enhancing small object detection capabilities. Experimental results on the DSTD maritime remote sensing dataset and the NWPU VHR-10 multi-category remote sensing dataset demonstrate that the improved algorithm achieves mean Average Precision(mAP) values of 94.9% and 94.5% respectively at an Intersection over Union(IoU) threshold of 0.5, showing improvements of 1% and 1.3% respectively over the original RT-DETR algorithm. These results confirm the effectiveness and versatility of the improved algorithm in remote sensing image detection.
[1] 王文韬，何小海，张豫堃，等.基于改进STANet的遥感图像变化检测算法[J].无线电工程，2024,54(5):1226-1235.
[2] 程换新，矫立浩，骆晓玲，等.改进YOLOv8的遥感图像检测算法[J].无线电工程，2024,54(5):1155-1161.
[3] 余俊宇，刘孙俊，许桃.融合注意力机制的YOLOv7遥感小目标检测算法研究[J].计算机工程与应用，2023,59(20):167-175.
[4] GIRSHICK R,DONAHUE J,DARRELL T,et al.Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition.Columbus:IEEE,2014:580-587.
[5] GIRSHICK R.Fast R-CNN[C]// 2015 IEEE International Conference on Computer Vision (ICCV).Santiago:IEEE,2015:1440-1448.
[6] REN S Q,HE K M,GIRSHICK R,et al.Faster R-CNN:Towards Real-time Object Detection with Region Proposal Networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(6):1137-1149.
[7] HE K M,GEORGIA G,PIOTR D,et al.Mask R-CNN[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2020,42(2):386-397.
[8] REDMON J,DIVVALA S,GIRSHICK R,et al.You Only Look Once:Unified,Real-time Object Detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE,2016:779-788.
[9] REDMON J,FARHADI A.YOLOv3:An Incremental Improvement[EB/OL].(2018-04-08) [2024-05-04].https://arxiv.org/abs/1804.02767.
[10] BOCHKOVSKIY A,WANG C Y,LIAO H Y M.YOLOv4:Optimal Speed and Accuracy of Object Detection[EB/OL].(2020-04-23) [2024-05-04].https://arxiv.org/abs/2004.10934.
[11] REDMON J,FARHADI A.YOLO9000:Better,Faster,Stronger[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition.Honolulu:IEEE,2017:7263-7271.
[12] LIU W,ANGUELOV D,ERHAN D,et al.SSD:Single Shot Multibox Detector[C]// 14th European Conference on Computer Vision.Amsterdam:Springer,2016:21-37.
[13] LIN T Y,GOYAL P,GIRSHICK R,et al.Focal Loss for Dense Object Detection[C]//2017 IEEE International Conference on Computer Vision.Venice:IEEE,2017:2980-2988.
[14] CARION N,MASSA F,SYNNAEVE G,et al.End-to-End Object Detection with Transformers[C]//16th European Conference on Computer Vision.Glasgow:Springer,2020:213-229.
[15] ZHAO Y A,LV W Y,XU S L,et al.DETRs Beat YOLOs on Real-time Object Detection[EB/OL].(2023-04-17)[2024-04-14].https://arxiv.org/abs/2304.08069.
[16] 金睿蛟，王堃，刘敏豪，等.基于DETR和改进去噪训练的光学遥感图像多尺度旋转目标检测(特邀)[J].激光与光电子学进展，2024,61(2):356-366.
[17] 刘敏豪，王堃，金睿蛟，等.基于改进RoI Transformer的遥感图像多尺度旋转目标检测[J].应用光学，2023,44(5):1010-1021.
[18] 周孟然，王澳.基于DETR的轻量级遥感图像目标检测算法[J/OL].(2024-03-29) [2024-05-04].http://kns.cnki.net/kcms/detail/50.1155.N.20240328.1703.004.html.
[19] ZHOU Y,CHEN S L,ZHAO J Q,et al.CLT-Det:Correlation Learning Based on Transformer for Detecting Dense Objects in Remote Sensing Images[J].IEEE Transactions on Geoscience and Remote Sensing,2022,60:4708915.
[20] WOO S H,PARK J C,LEE J Y,et al.CBAM:Convolutional Block Attention Module[C]//15th European Conference on Computer Vision (ECCV).Munich:Springer,2018:3-19.
[21] XU L Z,YAN W,JI J S.The Research of a Novel WOG-YOLO Algorithm for Autonomous Driving Object Detection[J].Scientific Reports,2023,13:3699.
[22] HE K M,ZHANG X Y,REN S Q,et al.Deep Residual Learning for Image Recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE,2016:770-778.
[23] CARON M,TOUVRON H,MISRA I,et al.Emerging Properties in Self-supervised Vision Transformers[C]//2021 IEEE/CVF International Conference on Computer Vision.Montreal:IEEE,2021:9650-9660.
[24] LECUN Y,BOSER B,DENKER J S,et al.Backpropagation Applied to Handwritten Zip Code Recognition[J].Neural Computation,1989,1(4):541-551.
[25] 何维娟，江涛，王林飞，等.舰船遥感图像数据集DSTD的构建研究[J].计算机技术与发展，2022,32(7):82-86.
[26] LI K,WAN G,CHENG G,et al.Object Detection in Optical Remote Sensing Images:A Survey and a New Benchmark[J].ISPRS Journal of Photogrammetry and Remote Sensing,2020,159:296-307.
基本信息:
中图分类号:TP751
引用信息:
[1]白金燕,江涛,魏玉梅,等.基于改进RT-DETR的遥感图像检测算法[J].无线电工程,2025,55(02):334-342.
基金信息:
国家自然科学基金(61363022)~~
2024-09-25
2024-09-25
2024-09-25