An Optimized YOLOv5s-rd Framework for Efficient Target Detection in Remote Sensing Images
Abstract
Remote sensing image object detection methods are crucial for applications related to aircraft, ships, vehicles, and buildings. Traditional methods, relying on manually designed features, suffer from high computational complexity, leading to low detection efficiency and stability. In response, we present an enhanced remote sensing image object detection approach, YOLOv5s - rd, which is built upon an optimized YOLOv5s.Our method integrates structural enhancements, refined loss functions, and advanced data augmentation strategies. These improvements include optimizing the number and orientation of rotation anchors to better handle target scale diversity and rotation changes in remote sensing images. We also adjust Gaussian distribution parameters, which is beneficial for dealing with the challenges of complex backgrounds. Additionally, we calibrate the weights of the weak - supervision branch, considering the fact that the number of objects in remote sensing images is usually small, aiming to improve the model's performance with limited labeled data. We conduct experiments on multiple public datasets: DOTA, HRSC2016, UCAS - AOD, and Northwest University VHR - 10. The results demonstrate that YOLOv5s - rd outperforms traditional and existing deep - learning methods in detection performance. Specifically, on the DOTA dataset, it achieves a mean average precision (mAP) of 80.4% and 41.2 FPS; on the UCAS - AOD dataset, 96.7% mAP and 40.3 FPS; on the HRSC2016 dataset, 93.2% mAP and 38.7 FPS; and on the Northwest University VHR - 10 dataset, 95.2% mAP and 39.7 FPS. Moreover, its computational complexity (FLOPs) is only 11.0B, surpassing most of the compared methods. By combining these novel optimizations, our YOLOv5s - rd not only enhances the robustness and effectiveness of the detection model but also significantly improves performance and reliability compared to existing methods, providing a new solution for remote sensing image object detection.
Full Text:
PDFReferences
Bi FK, Kong LZ, Feng ST, Han JH, Bian MM, Li Y. Refined Regression Detector for Multiclass-Oriented Target in Optical Remote Sensing Images. Journal of Applied Remote Sensing, 2023, 17: 15. DOI:10.1117/1.Jrs.17.026501.
Chen CC, Zeng WM, Zhang XL. HFPNet: Super Feature Aggregation Pyramid Network for Maritime Remote Sensing Small-Object Detection. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023, 16: 5973-5989. DOI:10.1109/jstars.2023.3286483.
Chen CY, Gong WG, Chen YL, Li WH. Object Detection in Remote Sensing Images Based on a Scene-Contextual Feature Pyramid Network. Remote Sensing, 2019, 11, 339. DOI:10.3390/rs11030339.
Chen H, Zhang LB, Ma J, Zhang J. Target Heat-Map Network: An End-to-End Deep Network for Target Detection in Remote Sensing Images. Neurocomputing, 2019, 331: 375-387. DOI:10.1016/j.neucom.2018.11.044.
Chen HB, Jiang S, He GH, Zhang BY, Yu H. TEANS: A Target Enhancement and Attenuated Nonmaximum Suppression Object Detector for Remote Sensing Images. IEEE Geoscience and Remote Sensing Letters, 2021, 18: 632-636. DOI:10.1109/lgrs.2020.2983070.
Chen LQ, Shi WX, Fan C, Zou L, Deng DX. A Novel Coarse-to-Fine Method of Ship Detection in Optical Remote Sensing Images Based on a Deep Residual Dense Network. Remote Sensing, 2020, 12: 21. DOI:10.3390/rs12193115.
Chen ST, Wang HJ, Mukherjee M, Xu X. Collaborative Learning-Based Network for Weakly Supervised Remote Sensing Object Detection. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023, 16: 7907-7918. DOI:10.1109/jstars.2022.3223845.
Chen T, Lu ZY, Yang Y, Zhang YX, Du B, Plaza A. A Siamese Network Based U-Net for Change Detection in High Resolution Remote Sensing Images. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, 15: 2357-2369. DOI:10.1109/jstars.2022.3157648.
Chen Y, Liu Q, Wang T, Wang B, Meng XL. Rotation-Invariant and Relation-Aware Cross-Domain Adaptation Object Detection Network for Optical Remote Sensing Images. Remote Sensing, 2021, 13: 22. DOI:10.3390/rs13214386.
Cheng B, Li ZZ, Wu QQ, Li B, Yang HH, Qing L, Qi B. Multi-Class Objects Detection Method in Remote Sensing Image Based on Direct Feedback Control for Convolutional Neural Network. IEEE Access, 2019, 7: 144691-144709. DOI:10.1109/access.2019.2943346.
Cheng B, Li ZZ, Xu BT, Dang CJ, Deng JQ. Target Detection in Remote Sensing Image Based on Object-and-Scene Context Constrained CNN. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 5. DOI:10.1109/lgrs.2021.3087597.
Cheng B, Li ZZ, Xu BT, Yao X, Ding ZQ, Qin TQ. Structured Object-Level Relational Reasoning CNN-Based Target Detection Algorithm in a Remote Sensing Image. Remote Sensing, 2021, 13: 26. DOI:10.3390/rs13020281.
Cheng G, Si YJ, Hong HL, Yao XW, Guo L. Cross-Scale Feature Fusion for Object Detection in Optical Remote Sensing Images. IEEE Geoscience and Remote Sensing Letters, 2021, 18: 431-435. DOI:10.1109/lgrs.2020.2975541.
Deng ZP, Sun H, Zhou SL, Zhao JP, Lei L, Zou HX. Multi-scale Object Detection in Remote Sensing Imagery with Convolutional Neural Networks. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 145(Part A): 3-22. DOI:10.1016/j.isprsjprs.2018.04.003.
Fang K, Ouyang JQ, Hu BW: Swin-HSTPS. Research on Target Detection Algorithms for Multi-Source High-Resolution Remote Sensing Images. Sensors, 2021, 21: 16. DOI:10.3390/s21238113.
Feng XX, Yao XW, Cheng G, Han JG, Han JW. SAENet: Self-Supervised Adversarial and Equivariant Network for Weakly Supervised Object Detection in Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-11. DOI:10.1109/tgrs.2021.3105575.
Feng YQ, Wang LW, Zhang MB. A Multi-Scale Target Detection Method for Optical Remote Sensing Images. Multimedia Tools and Applications, 2019, 78: 8751-8766. DOI:10.1007/s11042-018-6325-6.
Guo JX, Wang Z, Zhang SW. FESSD: Feature Enhancement Single Shot MultiBox Detector Algorithm for Remote Sensing Image Target Detection. Electronics, 2023, 12: 20. DOI:10.3390/electronics12040946.
Han QZ, Yin Q, Zheng X, Chen ZY. Remote Sensing Image Building Detection Method Based on Mask R-CNN. Complex & Intelligent Systems, 2022, 8: 1847-1855. DOI:10.1007/s40747-021-00322-z.
Han WX, Kuerban A, Yang YC, Huang ZT, Liu BH, Gao J: Multi-Vision Network for Accurate and Real-Time Small Object Detection in Optical Remote Sensing Images. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 5. DOI:10.1109/lgrs.2020.3044422.
Han XF, Jiang T, Zhao ZF, Lei ZT. Research on Remote Sensing Image Target Recognition Based on Deep Convolution Neural Network. International Journal of Pattern Recognition and Artificial Intelligence, 2020, 34: 20. DOI:10.1142/s0218001420540154.
Hou YJ, Shi G, Zhao YX, Wang F, Jiang X, Zhuang RJ, Mei YF, Ma XJ. R-YOLO: A YOLO-Based Method for Arbitrary-Oriented Target Detection in High-Resolution Remote Sensing Images. Sensors, 2022, 22: 16. DOI:10.3390/s22155716.
Hu Q, Li RS, Xu Y, Pan CF, Niu CY, Liu W. Toward Aircraft Detection and Fine-Grained Recognition from Remote Sensing Images. Journal of Applied Remote Sensing, 2022, 16: 18. DOI:10.1117/1.Jrs.16.024516.
Huang W, Li GY, Chen QQ, Ju M, Qu JT: CF2PN. A Cross-Scale Feature Fusion Pyramid Network Based Remote Sensing Target Detection. Remote Sensing, 2021, 13: 22. DOI:10.3390/rs13050847.
Huang W, Li GY, Jin BH, Chen QQ, Yin JR, Huang L. Scenario Context-Aware-Based Bidirectional Feature Pyramid Network for Remote Sensing Target Detection. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 5. DOI:10.1109/lgrs.2021.3135935.
Ji FC, Ming DP, Zeng BC, Yu JW, Qing YZ, Du TY, Zhang XY. Aircraft Detection in High Spatial Resolution Remote Sensing Images Combining Multi-Angle Features Driven and Majority Voting CNN. Remote Sensing, 2021, 13: 17. DOI:10.3390/rs13112207.
Jia HC, Guo Q, Chen J, Wang F, Wang HP, Xu F. Adaptive Component Discrimination Network for Airplane Detection in Remote Sensing Images. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 7699-7713. DOI:10.1109/jstars.2021.3098296.
Li B, Xie XY, Wei XX, Tang WT. Ship Detection and Classification from Optical Remote Sensing Images: A Survey. Chinese Journal of Aeronautics, 2021, 34: 145-163. DOI:10.1016/j.cja.2020.09.022.
Li BB, Zhou Y, Xie DH, Zheng LJ, Wu Y, Yue JB, Jiang SW. Stripe Noise Detection of High-Resolution Remote Sensing Images Using Deep Learning Method. Remote Sensing, 2022, 14: 28. DOI:10.3390/rs14040873.
Li CM, Gao HM, Yang Y, Qu XY, Yuan WJ. Segmentation Method of High-Resolution Remote Sensing Image for Fast Target Recognition. International Journal of Robotics & Automation, 2019, 34: 216-224. DOI:10.2316/j.2019.206-0114.
Li JX, Zhang ZL, Tian Y, Xu YP, Wen YH, Wang SC. Target-Guided Feature Super-Resolution for Vehicle Detection in Remote Sensing Images. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 5. DOI:10.1109/lgrs.2021.3112172.
Li QY, Chen YS, Zeng Y. Transformer with Transfer CNN for Remote-Sensing-Image Object Detection. Remote Sensing, 2022, 14: 21. DOI:10.3390/rs14040984.
Li RH, Shen Y. YOLOSR-IST: A Deep Learning Method for Small Target Detection in Infrared Remote Sensing Images Based on Super-Resolution and YOLO. Signal Processing, 2023, 208: 12. DOI:10.1016/j.sigpro.2023.108962.
Li S, Xu YL, Zhu MM, Ma SP, Tang H. Remote Sensing Airport Detection Based on End-to-End Deep Transferable Convolutional Neural Networks. IEEE Geoscience and Remote Sensing Letters, 2019, 16: 1640-1644. DOI:10.1109/lgrs.2019.2904076.
Li XG, Men FF, Lv SS, Jiang X, Pan MA, Ma Q, Yu HB. Vehicle Detection in Very-High-Resolution Remote Sensing Images Based on an Anchor-Free Detection Model with a More Precise Foveal Area. with a More Precise Foveal Area. ISPRS International Journal of Geo-Information, 2021, 10: 22. DOI:10.3390/ijgi10080549.
Li Y, Xu QZ, He ZF, Li W. Progressive Task-Based Universal Network for Raw Infrared Remote Sensing Imagery Ship Detection. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 13. DOI:10.1109/tgrs.2023.3275619.
Li YT, Wu ZH, Li L, Yang DN, Pang HF. Improved YOLOv3 Model for Vehicle Detection in High-Resolution Remote Sensing Images. Journal of Applied Remote Sensing, 2021, 15: 15. DOI:10.1117/1.Jrs.15.026505.
Li Z, Yuan JH, Li GX, Wang H, Li XC, Li D, Wang XH. RSI-YOLO: Object Detection Method for Remote Sensing Images Based on Improved YOLO. Sensors, 2023, 23: 21. DOI:10.3390/s23146414.
Li ZC, Yang RL, Cai WW, Xue YF, Hu YW, Li LJ. LLAM-MDCNet for Detecting Remote Sensing Images of Dead Tree Clusters. Remote Sensing, 2022, 14: 20. DOI:10.3390 /rs14153684.
Liu HJ, Du JX, Zhang Y, Zhang HB. Performance Analysis of Different DCNN Models in Remote Sensing Image Object Detection. Eurasip Journal on Image and Video Processing, 2022, 2022: 18. DOI:10.1186/s13640-022-00586-6.
DOI: https://doi.org/10.31449/inf.v49i18.7849
This work is licensed under a Creative Commons Attribution 3.0 License.