ABSTRACT
Space infrared sensors have become the "eyes" of space activities, there is currently no established identification method applied to the embedded deployment of space infrared sensor, despite it being a crucial device for aerospace detection due to the influence of deep learning target detection model embedded deployment difficulty, large sample demand, model parameter weight, low computing efficiency, and other factors. In this paper, the YOLOv5 detection model's network structure and activation function are lightened up. Next, the algorithm model's adaptive learning rate optimization algorithm is used to build the lightened-up YOLOv5 infrared image target recognition model. And the image samples were preprocessed with the intention of addressing the poor image quality brought on by background interference and multi-pose acquisition. Then, a method of sample amplification is suggested in light of the issues with the deep learning model. Consequently, the deep learning lightweight model applied in this study is one-third smaller than the conventional model, which dramatically increases target detection speed without significantly lowering detection rate. In addition to investigates ideas for the embedded deployment and recognition of infrared images in small sample space.
- Y XING, C WEI, L TANG, Development of au-tonomous sensing and control technology for extraterres-trial mobile exploration unmanned sys-tems[J].Aerospace Control and Application, 2021, 47(06): 1-8 (in Chinese).Google Scholar
- LIU Yinian, SUN Dexin, HAN Bo, Development of the visible short-wave infrared hyperspectral cam-era of Zuiyuan-1 02D satellite [J]. Spacecraft Engi-neering, 2020, 29(6): 85-92 (in Chinese).Google Scholar
- C PAN, L KANG, M LUO, Development Status and Application of Space Infrared Camera Optical Tech-nology[J]. Infrared Technology, 2022, 44(11): 1186-1194 (in Chinese).Google Scholar
- LI H G, YU R N, DING W R. Research development of small object tracking based onDLM[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(7): 024691(in Chinese).Google Scholar
- [5] SUN Y, Multisensor Fusion and Explicit Semantic Preserving-Based Deep Hashing for Cross-Modal Re mote Sensing Image Retrieval[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60(8): 1-14.Google Scholar
- L PU, X ZHANG.DLM based UAV vision object detection and tracking[J]. Journal of Beijing Uni-versity of Aeronautics and Astronautics, 2022, 48(05): 872-880 (in Chinese).Google Scholar
- X SU, R GUAN, Q WANG, Ice area and thick-ness detection method based on depth learning[J]. Acta Aeronautica et Astronautica Sinica, 2023, 1-9 (in Chi-nese).Google Scholar
- Q WANG, L WU, Y WANG, An infrared small target detection method based on key point[J]. Acta Aero-nautica et Astronautica Sinica, 2023, 44(10): 289-299 (in Chinese).Google Scholar
- Q ZHAO, Y WU, Y YUAN. Progress of ship detection and recognition methods in optical remote sensing imag-es[J]. Acta Aeronautica et Astronautica Sinica, 2023, 1-34 (in Chinese).Google Scholar
- HE K, ZHANG X, REN S, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015,37(9): 1904-1916.Google ScholarDigital Library
- GIRSHICK R, DONAHUE J, DARRELL T, Rich feature hierarchies for accurate object detection and semantic segmentation [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recogni-tion, Washington DC, United States, 2014:580587.Google Scholar
- GIRSHICK R. Fast r-cnn[C]//Proceedings of the IEEE international conference on computer vision. 2015: 1440-1448.Google Scholar
- REN S, HE K, GIRSHICK R, Faster r-cnn: To-wards real-time object detection with region proposal networks[C]//Advances in neural information pro-cessing systems. 2015: 91-99.Google Scholar
- LIU W, ANGUELOV D, ERHAN D, SSD: Single shot multibox detector[C]//European conference on computer vision. Springer, Cham, 2016:21-37.Google Scholar
- LIN T Y, GOYAL P, GIRSHICK R, Focal loss for dense object detection[C]//Proceedings of the IEEE In-ternational Conference on Computer Vision, Hawaii, USA, 2017:2980-2988.Google Scholar
- LAW H, DENG J. Corner Net: detecting objects as paired keypoints[J]. International Journal of Computer Vision, 2018, 75(6): 765-781Google Scholar
- DUAN K W, BAI S, XIE L X, Center Net: key-point triplets for object detection[C]. 2019 IEEE/CVF International Conference on Computer Vision, 2019: 6568-6577.Google Scholar
- REDMON J, DIVVALA S, GIRSHICK R, You only look once: Unified, real-time object detec-tion[C]//Proceedings of the IEEE conference on com-puter vision and pattern recognition.2016: 779-788.Google Scholar
- REMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of the IEEE conference on computer vision and pattern recognition.2017:7263-7271.Google Scholar
- REMON J, FARHADI A. YOLOv3:an incremental im-provement [J]. arXiv preprint arXiv:1804.02767, 2018.Google Scholar
- BOCHKOVSKIY A, WANG C Y, LIAO H. YOLOv4: Optimal Speed and Accuracy of Object Detection[J]. 2020.Google Scholar
- JIANG P Y, ERGU D J, LIU F Y, A review of Yolo algorithm developments[J]. Procedia Computer Science,2022,199:1066-1073.Google Scholar
- GE Z, LIU S T, WANG F, Yolox: Exceeding yolo series in 2021[J]. arXiv preprint arXiv:2107.08430, 2021.Google Scholar
- LIN T Y, DOLLAR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]// The IEEE Conference on Computer Vision and Pattern Recogni-tion (CVPR), July 21-26, 2017, Honolulu, HI, USA. New York : IEEE, 2017:2117-2125Google Scholar
Index Terms
- Lightweight recognition method of infrared sensor image based on deep learning method
Recommendations
Facial expression recognition from near-infrared videos
Facial expression recognition is to determine the emotional state of the face regardless of its identity. Most of the existing datasets for facial expressions are captured in a visible light spectrum. However, the visible light (VIS) can change with ...
Image Target Detection and Recognition Method Using Deep Learning
Image target detection and recognition had been widely used in many fields. However, the existing methods had poor robustness; they not only had high error rate of target recognition but also had high dependence on parameters, so they were limited in ...
Face Recognition Based on Deep Learning
Human Centered ComputingAbstractAs one of the non-contact biometrics, face representation had been widely used in many circumstances. However conventional methods could no longer satisfy the demand at present, due to its low recognition accuracy and restrictions of many ...
Comments