OSAF-Net: A one-stage anchor-free detector for small-target crop pest detection
Tóm tắt
Multi-class crop pest detection in massive images is a practically challenging problem. Recently, convolutional neural networks (CNN) based approaches have shown promise in detecting crop pests, but there are still significant obstacles to overcome. Two primary challenges include the highly similar physical appearance of some categories, making it difficult to distinguish the specific categories manually, and the multi-scale characteristics of pest objects, leading to numerous false negative detections, especially for small pests. To address the above problems, we propose a one-stage anchor-free detection network (OSAF-Net) with strong performance and robustness. Firstly, a dynamic training sample selection (DTSS) method is devised to capture high-quality training examples that contain multi-scale contextual information to improve the detection performance of pests with diverse scales. Secondly, to mitigate the disturbance from the similar physical appearance of pests, a dynamic detection head (DDH) is introduced to accurately obtain more representative semantic features to locate and distinguish pest objects. The proposed DTSS and DDH methods are stable for implementation. They can be combined with existing state-of-the-art detection network architectures. Extensive experiments conducted on two datasets, CropPest24 and MPD2018, demonstrate that our proposed method has a competitive performance, achieving AP50 of 77.3% on CropPest24 and 81.3% on MPD2018.
Tài liệu tham khảo
Ebrahimi MA, Khoshtaghaz MH, Minaei S, Jamshidi B (2017) Vision-based pest detection based on SVM classification method. Comput Electron Agric 137:52–58. https://doi.org/10.1016/j.compag.2017.03.016
Espinoza K, Valera DL, Torres JA et al (2016) Combination of image processing and artificial neural networks as a novel approach for the identification of Bemisia tabaci and Frankliniella occidentalis on sticky traps in greenhouse agriculture. Comput Electron Agric 127:495–505. https://doi.org/10.1016/j.compag.2016.07.008
Samanta R, Ghosh I (2012) Tea insect pests classification based on artificial neural networks. International Journal of Computer Engineering Science (IJCES) 2:1–13
Liu C, Ding W, Yang J et al (2020) Aggregation Signature for Small Object Tracking. IEEE Trans on Image Process 29:1738–1747. https://doi.org/10.1109/TIP.2019.2940477
Ng S-C, Kwok C-P (2020) An Intelligent Traffic Light System Using Object Detection and Evolutionary Algorithm for Alleviating Traffic Congestion in Hong Kong. Int J Comput Intell Syst 13:802–809. https://doi.org/10.2991/ijcis.d.200522.001
Yu D Ji S (2021) A New Spatial-Oriented Object Detection Framework for Remote Sensing Images. IEEE Trans Geosci Remote Sensing 1–1. https://doi.org/10.1109/TGRS.2021.3127232
Ren S, He K, Girshick R, Sun J (2017) Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Trans Pattern Anal Mach Intell 39:1137–1149. https://doi.org/10.1109/TPAMI.2016.2577031
Cai Z, Vasconcelos N (2021) Cascade R-CNN: High Quality Object Detection and Instance Segmentation. IEEE Trans Pattern Anal Mach Intell 43:1483–1498. https://doi.org/10.1109/TPAMI.2019.2956516
Redmon J, Divvala SK, Girshick RB, Farhadi A (2016) You Only Look Once: Unified, Real-Time Object Detection. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, NV, USA, June 27–30, 2016. IEEE Computer Society, pp 779–788
Hsu W-Y, Lin W-Y (2021) Ratio-and-Scale-Aware YOLO for Pedestrian Detection. IEEE Trans Image Process 30:934–947. https://doi.org/10.1109/TIP.2020.3039574
Xu K, Xu Y, Xing Y, Liu Z (2023) YOLO-F: YOLO for Flame Detection. Int J Pattern Recognit Artif Intell 37:. https://doi.org/10.1142/S0218001422500434
Liu W, Anguelov D, Erhan D, et al (2016) SSD: Single Shot MultiBox Detector. In: Leibe B, Matas J, Sebe N, Welling M (eds) Computer Vision - ECCV 2016 - 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I. Springer, pp 21–37
Lin T-Y, Goyal P, Girshick R et al (2020) Focal Loss for Dense Object Detection. IEEE Trans Pattern Anal Mach Intell 42:318–327. https://doi.org/10.1109/TPAMI.2018.2858826
Jiao L, Dong S, Zhang S et al (2020) AF-RCNN: An anchor-free convolutional neural network for multi-categories agricultural pest detection. Comput Electron Agric 174:105522. https://doi.org/10.1016/j.compag.2020.105522
Liu L, Wang R, Xie C et al (2019) PestNet : an end-to-end deep learning approach for large-scale multi-class pest detection and classification. IEEE Access 7:45301–45312
Dong S, Wang R, Liu K et al (2021) CRA-Net: A channel recalibration feature pyramid network for detecting small pests. Comput Electron Agric 191:106518. https://doi.org/10.1016/j.compag.2021.106518
Wang F, Wang R, Xie C et al (2021) Convolutional neural network based automatic pest monitoring system using hand-held mobile image analysis towards non-site-specific wild environment. Comput Electron Agric 187:106268. https://doi.org/10.1016/j.compag.2021.106268
Wang R, Liu L, Xie C et al (2021) AgriPest: A Large-Scale Domain-Specific Benchmark Dataset for Practical Agricultural Pest Detection in the Wild. Sensors 21:1601
Dai J, Qi H, Xiong Y, et al (2017) Deformable Convolutional Networks. In: IEEE Int Conf Comput Vis, ICCV 2017, Venice, Italy, October 22–29, 2017. IEEE, pp 764–773
Redmon J, Farhadi A (2018) Yolov3: An incremental improvement. arXiv preprint arXiv:180402767
Law H, Deng J (2020) CornerNet: Detecting Objects as Paired Keypoints. Int J Comput Vision 128:642–656. https://doi.org/10.1007/s11263-019-01204-1
Tian Z, Shen C, Chen H, He T (2022) FCOS: A Simple and Strong Anchor-Free Object Detector. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 44:1922–1933. https://doi.org/10.1109/TPAMI.2020.3032166
Hoiem D, Chodpathumwan Y, Dai Q (2012) Diagnosing Error in Object Detectors. In: Fitzgibbon A, Lazebnik S, Perona P, et al (eds) Computer Vision - Eccv 2012, Pt Iii. pp 340–353
He K, Zhang X, Ren S, Sun J (2016) Deep Residual Learning for Image Recognition. In: 2016 IEEE Conf Comput Vis Pattern Recognit, CVPR 2016, Las Vegas, NV, USA, June 27–30, 2016. pp 770–778
Gao S, Cheng M, Zhao K et al (2021) Res2Net: A New Multi-Scale Backbone Architecture. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 43:652–662. https://doi.org/10.1109/TPAMI.2019.2938758
Xie S, Girshick RB, Dollár P, et al (2017) Aggregated Residual Transformations for Deep Neural Networks. In: 2017 IEEE Conf Comput Vis Pattern Recognit, CVPR 2017, Honolulu, HI, USA, July 21–26, 2017. pp 5987–5995
Liu Z, Lin Y, Cao Y, et al (2021) Swin transformer: Hierarchical vision transformer using shifted windows. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp 10012–10022
Shifeng Dong, Rujing Wang, Jianming Du (2022) CropPest24-Dataset. https://github.com/observar/CropPest24