超越单一感知的农田害虫检测算法MRA-YOLOX

doi:10.3778/j.issn.1002-8331.2305-0318

摘要/Abstract

摘要： 目标检测技术正逐步应用于农业，然而在农田害虫检测的运用中仍存在检测速度慢、检测准确率偏低的问题，且仅仅预测害虫的种类和位置信息不足以满足复杂的工程需求。提出一种可以额外预测害虫状态信息的融合MAE和YOLOX算法的高速高精度农田害虫检测模型MRA-YOLOX（masked autoencoders and rapid aim detection-exceeding YOLO）。算法构建包含近4万张图片以及5万余标注的数据集TDBFP（target detection dataset be used for farmland pests），TDBFP数据集标注了10种害虫的生长状态、物种类别以及位置，以便更好地把握害虫信息，从而更准确地制定对策。修改YOLOX模型的解耦头及loss，额外输出生长状态，以改进模型预测更多信息；将ECA（efficient channel attention）和SA（shuffle attention）注意力机制进行有机融合，并插入backbone与FPN（feature pyramid networks）的连接过程以及FPN的通道堆叠过程，以便能够增强获得全局信息和丰富上下文信息的能力，从而取得比单一注意力机制更好的效果；将MAE中自监督解码器部分插入YOLOX的数据增强部分，扩大感受野，增强识别细粒度，获得超越mixup和mosaic的数据增强效果。实验结果表明，当需要同时感知目标的状态、分类和位置时，MRA-YOLOX相较于原始YOLOX模型对于TDBFP数据集的检测精度mAP@0.5由60.1%上升到88.2%，平均检测准确率提高了18.8个百分点，且检测帧率达到145?FPS，可以用于更复杂的工程实践。

关键词: 掩码自编码器（MAE）, 注意力机制, YOLOX, 害虫识别, 状态检测

Abstract: At present, target detection technology is gradually applied in agriculture, but there are still problems in the application of farmland pest detection, such as slow detection speed and low detection accuracy, and only predicting the type and location information of pests is not enough to meet the complex engineering needs. In this paper, a high speed and high precision farmland pest detection model MRA-YOLOX (masked autoencoders and rapid aim detection-exceeding YOLO), which can be used to predict additional pest status information, is proposed by fusing MAE and YOLOX algorithm. By constructing nearly 40?000 pictures and more than 50?000 labels dataset TDBFP (target detection dataset be used for farmland pests), the TDBFP dataset labels the growth status, species category and location of 10 kinds of pests, so as to better grasp the information of pests and develop more accurate countermeasures. Firstly, the decoupling head and loss of YOLOX model are modified to output additional growth states to improve model prediction. Secondly, ECA and SA attention mechanisms are organically integrated, and the connection process between backbone and FPN and the channel stacking process of FPN are inserted, so as to enhance the ability to obtain global information and enrich context information and achieve better results than a single attention mechanism. Finally, the self-supervised decoder part of MAE is inserted into the data enhancement part of YOLOX in order to expand the receptive field, enhance the recognition granularity, and obtain the data enhancement effect beyond mixup and mosaic. Experimental results show that when it is necessary to perceive the state, classification and position of the target at the same time, compared with the original YOLOX model, the detection accuracy of MLA-YOLOX for TDBFP dataset mAP@0.5 increases from 60.1% to 88.2%, the average detection accuracy increases by 18.8 percentage points, and the detection frame rate reaches 145?FPS, MLA-YOLOX can be used for more complex engineering.

Key words: masked autoencoders (MAE), attention mechanism, YOLOX, pest detection, state detection

王中天, 邹颖波, 吴昌霖, 李新. 超越单一感知的农田害虫检测算法MRA-YOLOX[J]. 计算机工程与应用, 2024, 60(16): 206-216.

WANG Zhongtian, ZOU Yingbo, WU Changlin, LI Xin. MRA-YOLOX for Pest Detection in Farmland Beyond Single Perception[J]. Computer Engineering and Applications, 2024, 60(16): 206-216.

参考文献

[1] 罗锡文, 廖娟, 胡炼, 等. 我国智能农机的研究进展与无人农场的实践[J]. 华南农业大学学报, 2021, 42(6): 8-17.
LUO X W, LIAO J, HU L, et al. Research progress of intelligent agricultural machinery and practice of unmanned farm in China[J]. Journal of South China Agricultural University, 2021, 42(6): 8-17.
[2] CHENNG X, ZHANG Y, CHEN Y, et al. Pest identification via deep residual learning in complex background[J]. Computers and Electronics in Agriculture, 2017, 141: 351-356.
[3] ZHENG L, SHEN L. Scalable person re-identification: a benchmark[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision, 2015: 1116-1124.
[4] 翟肇裕, 曹益飞, 徐焕良, 等. 农作物病虫害识别关键技术研究综述[J]. 农业机械学报, 2021, 52(7): 1-18.
ZHAI Z Y, CAO Y F, XU H L, et al. Review of key techniques for crop disease and pest detection[J]. Transactions of the Chinese Society for Agricultural Machinery, 2021, 52(7): 1-18.
[5] 赵路欢, 张玉波. 农业害虫专家系统信息化平台的构建[J]. 安徽农业科学, 2016, 44(15): 255-256.
ZHAO L H, ZHANG Y B. Construction of expert system information platform of agricultural pests[J]. Journal of Anhui Agriculture, 2016, 44(15): 255-256.
[6] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, Jun 23-28, 2014. Washington: IEEE Computer Society, 2014: 580-587.
[7] GIRSHICK R. Fast R-CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision, Santiago, Dec 7-13, 2015. Piscataway: IEEE, 2015: 1440-1448.
[8] JOSEPH R, SANTOSH D. You only look once: unified, real-time object detection[J]. arXiv:1506.02640, 2015.
[9] WANG C Y, ALEXEY B. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[J]. arXiv:2207.02696, 2022.
[10] GE Z, LIU S. YOLOX: ?exceeding?YOLO?series?in?2021[J]. arXiv:2107.08430, 2021.
[11] CHEN Q, WANG Y. You only look one-level feature[J]. arXiv:2103.09460, 2021.
[12] TENG Y, ZHANG J. MSR-RCNN: a multi-class crop pest detection network based on a multi-scale super-resolution feature enhancement module[J]. Frontiers in Plant Science, 2022, 13: 810546.
[13] 王昕, 董琴, 杨国宇. 基于优化CBAM改进YOLOv5的农作物病虫害识别[J]. 计算机系统应用, 2023, 32(7): 261-268.
WANG X, DONG Q, YANG G Y. YOLOv5 improved by optimized CBAM for crop pest identification[J]. Computer Systems?&?Applications, 2023, 32(7): 261-268.
[14] 沈怀艳, 吴云. 基于MSFA-Net的肝脏CT图像分割方法[J]. 计算机科学与探索, 2023, 17(3): 646-656.
SHEN H Y, WU Y. Liver CT image segmentation method based on MSFA-Net[J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(3): 646-656.
[15] 夏鸿斌, 肖奕飞, 刘渊. 融合自注意力机制的长文本生成对抗网络模型[J]. 计算机科学与探索, 2022, 16(7): 1603-1610.
XIA X B, XIAO Y F, LIU Y. Long text generation adversarial network model with self-attention mechanism[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(7): 1603-1610.
[16] 程艳, 蔡壮, 吴刚, 等. 结合自注意力特征过滤分类器和双分支GAN的面部表情识别[J]. 模式识别与人工智能, 2022, 35(3): 243-253.
CHENG Y, CAO Z, WU G, et al. Facial expression recognition combining self-attention feature filtering classifier and two-branch GAN[J]. Pattern Recognition and Artificial Intelligence, 2022, 35(3): 243-253.
[17] JIA L Q, WANG T. MobileNet-CA-YOLO: an improved YOLOv7 based on the MobileNetV3 and attention mechanism for rice pests and diseases detection[J]. Agriculture, 2023, 13(7): 1285.
[18] XIANG Q C, HUANG X N. Yolo-Pest: ?an?insect?pest?object detection algorithm via CAC3 module[J]. Sensors, 2023, 23(6): 3221.
[19] WANG Q, WU B, ZHU P, et al. ECA-Net: efficient?channel?attention?for?deep?convolutional?neural?networks[J]. arXiv:1910.03151, 2019.
[20] FENG C, YAN Y. Exploring separable attention for multi-contrast MR image super-resolution[J]. arXiv:2109.01664, 2021.
[21] HE K, CHEN X. Masked autoencoders are scalable vision learners[J]. arXiv:2111.06377, 2021.
[22] HJELMAS E, LOW B K. Face detection: a survey[J]. Computer Vision and Image Understanding, 2001(83): 236-274.
[23] HUANG G B, MATTAR M, BERG T, et al. Labeled faces in the wild: a database for studying face recognition in unconstrained environments[EB/OL]. (2008)[2023-03-10]. https://api.semanticscholar.org/CorpusID:88166.
[24] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//Proceedings of the 2015 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2005: 886-893.
[25] DOLLAR P, WOJEK C, SCHIELE B, et al. Pedestrian detection: an evaluation of the state of the art[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(4): 743-761.
[26] TSUNG Y L, MAIRE M, BELONGIE S, et al. Microsoft COCO: common objects in context[C]//Proceedings of the 13th European Conference on Computer Vision, Zurich, Sep 6-12, 2014. Cham: Springer, 2014: 740-755.
[27] EVERINGHAM M, GOOL L V, CHRISTOPHER K I W, et al. The pascal visual object classes (VOC) challenge[J]. International Journal of Computer Vision, 2010, 88(2): 303-338.
[28] DENG J, DONG W, SOCHERE R, et al. ImageNet: a large scale hierarchical image database[C]//Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, 2009: 248-255.
[29] LAO S, WANG D, LI F, et al. Human running detection: benchmark and baseline[J]. Computer Vision and Image Understanding, 2016, 153: 143-150.
[30] LIU X, DENG Z, LU H, et al. Benchmark for road marking detection: dataset specification and performance baseline[C]//Proceedings of the 2017 IEEE 20th International Conference on Intelligent Transportation Systems, Yokohama, 2017: 1-6.
[31] NAUDE J, JOUBERT D. The aerial elephant dataset: a new public benchmark for aerial object detection[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, Jun 16-20, 2019: 48-55.
[32] WOSNER O, FARJON G, BARHILLEL A. Object detection in agricultural contexts: a multiple resolution benchmark and comparison to human[J]. Computers and Electronics in Agriculture, 2021, 189: 106404.
[33] WOO S, PARK J. CBAM: convolutional block attention module[J]. arXiv:1807.06521, 2018.
[34] XU H, DING S. Masked autoencoders are robust data augmentors[J]. arXiv:2206.04846, 2022.