Forest pest monitoring and prevention based on UAV image detection

doi:10.14088/j.cnki.issn0439-8114.2024.08.044

Abstract

Abstract: In order to solve the problem of low efficiency and poor effect of existing pest control methods, which required a lot of manpower and material resources, the research built a forest pest detection framework based on deep learning, which transferred the feature information extracted from the shallow network to the deep network, and made lightweight improvements to the model through pruning and batch normalization folding. The results showed that, when each model tended to be stable during training, the average accuracy of the improved YOLOv4 model reached 97.38%, and compared with the original YOLOv4 model, the computing cost and storage requirements were reduced by 17.81 percent points and 23.38%, respectively. The average detection accuracy was 12.75 percent points higher than before.

Key words: unmanned aerial vehicle（UAV）, pest control, image detection, YOLOv4 model

CLC Number:

S763
TP751

QIU Ya-lin, LIU Xiang-long, HE Xiao-jun, ZHAO Qing-long, JIA Cun-fang. Forest pest monitoring and prevention based on UAV image detection[J]. HUBEI AGRICULTURAL SCIENCES, 2024, 63(8): 262-266.

References

[1] 慕君林, 马博, 王云飞, 等. 基于深度学习的农作物病虫害检测算法综述[J]. 农业机械学报, 2023, 54(2):301-313.
[2] 张家盛, 梁进兴. 基于深度学习的无人机巡检架空输电线路金具锈蚀缺陷检测方法[J]. 湖南电力, 2022, 42(5):75-78.
[3] JABER N, DAHER N A, ASMAR D C.Deep-Pest-Detector: Automated detection and localization of processionary moth nests on pine trees via aerial drones and deep neural networks[J]. Applied engineering in agriculture, 2022, 28(2):253-272.
[4] 俞天宇, 倪文俭, 刘见礼, 等. 大兴安岭林区无人机可见光影像散发枯立木识别算法[J]. 遥感学报, 2021, 25(3):725-736.
[5] ALBATTAH W, MASOOD M, JAVED A, et al.Custom cornernet: A drone-based improved deep learning technique for large-scale multiclass pest localization and classification[J]. Complex & intelligent systems, 2023, 9(2):1299-1316.
[6] 白雪松, 吴建平, 景文超, 等. 基于改进残差网络的农作物病虫害检测研究[J]. 计算机技术与发展, 2023, 33(5):145-151.
[7] DEMIR K, TÜMEN V. Drone-assisted automated plant diseases identification using spiking deep conventional neural learning[J]. Ai communications, 2021, 34(2):147-162.
[8] 秦彩杰, 李勇. 作物病虫害自动识别技术研究——基于视频监控和无人机平台[J]. 农机化研究, 2021, 43(10):42-45.
[9] ODEMER R, ODEMER F, LIEBIG G, et al.Temporal increase of varroa mites in trap frames used for drone brood removal during the honey bee season[J]. Journal of applied entomology,2022,146(9):1207-1211.
[10] 刘文, 陆小锋, 毛建华, 等. 基于机载计算机的无人机智能巡检方案[J]. 计算机测量与控制, 2022, 30(7):181-186.
[11] NATHAN M G, BREWER M J.A special collection: Drones to improve insect pest management[J]. Journal of economic entomology, 2021, 114(5):1853-1856.
[12] 周焱, 刘文萍, 骆有庆, 等. 基于深度学习的小目标受灾树木检测方法[J]. 林业科学, 2021, 57(3):98-107.
[13] GUMAEI A, AL-RAKHAMI M, HASSAN M, et al.Deep learning and blockchain with edge computing for 5G-Enabled drone identification and flight mode detection[J]. IEEE network, 2021, 35(1):94-100.
[14] ISHAKH P, DZURISIN J, FOLLETT P.Evaluation of macadamia felted coccid (Hemiptera: Eriococcidae) damage and cultivar susceptibility using imagery from a small unmanned aerial vehicle (sUAV), combined with ground truthing[J]. Pest management science, 2022, 78(11):4533-4543.
[15] 李浩, 方伟泉, 李浪浪, 等. 基于深度学习的松材线虫病害松木识别[J]. 林业工程学报, 2021, 6(6):142-147.