基于改进YOLOv4的林业害虫检测

doi:10.3969/j.issn.1004-1524.2022.06.21

浙江农业学报 ›› 2022, Vol. 34 ›› Issue (6): 1306-1315.DOI: 10.3969/j.issn.1004-1524.2022.06.21

基于改进YOLOv4的林业害虫检测

陈道怀¹(), 汪杭军²^,^*()

1.浙江农林大学信息工程学院,浙江杭州 311300
2.浙江农林大学暨阳学院工程技术学院,浙江诸暨 311800

收稿日期:2021-07-03 出版日期:2022-06-25 发布日期:2022-06-30
通讯作者: 汪杭军
作者简介:*汪杭军,E-mail: whj@zafu.edu.cn
陈道怀（1997—）,男,浙江长兴人,硕士研究生,主要从事图像处理与模式识别研究。E-mail: 1310451391@qq.com
基金资助:
浙江省基础公益研究计划(LGN19C140006);绍兴市科技计划(2018C20013)

Detection of forest pests based on improved YOLOv4

CHEN Daohuai¹(), WANG Hangjun²^,^*()

1. College of Information Engineering, Zhejiang A&F University, Hangzhou 311300, China
2. College of Engineering and Technology, Jiyang College of Zhejiang A&F University, Zhuji 311800, Zhejiang, China

Received:2021-07-03 Online:2022-06-25 Published:2022-06-30
Contact: WANG Hangjun

摘要/Abstract

摘要：

为了提高林业害虫检测的准确性,提出一种基于YOLOv4的改进算法。首先,基于智能害虫捕捉装置拍摄的图像,制作害虫数据集,采用K-means算法对样本数据集的目标框进行聚类分析,基于DIoU-NMS算法实现对害虫的计数功能;然后,在模型的路径聚合网络（PANet）结构上增加特征融合和104×104层级特征检测图,以提升对小个体害虫的识别率;最后,根据模型检测效率和复杂度,调整模型中的尺度特征图组合,在保证检测准确度的基础上,提升检测效率,并精简模型。试验结果表明,改进的YOLOv4模型的平均识别精度比传统YOLOv4模型提高了1.6百分点,且对于小个体害虫的识别效果更好,模型复杂度和模型参数量分别减少了11.9%、33.2%,检测速度提升了11.1%,更适于应用部署。

关键词: 林业害虫, 害虫检测, 深度学习

Abstract:

Abstract：In order to improve the accuracy of pest detection in the forest, an improved algorithm based on YOLOv4 was proposed. First, based on the image captured by the intelligent pest capture device, the pest data set was made, and K-means algorithm was used to cluster the target frame of the sample data set. Based on the DIoU-NMS algorithm, the counting function of pests was realized. Then, feature fusion was added to the path aggregation network (PANet) structure of the model, as well as 104×104 hierarchical feature detection graph, to increase the recognition accuracy of small-sized pests. Finally, according to the efficiency and complexity of model detection, the combination of scale feature maps in the model was adjusted to ensure the detection accuracy, improve the detection efficiency and simplify the model. The experimental results showed that the mean average precision of the improved YOLOv4 model was 1.6 percent higher than that of the traditional YOLOv4 model with a better performance on the detection of small-sized pests. Besides, its speed was improved by 11.1 percent, and the model complexity and model parameters were reduced by 11.9 percent and 33.2 percent, respectively, as compared with the traditional YOLOv4 model, which was more suitable for application deployment.

Key words: forest pests, pest detection, deep learning

中图分类号:

S431.9

陈道怀, 汪杭军. 基于改进YOLOv4的林业害虫检测[J]. 浙江农业学报, 2022, 34(6): 1306-1315.

CHEN Daohuai, WANG Hangjun. Detection of forest pests based on improved YOLOv4[J]. Acta Agriculturae Zhejiangensis, 2022, 34(6): 1306-1315.

图/表 12

参考文献 21

[1]	顾晓丽, 潘洁, 张衡, 等. 基于物联网架构的我国森林病虫害监测研究进展[J]. 世界林业研究, 2015, 28(2): 48-53.
	GU X L, PAN J, ZHANG H, et al. Review of forest diseases and pests monitoring research in China based on Internet of Things framework[J]. World Forestry Research, 2015, 28(2): 48-53. (in Chinese with English abstract)
[2]	贾少鹏, 高红菊, 杭潇. 基于深度学习的农作物病虫害图像识别技术研究进展[J]. 农业机械学报, 2019, 50(S1): 313-317.
	JIA S P, GAO H J, HANG X. Research progress on image recognition technology of crop pests and diseases based on deep learning[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(S1): 313-317. (in Chinese with English abstract)
[3]	邱道尹, 张红涛, 刘新宇, 等. 基于机器视觉的大田害虫检测系统[J]. 农业机械学报, 2007, 38(1): 120-122.
	QIU D Y, ZHANG H T, LIU X Y, et al. Design of detection system for agriculture field pests based on machine vision[J]. Transactions of the Chinese Society for Agricultural Machinery, 2007, 38(1): 120-122. (in Chinese with English abstract)
[4]	WEN C L, GUYER D. Image-based orchard insect automated identification and classification method[J]. Computers and Electronics in Agriculture, 2012, 89: 110-115. DOI URL
[5]	LIU T, CHEN W, WU W, et al. Detection of aphids in wheat fields using a computer vision technique[J]. Biosystems Engineering, 2016, 141: 82-93. DOI URL
[6]	朱莉, 罗靖, 徐胜勇, 等. 基于颜色特征的油菜害虫机器视觉诊断研究[J]. 农机化研究, 2016, 38(6): 55-58.
	ZHU L, LUO J, XU S Y, et al. Machine vision recognition of rapeseed pests based on color feature[J]. Journal of Agricultural Mechanization Research, 2016, 38(6): 55-58. (in Chinese with English abstract)
[7]	李文勇, 李明, 陈梅香, 等. 基于机器视觉的作物多姿态害虫特征提取与分类方法[J]. 农业工程学报, 2014, 30(14): 154-162.
	LI W Y, LI M, CHEN M X, et al. Feature extraction and classification method of multi-pose pests using machine vision[J]. Transactions of the Chinese Society of Agricultural Engineering, 2014, 30(14): 154-162. (in Chinese with English abstract)
[8]	张睿珂, 陈梅香, 李明, 等. 基于机器视觉的蛾类三维姿态中前翅间夹角计算方法[J]. 林业科学, 2017, 53(11): 120-130.
	ZHANG R K, CHEN M X, LI M, et al. Method of extracting forewings angle of 3D pose for the moth based on machine vision[J]. Scientia Silvae Sinicae, 2017, 53(11): 120-130. (in Chinese with English abstract)
[9]	田冉, 陈梅香, 董大明, 等. 红外传感器与机器视觉融合的果树害虫识别及计数方法[J]. 农业工程学报, 2016, 32(20): 195-201.
	TIAN R, CHEN M X, DONG D M, et al. Identification and counting method of orchard pests based on fusion method of infrared sensor and machine vision[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(20): 195-201. (in Chinese with English abstract)
[10]	肖德琴, 张玉康, 范梅红, 等. 基于视觉感知的蔬菜害虫诱捕计数算法[J]. 农业机械学报, 2018, 49(3): 51-58.
	XIAO D Q, ZHANG Y K, FAN M H, et al. Vegetable pest counting algorithm based on visual perception[J]. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(3): 51-58. (in Chinese with English abstract)
[11]	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.
	ZHOU Y, LIU W P, LUO Y Q, et al. Small object detection for infected trees based on the deep learning method[J]. Scientia Silvae Sinicae, 2021, 57(3): 98-107. (in Chinese with English abstract)
[12]	杨国国, 鲍一丹, 刘子毅. 基于图像显著性分析与卷积神经网络的茶园害虫定位与识别[J]. 农业工程学报, 2017, 33(6): 156-162.
	YANG G G, BAO Y D, LIU Z Y. Localization and recognition of pests in tea plantation based on image saliency analysis and convolutional neural network[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(6): 156-162. (in Chinese with English abstract)
[13]	武英洁, 房世波, Piotr Chudzik, 等. 基于Faster R-CNN的野外环境中蝗虫快速识别[J]. 气象与环境学报, 2020, 36(6): 137-143.
	WU Y J, FANG S B, CHUDZIK P, et al. Rapid identification of locust on fields based on Faster R-CNN[J]. Journal of Meteorology and Environment, 2020, 36(6): 137-143. (in Chinese)
[14]	张博, 张苗辉, 陈运忠. 基于空间金字塔池化和深度卷积神经网络的作物害虫识别[J]. 农业工程学报, 2019, 35(19): 209-215.
	ZHANG B, ZHANG M H, CHEN Y Z. Crop pest identification based on spatial pyramid pooling and deep convolution neural network[J]. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(19): 209-215. (in Chinese with English abstract)
[15]	李衡霞, 龙陈锋, 曾蒙, 等. 一种基于深度卷积神经网络的油菜虫害检测方法[J]. 湖南农业大学学报(自然科学版), 2019, 45(5): 560-564.
	LI H X, LONG C F, ZENG M, et al. A detecting method for the rape pests based on deep convolutional neural network[J]. Journal of Hunan Agricultural University (Natural Sciences), 2019, 45(5): 560-564. (in Chinese with English abstract)
[16]	李想. 基于物联网的虫情监测系统[D]. 北京: 北京林业大学, 2019.
	LI X. The pest monitoring system based on the Internet of Things[D]. Beijing: Beijing Forestry University, 2019. (in Chinese with English abstract)
[17]	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition, June 27-30, 2016, Las Vegas, NV, USA: IEEE, 2016: 779-788.
[18]	王金鹏, 高凯, 姜洪喆,等. 基于改进的轻量化卷积神经网络火龙果检测方法(英)[J]. 农业工程学报, 2020, 36(20): 218-225.
	WANG J P, GAO K, JIANG H Z, et al. Method for detecting dragon fruit based on improved lightweight convolutional neural network[J]. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36(20): 218-225.(in English)
[19]	蔡舒平, 孙仲鸣, 刘慧, 等. 基于改进型YOLOv4的果园障碍物实时检测方法[J]. 农业工程学报, 2021, 37(2): 36-43.
	CAI S P, SUN Z M, LIU H, et al. Real-time detection methodology for obstacles in orchards using improved YOLOv4[J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(2): 36-43. (in Chinese with English abstract)
[20]	LIU S, QI L, QIN H F, et al. Path aggregation network for instance segmentation[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 18-23, 2018, Salt Lake City, UT, USA: IEEE, 2018: 8759-8768.
[21]	鞠默然, 罗海波, 王仲博, 等. 改进的YOLO V3算法及其在小目标检测中的应用[J]. 光学学报, 2019, 39(7): 0715004.
	JU M R, LUO H B, WANG Z B, et al. Improved YOLO V3 algorithm and its application in small target detection[J]. Acta Optica Sinica, 2019, 39(7): 0715004. (in Chinese with English abstract)

编辑推荐

Metrics

阅读次数

全文

729

HTML			PDF

最新录用	在线预览	正式出版	最新录用	在线预览	正式出版
0	0	17	0	0	712

来源	本网站	其他网站

次数	449	280
比例	62%	38%

摘要

428

最新录用	在线预览	正式出版

0	0	428

	来源	本网站

	次数	428
	比例	100%

害虫种类 Pest species	害虫尺寸 Pest size	样本数量 Sample quantity	害虫种类 Pest species	害虫尺寸 Pest size	样本数量 Sample quantity
美苔蛾Miltochrista miniata	小Small	1 060	斜矛丽金龟Callistethus plagiicollis	中Medium	1 057
黄土苔蛾Eilema nigripoda	小Small	1 067	铜绿丽金龟Anomala corpulenta Motschulsky	大Big	1 289
条纹小斑蛾Thyrassia penangae	小Small	1 057	毛黄鳃金龟Holotrichia trichophora	大Big	1 156
突背斑红蝽Physopelta gutta	小Small	1 189	茶斑蛾Eterusia aedea	大Big	1 081
八点灰灯蛾Creatonotos transiens	中Medium	1 060	蚱蜢Locusta migratoria	大Big	903

害虫种类 Pest species	害虫尺寸 Pest size	样本数量 Sample quantity	害虫种类 Pest species	害虫尺寸 Pest size	样本数量 Sample quantity
美苔蛾Miltochrista miniata	小Small	1 060	斜矛丽金龟Callistethus plagiicollis	中Medium	1 057
黄土苔蛾Eilema nigripoda	小Small	1 067	铜绿丽金龟Anomala corpulenta Motschulsky	大Big	1 289
条纹小斑蛾Thyrassia penangae	小Small	1 057	毛黄鳃金龟Holotrichia trichophora	大Big	1 156
突背斑红蝽Physopelta gutta	小Small	1 189	茶斑蛾Eterusia aedea	大Big	1 081
八点灰灯蛾Creatonotos transiens	中Medium	1 060	蚱蜢Locusta migratoria	大Big	903

模型Model	特征图Feature map
YOLOv4	13×13、26×26、52×52
Improved YOLOv4-1	13×13、26×26、52×52、104×104
Improved YOLOv4-2	26×26、52×52、104×104
Improved YOLOv4-3	52×52、104×104
Improved YOLOv4-4	104×104

模型Model	特征图Feature map
YOLOv4	13×13、26×26、52×52
Improved YOLOv4-1	13×13、26×26、52×52、104×104
Improved YOLOv4-2	26×26、52×52、104×104
Improved YOLOv4-3	52×52、104×104
Improved YOLOv4-4	104×104

模型 Model	MAP/%	FPS	FLOPs/B	MP/MB
YOLOv3	91.9	41	65	240
YOLOv4	94.8	36	59	250
Improved YOLOv4-1	96.1	30	69	254
Improved YOLOv4-2	96.3	33	64	187
Improved YOLOv4-3	96.0	38	58	171
Improved YOLOv4-4	96.4	40	52	167

基于改进YOLOv4的林业害虫检测

Detection of forest pests based on improved YOLOv4

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 21

相关文章 4

编辑推荐

Metrics

本文评价

编号 Number	害虫尺寸 Pest size	YOLOv3/%	YOLOv4/ %	Improved YOLOv4-4/ %
1	小Small	89.2	93.1	96.0
2	小Small	91.4	91.7	94.9
3	小Small	91.5	94.4	96.0
4	小Small	92.8	93.0	94.4
5	中Medium	97.1	98.3	98.0
6	中Medium	96.1	97.7	96.9
7	大Big	98.3	99.1	99.1
8	大Big	97.0	96.9	97.7
9	大Big	98.5	99.0	99.0
10	大Big	95.8	98.9	97.5
11	—	62.5	80.9	90.9
MAP	—	91.9	94.8	96.4

[1]	张晴晴, 刘连忠, 宁井铭, 吴国栋, 江朝晖, 李孟杰, 李栋梁. 基于YOLOV3优化模型的复杂场景下茶树嫩芽识别[J]. 浙江农业学报, 2021, 33(9): 1740-1747.
[2]	鲍烈, 王曼韬, 刘江川, 文波, 明月. 基于卷积神经网络的小麦产量预估方法[J]. 浙江农业学报, 2020, 32(12): 2244-2252.
[3]	王彦翔, 张艳, 杨成娅, 孟庆龙, 尚静. 基于深度学习的农作物病害图像识别技术进展[J]. 浙江农业学报, 2019, 31(4): 669-676.
[4]	杨国亮, 许楠, 康乐乐, 龚曼, 洪志阳. 基于参数指数非线性残差神经网络的脐橙病变叶片识别[J]. 浙江农业学报, 2018, 30(6): 1073-1081.