A new cotton aphid image recognition algorithm and software based on YOLOv8

doi:10.12398/j.issn.2096-7217.2023.03.005

Abstract

Abstract:

In order to solve the problems of high difficulty and low efficiency of manual counting in scientific research of cotton aphid population prediction and control， a cotton aphid image recognition algorithm based on YOLO neural network was proposed and developed into software. First， the images of artificially inoculated cotton aphids were taken by mobile phone for 15 consecutive days， 50 clear images were selected and cut into 6 sub-images， and then the training set and test set were obtained by using LabelImg software for manual labeling. Then， 10 models from the series of YOLOv5 and YOLOv8， whose training parameters were set as the same（batch size was 32， iteration was 100 rounds， initial learning rate was 0.01， and periodic learning rate was 0.01）， were selected and trained by using the server of the AutoDL platform. Finally， the trained models were tested， and the YOLOv8l model showed the best overall performance， with mAP50 reaching 0.926. In order to provide users with convenient and easy-to-use man-machine software， the front end of the software was developed by using PYQT5， to realize the functions of reading and counting of cotton aphid pictures， visualizing results and exporting results to Excel. The back end of the software adopted an image processing method of "splitting-detection-merging"， which ensuring the efficient detection of YOLO model on small targets. After testing， the software had an average precision of 0.945 for counting dead cotton aphids and live cotton aphids， which was comparable to manual counting and had good practical value. This research may provide an intelligent detection tool for researchers related to cotton aphid control and also provided key operation information for precise operation in scenes such as unmanned farms.

Key words: cotton aphid, counting, image recognition, image processing, YOLO

CLC Number:

MA Pan, YANG Ziheng, WAN Hu, HE Shun, HUANG Yuan, XU Shengyong. A new cotton aphid image recognition algorithm and software based on YOLOv8[J]. Journal of Intelligent Agricultural Mechanization, 2023, 4(3): 42-49.

Figures/Tables 7

References 27

1	李越，朱鹤，单莹，等. 棉蚜对吡虫啉和氟啶虫胺腈抗药性监测及抗性机理研究进展［J］. 中国棉花， 2023， 50（2）： 1-5.
	LI Yue， ZHU He， SHAN Ying， et al. Resistance monitoring and resistant mechanism progress of Aphis gossypii to imidacloprid and sulfoxaflor ［J］. China Cotton， 2023， 50（2）： 1-5.
2	王小丽，吴娜，张玉栋，等. 取食蚜害后的棉花对棉蚜营养代谢的影响［J］. 应用昆虫学报， 2022， 59（4）： 854-861.
	WANG Xiaoli， WU Na， ZHANG Yudong， et al. Effects of aphid damage to cotton plants on the nutritional metabolism of Aphis gossypii ［J］. Chinese Journal of Applied Entomology， 2022， 59（4）： 854-861.
3	王晨，刘金萍，杨益众，等. 干旱胁迫对棉花上棉蚜种群增长影响的评价方法［J］. 植物保护， 2023， 49（1）： 193-199.
	WANG Chen， LIU Jinping， YANG Yizhong， et al. A method for assessing the effects of drought stress on population growth of Aphis gossypii Glover on cotton plants ［J］. Plant Protection， 2023， 49（1）： 193-199.
4	李文娟，张睿，蔡志华，等. 苯炔［3+2］环加成反应构建三氟甲基取代的苯并环状亚砜亚胺衍生物及其杀棉蚜活性研究［J］. 有机化学， 2022， 42（9）： 2832-2839.
	LI Wenjuan， ZHANG Rui， CAI Zhihua， et al. Construction and insecticidal activities of trifluoromethylated benzocyclicsulfoximine derivatives by ［3+2］ cycloaddition reaction of beznyne ［J］. Chinese Journal of Organic Chemistry， 2022， 42（9）： 2832-2839.
5	时英爽，高有华，蔡超，等. 新疆棉田吡虫啉残留动态及其防治棉蚜田间用药量确定［J］. 中国棉花， 2019， 46（10）： 25-29， 37.
	SHI Yingshuang， GAO Youhua， CAI Chao， et al. Imidacloprid residue dynamics in the cotton field of Xinjiang and determination of its dosage for controlling cotton aphid ［J］. China Cotton， 2019， 46（10）： 25-29， 37.
6	丁瑞丰，朱晓华，阿克旦·吾外士，等. 人工释放普通草蛉田间防治棉蚜效果研究［J］. 植物保护， 2015， 41（2）： 200-204.
	DING Ruifeng， ZHU Xiaohua， AKEDAN·WUWAISHI， et al. Control effects on Aphis gossypii （Glover） by releasing Chrysoperla carnea （Stephens） in cotton field ［J］. Plant Protection， 2015， 41（2）： 200-204.
7	KRIZHEVSKY A， SUTSKEVER I， HINTON G E. ImageNet classification with deep convolutional neural networks ［J］. Communications of the ACM， 2017， 60（6）： 84-90.
8	GIRSHICK R， DONAHUE J， DARRELL T， et al. Rich feature hierarchies for accurate object detection and semantic segmentation ［C］// Proceedings of the IEEE conference on computer vision and pattern recognition. 2014： 580-587.
9	卜俊怡，孙国祥，王迎旭，等. 基于诱虫板图像的温室番茄作物害虫识别与监测方法［J］. 南京农业大学学报， 2021， 44（2）： 373-383.
	BU Junyi， SUN Guoxiang， WANG Yingxu， et al. Identification and monitoring method of tomato crop pests in greenhouse based on trapping board image ［J］. Journal of Nanjing Agricultural University， 2021， 44（2）： 373-383.
10	ZHU C H， WANG J J， LIU H， et al. Insect identification and counting in stored grain： Image processing approach and application embedded in smartphones ［J］. Mobile Information Systems， 2018： 1-5.
11	俞浩，吕美巧，刘丽敏，等. 高光谱成像与图像结合进行油菜角果蚜虫侵染的定位识别［J］. 光谱学与光谱分析， 2017， 37（10）： 3193-3197.
	YU hao， Meiqiao LÜ， LIU Limin， et al. Identification of aphid infection on rape pods using hyperspectral imaging combined with image processing ［J］. Spectroscopy and Spectral Analysis， 2017， 37（10）： 3193-3197.
12	HAYASHI M， TAMAI K， OWASHI Y， et al. Automated machine learning for identification of pest aphid species （Hemiptera： Aphididae）［J］. Applied Entomology and Zoology， 2019， 54： 487-490.
13	SHAJAHAN S， SIVARAJAN S， MAHARLOOEI M， et al. Identification and counting of soybean aphids from digital images using shape classification ［J］. Transactions of the ASABE， 2017， 60（5）： 1467-1477.
14	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.
15	王林惠，兰玉彬，刘志壮，等. 便携式柑橘虫害实时检测系统的研制与试验［J］. 农业工程学报， 2021， 37（9）： 282-288.
	WANG Linhui， LAN Yubin， LIU Zhizhuang， et al. Development and experiment of the portable real-time detection system for citrus pests ［J］. Transactions of the Chinese Society of Agricultural Engineering， 2021， 37（9）： 282-288.
16	CHEN J， FAN Y Y， WANG T， et al. Automatic segmentation and counting of aphid nymphs on leaves using convolutional neural networks ［J］. Agronomy， 2018， 8（8）： 129.
17	董伟，钱蓉，张洁，等. 基于深度学习的蔬菜鳞翅目害虫自动识别与检测计数［J］. 中国农业科技导报， 2019， 21（12）： 76-84.
	DONG Wei， QIAN Rong， ZHANG Jie， et al. Vegetable lepidopteran pest auto recognition and detection counting based on deep learning ［J］. Journal of Agricultural Science and Technology， 2019， 21（12）： 76-84.
18	YAN T Y， XU W， LIN J， et al. Combining multi-dimensional convolutional neural network （CNN） with visualization method for detection of Aphis gossypii Glover infection in cotton leaves using hyperspectral imaging ［J］. Frontiers in Plant Science， 2021， 12： 604510.
19	CHEN P， LI W L， YAO S J， et al. Recognition and counting of wheat mites in wheat fields by a three-step deep learning method ［J］. Neurocomputing， 2021， 437： 21-30.
20	陈娟，陈良勇，王生生，等. 基于改进残差网络的园林害虫图像识别［J］. 农业机械学报， 2019， 50（5）： 187-195.
	CHEN Juan， CHEN Liangyong， WANG Shengsheng， et al. Pest image recognition of garden based on improved residual network ［J］. Transactions of the Chinese Society for Agricultural Machinery， 2019， 50（5）： 187-195.
21	姚青，姚波，吕军，等. 基于双线性注意力网络的农业灯诱害虫细粒度图像识别研究［J］. 中国农业科学， 2021， 54（21）： 4562-4572.
	YAO Qing， YAO Bo， Jun LÜ， et al. Research on fine-grained image recognition of agricultural light-trap pests based on bilinear attention network ［J］. Scientia Agricultura Sinica， 2021， 54（21）： 4562-4572.
22	曾伟辉，张文凤，陈鹏，等. 基于SCResNeSt的低分辨率水稻害虫图像识别方法［J］. 农业机械学报， 2022， 53（9）： 277-285.
	ZENG Weihui， ZHANG Wenfeng， CHEN Peng， et al. Low-resolution rice pest image recognition based on SCResNeSt ［J］. Transactions of the Chinese Society for Agricultural Machinery， 2022， 53（9）： 277-285.
23	ZHONG Y H， GAO J Y， LEI Q L， et al. A vision-based counting and recognition system for flying insects in intelligent agriculture［J］. Sensors， 2018， 18（5）： 1489.
24	PEI H R， LIU K， ZHAO X J， et al. Enhancing aphid detection framework based on ORB and convolutional neural networks ［J］. Scientific Reports， 2020， 10（1）： 18697.
25	REDMON J， DIVVALA S， GIRSHICK R， et al. You only look once： Unified， real-time object detection ［C］//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016： 779-788.
26	ZHANG S H， YANG H K， YANG C H， et al. Edge device detection of tea leaves with one bud and two leaves based on ShuffleNetv2-YOLOv5-Lite-E ［J］. Agronomy， 2023， 13（2）： 577.
27	邵延华，张铎，楚红雨，等. 基于深度学习的YOLO目标检测综述［J］. 电子与信息学报， 2022， 44（10）： 3697-3708.
	SHAO Yanhua， ZHANG Duo， CHU Hongyu， et al. A review of YOLO object detection based on deep learning ［J］. Journal of Electronics & Information Technology， 2022， 44（10）： 3697-3708.

模型	深度 /times	通道 /times	参数量 /M	复杂度 /B
YOLOv5n	0.33	0.25	1.9	4.5
YOLOv5s	0.33	0.50	7.2	16.5
YOLOv5m	0.67	0.75	21.2	49.0
YOLOv5l	1.00	1.00	46.5	109.1
YOLOv5x	1.33	1.25	86.7	205.7
YOLOv8n	0.33	0.25	3.2	8.7
YOLOv8s	0.33	0.50	11.2	28.6
YOLOv8m	0.67	0.75	25.9	78.9
YOLOv8l	1.00	1.00	43.7	165.2
YOLOv8x	1.33	1.25	68.2	257.8

模型	深度 /times	通道 /times	参数量 /M	复杂度 /B
YOLOv5n	0.33	0.25	1.9	4.5
YOLOv5s	0.33	0.50	7.2	16.5
YOLOv5m	0.67	0.75	21.2	49.0
YOLOv5l	1.00	1.00	46.5	109.1
YOLOv5x	1.33	1.25	86.7	205.7
YOLOv8n	0.33	0.25	3.2	8.7
YOLOv8s	0.33	0.50	11.2	28.6
YOLOv8m	0.67	0.75	25.9	78.9
YOLOv8l	1.00	1.00	43.7	165.2
YOLOv8x	1.33	1.25	68.2	257.8

模型	P	R	mAP50	mAP50-95
Yolov5n	0.881	0.781	0.876	0.421
Yolov5s	0.823	0.837	0.888	0.456
Yolov5m	0.859	0.865	0.901	0.484
Yolov5l	0.902	0.84	0.904	0.492
Yolov5x	0.901	0.847	0.901	0.5
Yolov8n	0.878	0.765	0.869	0.457
Yolov8s	0.856	0.821	0.89	0.47
Yolov8m	0.88	0.865	0.919	0.519
Yolov8l	0.917	0.847	0.926	0.552
Yolov8x	0.955	0.81	0.923	0.559

模型	P	R	mAP50	mAP50-95
Yolov5n	0.881	0.781	0.876	0.421
Yolov5s	0.823	0.837	0.888	0.456
Yolov5m	0.859	0.865	0.901	0.484
Yolov5l	0.902	0.84	0.904	0.492
Yolov5x	0.901	0.847	0.901	0.5
Yolov8n	0.878	0.765	0.869	0.457
Yolov8s	0.856	0.821	0.89	0.47
Yolov8m	0.88	0.865	0.919	0.519
Yolov8l	0.917	0.847	0.926	0.552
Yolov8x	0.955	0.81	0.923	0.559

	MAE	RMSE	Accuracy
活棉蚜	1.60	2.68	0.96
死棉蚜	0.94	2.22	0.93
总数	2.18	3.37	0.96