Foxtail Millet Ear Detection Approach Based on YOLOv4 and Adaptive Anchor Box Adjustment

doi:10.12133/j.smartag.2021.3.1.202102-SA066

Abstract

Abstract:

谷穗的检测和计数对于预测谷子产量和育种至关重要。但是，传统的谷穗计数主要基于人工统计，既费时又费力。为解决上述问题，本研究首先建立了一个包含784张图像和10,000个谷穗样本的谷穗检测数据集。提出了一种基于YOLOv4和自适应锚框调整的谷穗检测方法，可快速准确地检测特定框中的谷穗。通过自适应地调整锚框，可生成符合谷穗目标的候选框，从而提升检测的准确率。为验证该方法的有效性，采用了多个标准，包括平均精度（mAP），F1得分（F1-Score），精度（Precision）和召回率（Recall）进行评价。此外，设计了对比试验验证所提出方法的有效性，包括与其他模型（YOLOv2，YOLOv3和Faster-RCNN）进行比较来评估模型的性能，评估模型在不同交并比（IOU）取值下的性能，评估模型在自适应锚框调整下的谷穗检测性能，评估引起模型评价标准变化的原因，以及评估模型在不同原始输入图像尺寸下的性能。试验结果表明，YOLOv4获得了良好的谷穗检测性能。YOLOv4的mAP达到78.99%，F1-score达到83.00%，Precision达到87%和Recall达到79.00%，在所有评价标准上均比其他比较模型高出8%。试验结果表明，该方法具有较好的准确性和高效性。

Key words: 谷穗检测, YOLOv4, 深度神经网络, 数据集, 自适应锚框调整

CLC Number:

HAO Wangli, YU Peiyan, HAO Fei, HAN Meng, HAN Jiwan, SUN Weirong, LI Fuzhong. Foxtail Millet Ear Detection Approach Based on YOLOv4 and Adaptive Anchor Box Adjustment[J]. Smart Agriculture, 2021, 3(1): 63-74.

References

1	LI Y. Millet breeding[M]. Beijing: China Agriculture Press, 1997: 22-23.
2	CHEN G. Analysis of the total contribution of millet yield components[J]. Miscellaneous Crops, 2000, 20(3): 25-26.
3	REDMON J, FARHADI A. YOLO9000: Better, faster, stronger[C]// In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, New York, USA: IEEE, 2017: 7263-7271.
4	TANG C, HU H, WEI P, et al. An improved Yolov3 algorithm to detect molting in swimming crabs against a complex background[J]. Aquacultural Engineering, 2020, 91(3): 102-115, .
5	CHENG Z, ZHANG F. Flower end-to-end detection based on YOLOv4 using a mobile device[J]. Wireless Communications and Mobile Computing, 2020(2): 1-9.
6	GONG B, ERGU D, CAI Y, et al. A method for wheat head detection based on Yolov4[EB/OL]. DOI:10.21203/rs.3.rs-86158/v1.
7	LU X. Research on wheat ear recognition based on image processing technology[D]. Shijiazhuang: Hebei Agricultural University, 2012.
8	SHI X. Detection and grading of wheat appearance quality based on image processing[D]. Zhengzhou: Henan University of Technology, 2013.
9	ZHANG Q, CHEN Y, LI Y, et al. Winter wheat ear detection and counting system based on convolutional neural network[J]. Transactions of the Chinese Society of Agricultural Machinery, 2019, 50(3): 144-150.
10	GAO Y. Research on field wheat ear detection method based on deep neural network[D]. Beijing: Beijing Forestry University, 2019.
11	ALKHUDAYDI T, ZHOU J, LGLESIA B D L. SpikeletFCN: Counting Spikelets from infield wheat crop images using fully convolutional networks[M]. Boca Raton: Artificial Intelligence and Soft Computing, 2019.
12	XIE Y, HE C, YU Z, et al. Optimization method of wheat ear detection cascade network in complex field scene[J]. Transactions of the Chinese Society of Agricultural Machinery, 2020, 51(12): 212-219.
13	TZUTA[EB/OL].[2021-2-10] .
14	KHARCHENKO V,CHYRKA I. Detection of airplanes on the ground using YOLO neural network[C]// 2018 IEEE 17th International Confer ence on Mathematical Methods in Electromagnetic Theory (MMET). Piscataway, New York, USA: IEEE, 2018: 294-297.
15	WANG C, LIAO H M, YEH I, et al. CSPNet: A New Backbone that can Enhance Learning Capability of CNN[C]// IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Piscataway, New York, USA: IEEE, 2020.
16	TAN X, WANG Z. Ping-pong table recognition based on YOLOv4 improved algorithm[J]. Science and Technology Innovation and Application, 2020(27): 74-76.
17	WANG J, WANG J, SONG J, et al. Optimized cartesian k-means[J]. Engineering IEEE Transactions on Knowledge & Data, 2015, 27(1): 180-192.
18	Microsoft. PASCAL-VOC2012[EB/OL]. (2012-02-20)[2019-08-02]. .
19	REZATOFIGHI H, TSOI N, GWAK J Y, et al. Generalized intersection over union: A metric and a loss for bounding box regression[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, New York, USA: IEEE, 2019: 658-666.

[1]	LI Yangde, MA Xiaohui, WANG Ji. Pineapple Maturity Analysis in Natural Environment Based on MobileNet V3-YOLOv4 [J]. Smart Agriculture, 2023, 5(2): 35-44.
[2]	LONG Jiehua, GUO Wenzhong, LIN Sen, WEN Chaowu, ZHANG Yu, ZHAO Chunjiang. Strawberry Growth Period Recognition Method Under Greenhouse Environment Based on Improved YOLOv4 [J]. Smart Agriculture, 2021, 3(4): 99-110.