A Rapid Detection Method for Wheat Seedling Leaf Number in Complex Field Scenarios Based on Improved YOLOv8

doi:10.12133/j.smartag.SA202403019

Abstract

Abstract:

[Objective] The enumeration of wheat leaves is an essential indicator for evaluating the vegetative state of wheat and predicting its yield potential. Currently, the process of wheat leaf counting in field settings is predominantly manual, characterized by being both time-consuming and labor-intensive. Despite advancements, the efficiency and accuracy of existing automated detection and counting methodologies have yet to satisfy the stringent demands of practical agricultural applications. This study aims to develop a method for the rapid quantification of wheat leaves to refine the precision of wheat leaf tip detection. [Methods] To enhance the accuracy of wheat leaf detection, firstly, an image dataset of wheat leaves across various developmental stages—seedling, tillering, and overwintering—under two distinct lighting conditions and using visible light images sourced from both mobile devices and field camera equipmen, was constructed. Considering the robust feature extraction and multi-scale feature fusion capabilities of YOLOv8 network, the foundational architecture of the proposed model was based on the YOLOv8, to which a coordinate attention mechanism has been integrated. To expedite the model's convergence, the loss functions were optimized. Furthermore, a dedicated small object detection layer was introduced to refine the recognition of wheat leaf tips, which were typically difficult for conventional models to discern due to their small size and resemblance to background elements. This deep learning network was named as YOLOv8-CSD, tailored for the recognition of small targets such as wheat leaf tips, ascertains the leaf count by detecting the number of leaf tips present within the image. A comparative analysis was conducted on the YOLOv8-CSD model in comparison with the original YOLOv8 and six other prominent network architectures, including Faster R-CNN, Mask R-CNN, YOLOv7, and SSD, within a uniform training framework, to evaluate the model's effectiveness. In parallel, the performance of both the original and YOLOv8-CSD models was assessed under challenging conditions, such as the presence of weeds, occlusions, and fluctuating lighting, to emulate complex real-world scenarios. Ultimately, the YOLOv8-CSD model was deployed for wheat leaf number detection in intricate field conditions to confirm its practical applicability and generalization potential. [Results and Discussions] The research presented a methodology that achieved a recognition precision of 91.6% and an mAP_0.5 of 85.1% for wheat leaf tips, indicative of its robust detection capabilities. This method exceled in adaptability within complex field environments, featuring an autonomous adjustment mechanism for different lighting conditions, which significantly enhanced the model's robustness. The minimal rate of missed detections in wheat seedlings' leaf counting underscored the method's suitability for wheat leaf tip recognition in intricate field scenarios, consequently elevating the precision of wheat leaf number detection. The sophisticated algorithm embedded within this model had demonstrated a heightened capacity to discern and focus on the unique features of wheat leaf tips during the detection process. This capability was essential for overcoming challenges such as small target sizes, similar background textures, and the intricacies of feature extraction. The model's consistent performance across diverse conditions, including scenarios with weeds, occlusions, and fluctuating lighting, further substantiated its robustness and its readiness for real-world application. [Conclusions] This research offers a valuable reference for accurately detecting wheat leaf numbers in intricate field conditions, as well as robust technical support for the comprehensive and high-quality assessment of wheat growth.

Key words: wheat leaves, tip detection, leaf counting, attention mechanism, YOLOv8, deep learning

HOU Yiting, RAO Yuan, SONG He, NIE Zhenjun, WANG Tan, HE Haoxu. A Rapid Detection Method for Wheat Seedling Leaf Number in Complex Field Scenarios Based on Improved YOLOv8[J]. Smart Agriculture, 2024, 6(4): 128-137.

Figures/Tables 15

Fig.1

Table 1

Fig.2

Table 2

Fig.3

Table 3

Table 4

Fig. 4

Table 5

Fig. 5

Fig. 6

Table 6

Fig. 7

Fig.8

Table 7

References 26

1	JIN Y C, LIU J Z, XU Z J, et al. Development status and trend of agricultural robot technology[J]. International journal of agricultural and biological engineering, 2021, 14(3): 1-19.
2	CHEN P P, LI Y, LIU X J, et al. Improving yield prediction based on spatio-temporal deep learning approaches for winter wheat: A case study in Jiangsu province, China[J]. Computers and electronics in agriculture, 2023, 213: ID 108201.
3	申雪懿, 李东升, 陈琛, 等. 小麦分蘖数目遗传研究进展与展望[J]. 麦类作物学报, 2023, 43(10): 1344-1350.
	SHEN X Y, LI D S, CHEN C, et al. Progress and prospect of genetic research on tiller number in wheat[J]. Journal of triticeae crops, 2023, 43(10): 1344-1350.
4	WANG C Y, BOCHKOVSKIY A, LIAO H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, New Jersey, USA: IEEE, 2023: 7464-7475.
5	ZOU Z X, CHEN K Y, SHI Z W, et al. Object detection in 20 years: A survey[J]. Proceedings of the IEEE, 2023, 111(3): 257-276.
6	HE Y, XIAO L. Structured pruning for deep convolutional neural networks: A survey[J]. IEEE transactions on pattern analysis & machine intelligence, 2024, 46(5): 2900-2919.
7	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, New Jersey, USA: IEEE, 2016: 770-778.
8	LU X C, JI J, XING Z Q, et al. Attention and feature fusion SSD for remote sensing object detection[J]. IEEE transactions on instrumentation measurement, 2021, 70: ID 3052575.
9	REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 39(6): 1137-1149.
10	CHEN W B, LIU M C, ZHAO C J, et al. MTD-YOLO: Multi-task deep convolutional neural network for cherry tomato fruit bunch maturity detection[J]. Computers and electronics in agriculture, 2024, 216: ID 108533..
11	XU X M, WANG L, LIANG X W, et al. Maize Seedling Leave Counting Based on Semi-Supervised Learning and UAV RGB Images[J]. Sustainability, 2023, 15(12): ID 9583.
12	DEB M, DHAL K G, DAS A, et al. A CNN-based model to count the leaves of rosette plants (LC-Net)[J]. Scientific reports, 2024, 14: ID 1496.
13	PRAVEEN KUMAR J, DOMNIC S. Image based leaf segmentation and counting in rosette plants[J]. Information processing in agriculture, 2019, 6(2): 233-246.
14	YIGIT E, SABANCI K, TOKTAS A, et al. A study on visual features of leaves in plant identification using artificial intelligence techniques[J]. Computers and electronics in agriculture, 2019, 156: 369-377.
15	XIE X, GE Y, WALIA H, et al. Leaf-counting in monocot plants using deep regression models[J]. Sensors (basel), 2023, 23(4): ID 1890.
16	KOLHAR S, JAGTAP J. Leaf segmentation and Counting for Phenotyping of Rosette plants using xception-style U-net and Watershed algorithm[M]// Communications in Computer and Information Science. Cham: Springer International Publishing, 2022: 139-150.
17	REIS D, KUPEC J, HONG J, et al. Real-time flying object detection with YOLOv8[EB/OL]. arXiv: 2305.09972, 2023.
18	ELFWING S, UCHIBE E, DOYA K. Sigmoid-weighted linear units for neural network function approximation in reinforcement learning[J]. Neural netw, 2018, 107: 3-11.
19	BRAUWERS G, FRASINCAR F. A general survey on attention mechanisms in deep learning[J]. IEEE transactions on knowledge and data engineering, 2023, 35(4): 3279-3298.
20	ZHU X Z, CHENG D Z, ZHANG Z, et al. An empirical study of spatial attention mechanisms in deep networks[C]// 2019 IEEE/CVF International Conference on Computer Vision (ICCV). Piscataway, New Jersey, USA: IEEE, 2019: 6688-6697.
21	HOU Q B, ZHOU D Q, FENG J S. Coordinate attention for efficient mobile network design[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, New Jersey, USA: IEEE, 2021: 13713-13722.
22	KUNC V, KLÉMA J. Three decades of activations: a comprehensive survey of 400 activation functions for neural networks[EB/OL]. arXiv: 2402.09092, 2024.
23	ZHENG Z, WANG P, REN D, et al. Enhancing geometric factors in model learning and inference for object detection and instance segmentation[J]. IEEE transactions on cybernetics, 2022, 52(8): 8574-8586.
24	PENG H, YU S. A systematic IoU-related method: Beyond simplified regression for better localization[J]. IEEE transactions on image processing, 2021, 30: 5032-5044.
25	HE K M, GKIOXARI G, DOLLAR P, et al. Mask R-CNN[C]// 2017 IEEE International Conference on Computer Vision (ICCV). Piscataway, New Jersey, USA: IEEE, 2017: 2961-2969.
26	张阳婷, 黄德启, 王东伟, 等. 基于深度学习的目标检测算法研究与应用综述[J]. 计算机工程与应用, 2023, 59(18): 1-13.
	ZHANG Y T, HUANG D Q, WANG D W, et al. Review on research and application of deep learning-based target detection algorithms[J]. Computer engineering and applications, 2023, 59(18): 1-13.

数据来源	不同生长期小麦图像数量/张			不同光照条件小麦图像数量/张
数据来源	出苗期	分蘖期	越冬期	晴天	阴天
手机	187	323	254	354	410
田间摄像头	—	62	44	48	58
总数	187	385	298	402	468

特征图/像素	20×20	40×40	80×80	160×160
感受野	大	中	较小	小
锚框	（116，90）（156，198）（373，326）	（30，61）（62，45）（59，119）	（10，13）（16，30）（33，23）	（5，7）（9，13）（11，15）

序号	添加CA注意力机制	替换SIoU损失函数	添加小目标检测层	P/%	R/%	mAP_0.5/%
1	×	×	×	89.3	73.6	81.8
2	√	×	×	90.2	74.5	83.2
3	×	√	×	89.7	74.1	82.2
4	×	×	√	90.0	76.8	84.9
5	√	√	√	91.6	80.0	85.1

模型	P/%	R/%	mAP_0.5/%	模型大小/MB
Faster R-CNN	70.9	50.4	58.7	108.2
Mask R-CNN	79.3	54.8	61.6	244.0
YOLOv7	87.2	62.5	71.8	74.8
YOLOv8	89.3	73.6	81.8	49.7
SSD	85.6	61.9	71.5	78.1
YOLOv8-CSD	91.6	80.0	85.1	52.2

图像采集设备	P/%	R/%	mAP_0.5/%
田间摄像头	90.18	78.16	82.43
手机	92.43	83.51	86.78