Grading Asparagus officinalis L. Using Improved YOLOv11

doi:10.12133/j.smartag.SA202501024

Abstract

Abstract:

[Objective] Asparagus officinalis L. is a perennial plant with a long harvesting cycle and fast growth rate. The harvesting period of tender stems is relatively concentrated, and the shelf life of tender stems is very short. Therefore, the harvested asparagus needs to be classified according to the specifications of asparagus in a short time and then packaged and sold. However, at this stage, the classification of asparagus specifications basically depends on manual work, and it is difficult for asparagus of different specifications to rely on sensory grading, which requires a lot of money and labor. To save labor costs, an algorithm based on asparagus stem diameter classification was developed using deep learning and computer vision technology. This method selected YOLOv11 as the baseline model and makes several improvements, aiming to study a lightweight model for accurate grading of post-harvest asparagus. [Methods] This dataset was obtained by cell phone photography of post-harvest asparagus using fixed camera positions. In order to improve the generalization ability of the model, the training set was augmented with data by increasing contrast, mirroring, and adjusting brightness. The data-enhanced training set includes a total of 2 160 images for training the model. And the test set and validation set include 90 and 540 images respectively for inference and validation of the model. In order to enhance the performance of the improved model, the following four improvements were made to the baseline model, respectively. First, the efficient channel attention (ECA) module was added to the twelfth layer of the YOLOv11 backbone network. The ECA enhanced asparagus stem diameter feature extraction by dynamically adjusting channel weights in the convolutional neural network and improved the recognition accuracy of the improved model. Second, the bi-directional feature pyramid network (BiFPN) module was integrated into the neck network. This module modified the original feature fusion method to automatically emphasize key asparagus features and improved the grading accuracy through multi-scale feature fusion. What's more, BiFPN dynamically adjusted the importance of each layer to reduce redundant computations. Next, the slim-neck module was applied to optimize the neck network. The slim-neck Module consisted of GSConv and VOVGSCSP. The GSConv module replaced the traditional convolutional. And the VOVGSCSP module replaced the C2k3 module. This optimization reduced computational costs and model size while improving the recognition accuracy. Finally, the original YOLOv11 detection head was replaced with an EfficientDet Head. EfficientDet Head had the advantages of light weight and high accuracy. This head co-training with BiFPN to enhance the effect of multi-scale fusion and improve the performance of the model. [Results and Discussions] In order to verify the validity of the individual modules introduced in the improved YOLOv11 model and the superiority of the performance of the improved model, ablation experiments and comparison experiments were conducted respectively. The results of the comparison test between different attentional mechanisms added to the baseline model showed that the ECA module had better performance than other attentional mechanisms in the post-harvest asparagus grading task. The YOLOv11-ECA had higher recognition accuracy and smaller model size, so the selection of the ECA module had a certain degree of reliability. Ablation experiments demonstrated that the improved YOLOv11 achieved 96.8% precision (P), 96.9% recall (R), and 92.5% mean average precision (mAP), with 4.6 GFLOPs, 1.67 × 10⁶ parameters, and a 3.6 MB model. The results of the asparagus grading test indicated that the localization frames of the improved model were more accurate and had a higher and higher confidence level. Compared with the original YOLOv11 model, the improved YOLOv11 model increased the precision, recall, and mean average precision by 2.6, 1.4, and 2.2 percentage points, respectively. And the floating-point operation, parameter quantity, and model size were reduced by 1.7 G, 9.1 × 10⁵, and 2.2 MB, respectively. Moreover, various improvements to the model could increase the accuracy of the model while ensuring that the model was light weight. In addition, the results of the comparative tests showed that the performance of the improved YOLOv11 model was better than those of SSD, YOLOv5s, YOLOv8n, YOLOv11, and YOLOv12. Overall, the improved YOLOv11 had the best overall performance, but still had some shortcomings. In terms of the real-time performance of the model, the inference speed of the improved model was not optimal, and the inference speed of the improved YOLOv11 was inferior to that of YOLOv5s and YOLOv8n. On this basis, to evaluate the inference speed of improved YOLOv11 and YOLOv11 used the aggregate test. The results of the Wilcoxon signed-rank test showed that the improved YOLOv11 had a significant improvement in inference speed compared to the original YOLOv11 model. [Conclusions] The improved YOLOv11 model demonstrated better recognition, lower parameters and floating-point operations, and smaller model size in the asparagus grading task. The improved YOLOv11 provided a theoretical foundation for intelligent post-harvest asparagus grading. Deploying the improved YOLOv11 model on asparagus grading equipment enables fast and accurate grading of post-harvest asparagus.

Key words: asparagus, object detection, image recognition, intelligent grading, YOLOv11

CLC Number:

YANG Qilang, YU Lu, LIANG Jiaping. Grading Asparagus officinalis L. Using Improved YOLOv11[J]. Smart Agriculture, doi: 10.12133/j.smartag.SA202501024.

Figures/Tables 15

Table 1

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Fig. 8

Table 2

Table 3

Fig. 9

Fig. 10

Table 4

Table 5

References 26

1	祁复蓉, 王福国, 周亚婷, 等. 设施芦笋高质高效栽培技术要点[J]. 农业科技与信息, 2024(7): 10-14.
	QI F R, WANG F G, ZHOU Y T, et al. Key points of high quality and efficient cultivation techniques of Asparagus in facilities[J]. Agricultural science-technology and information, 2024(7): 10-14.
2	CHEN Q, XIA C, SHI Y Y, et al. A novel approach for Asparagus comprehensive classification based on TOPSIS evaluation and SVM prediction[J]. Agronomy, 2024, 14(6): ID1175.
3	俞风娟, 王继涛, 汪洋, 等. 芦笋生育特性及高产栽培技术[J]. 宁夏农林科技, 2021, 62(3): 8-11.
	YU F J, WANG J T, WANG Y, et al. Growth characteristics and high-yield cultivation techniques of Asparagus [J]. Ningxia journal of agriculture and forestry science and technology, 2021, 62(3): 8-11.
4	朱德明, 程香平, 邱伊健, 等. 基于深度学习的农作物图像识别技术研究进展[J]. 江西科学, 2025, 43(1): 154-161.
	ZHU D M, CHENG X P, QIU Y J, et al. Research progress on crop image recognition alg orithm based on deep learning[J]. Jiangxi science, 2025, 43(1): 154-161.
5	张润池, 周云成, 侯玉涵, 等. 基于超深掩蔽与改进YOLOv8的不同成熟度番茄计数方法[J]. 农业工程学报, 2024, 40(24): 146-156.
	ZHANG R C, ZHOU Y C, HOU Y H, et al. Counting tomatoes with different maturities using ultra-depth masking and improved YOLOv8[J]. Transactions of the Chinese society of agricultural engineering, 2024, 40(24): 146-156.
6	袁杰, 谢霖伟, 郭旭, 等. 基于改进YOLO v7的苹果叶片病害检测方法[J]. 农业机械学报, 2024, 55(11): 68-74.
	YUAN J, XIE L W, GUO X, et al. Apple leaf disease detection method based on improved YOLO v7[J]. Transactions of the Chinese society for agricultural machinery, 2024, 55(11): 68-74.
7	杨昊霖, 王其欢, 李华彪, 等. 基于改进YOLOv5的田间复杂环境障碍物检测[J]. 中国农机化学报, 2024, 45(6): 216-222, 256, 2.
	YANG H L, WANG Q H, LI H B, et al. Obstacle detection in complex farmland environment based on improved YOLOv5[J]. Journal of Chinese agricultural mechanization, 2024, 45(6): 216-222, 256, 2.
8	李扬, 张萍, 苑进, 等. 白芦笋采收机器人视觉定位与采收路径优化方法[J]. 智慧农业(中英文), 2020, 2(4): 65-78.
	LI Y, ZHANG P, YUAN J, et al. Visual positioning and harvesting path optimization of white Asparagus harvesting robot[J]. Smart agriculture, 2020, 2(4): 65-78.
9	ZHAO X Y, HE Y X, ZHANG H T, et al. A quality grade classification method for fresh tea leaves based on an improved YOLOv8x-SPPCSPC-CBAM model[J]. Scientific reports, 2024, 14: ID 4166.
10	FAN Y R, CAI Y L, YANG H J. A detection algorithm based on improved YOLOv5 for coarse-fine variety fruits[J]. Journal of food measurement and characterization, 2024, 18(2): 1338-1354.
11	汪小旵, 李为民, 王琳, 等. 基于改进YOLACT++的成熟芦笋检测-判别-定位方法[J]. 农业机械学报, 2023, 54(7): 259-271.
	WANG X C, LI W M, WANG L, et al. Method of detection-discrimination-localization for mature Asparagus based on improved YOLACT + + algorithm[J]. Transactions of the Chinese society for agricultural machinery, 2023, 54(7): 259-271.
12	中华人民共和国农业部. 芦笋等级规格: NY/T 1585—2008 [S]. 北京: 中国农业出版社, 2008.
	Ministry of Agriculture of the PRC. Grades and specifications of asparagus: NY/T 1585—2008 [S]. BeiJing: China agriculture press, 2008.
13	KHANAM R, HUSSAIN M. YOLOv11: An overview of the key architectural enhancements[EB/OL]. arXiv: 2410.17725, 2024.
14	LIAO Y, LI L R, XIAO H Q, et al. YOLO-MECD: Citrus detection algorithm based on YOLOv11[J]. Agronomy, 2025, 15(3): ID 687.
15	WANG Q L, WU B G, ZHU P F, et al. ECA-net: Efficient channel attention for deep convolutional neural networks[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, New Jersey, USA. IEEE, 2020: 11531-11539.
16	LI L T, ZHAO Y D. Tea disease identification based on ECA attention mechanism ResNet50 network[J]. Frontiers in plant science, 2025, 16: ID1489655.
17	HE J J, WANG Y C, WANG Y T, et al. A lightweight road crack detection algorithm based on improved YOLOv7 model[J]. Signal, image and video processing, 2024, 18(1): 847-860.
18	WANG J L, QIN C C, HOU B B, et al. LCGSC-YOLO: A lightweight apple leaf diseases detection method based on LCNet and GSConv module under YOLO framework[J]. Frontiers in plant science, 2024, 15: ID1398277.
19	LI H, LI J, WEI H, et al. Slim-neck by GSConv: A better design paradigm of detector architectures for autonomous vehicles[EB/OL]. arXiv: 2206.02424, 2022.
20	SHI P, ZHANG Y Y, CAO Y Q, et al. DVCW-YOLO for printed circuit board surface defect detection[J]. Applied sciences, 2025, 15(1): ID 327.
21	张荣华, 白雪, 樊江川. 复杂场景下害虫目标检测算法: YOLOv8-Extend[J]. 智慧农业(中英文), 2024, 6(2): 49-61.
	ZHANG R H, BAI X, FAN J C. Crop pest target detection algorithm in complex scenes: YOLOv8-extend[J]. Smart agriculture, 2024, 6(2): 49-61.
22	周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.
	ZHOU Z H. Machine learning[M]. BeiJing: Tsinghua university press, 2016.
23	俞建峰. 深度学习: 智能机器人应用的理论与实践[M]. 北京: 化学工业出版社, 2024.
	YU J F. Deep learning : Theory and practice of intelligent robot applications[M]. BeiJing: Chemical industry press, 2024.
24	孙玉林. 计算机视觉从入门到进阶实战: 基于PyTorch[M]. 北京: 化学工业出版社, 2024.
	SUN Y L. Computer vision from entry to advanced practice : Based on PyTorch[M]. BeiJing: Chemical industry press, 2024.
25	YANG L, ZHANG R, LI L, et al. SimAM: A Simple, Parameter-Free Attention Module for Convolutional Neural Networks[C]// Proceedings of the International Conference on Machine Learning. New York, USA: PMLR, 2021: 11863-11874.
26	HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway, New Jersey, USA. IEEE, 2018: 7132-7141.

模型	精确率/%	召回率/%	平均精度均值/%	模型大小/MB
YOLOv11	94.2	95.5	90.3	5.2
YOLOv11-SimAM	94.6	97.0	91.0	5.5
YOLOv11-ECA	95.5	98.1	90.8	5.2
YOLOv11-SE	95.0	96.1	90.5	5.5

模型	+ECA	+BiFPN	+slim-neck	+EfficientDet Head	精确率/%	召回率/%	平均精度均值/%	模型大小/MB	浮点运算量/G	参数量×10⁶
YOLOv11	×	×	×	×	94.2	95.5	90.3	5.2	6.3	2.58
	√	×	×	×	95.5	98.1	90.8	5.2	6.3	2.58
	×	√	×	×	94.6	97.0	91.4	4.0	6.3	1.92
	×	×	√	×	95.0	97.3	90.8	5.2	5.9	2.57
	×	×	×	√	97.0	96.7	90.7	4.7	5.1	2.31
	×	√	×	√	95.1	98.0	90.7	3.9	5.2	1.73
	×	√	√	√	94.8	96.3	91.6	3.6	4.6	1.67
	√	√	√	√	96.8	96.9	92.5	3.6	4.6	1.67

模型	精确率/%	召回率/%	平均精度均值/%	模型大小/MB	参数量×106	浮点运算量/G	推理速度FPS
SSD	90.6	91.6	87.4	95.5	24.9	31.4	171.5
YOLOv5s	92.4	96.9	74.9	14.4	7.03	16.0	244.1
YOLOv8n	95.6	97.1	85.8	6.2	3.01	8.2	271.4
YOLOv11	94.2	95.5	90.3	5.8	2.58	6.3	194.1
YOLOv12	94.7	96.3	91.0	5.4	2.52	6.0	195.2
改进的YOLOv11	96.8	96.9	92.5	3.6	1.67	4.6	204.0

[1]	NIU Ziang, QIU Zhengjun. Extraction Method of Maize Plant Skeleton and Phenotypic Parameters Based on Improved YOLOv11-Pose [J]. Smart Agriculture, 2025, 7(2): 95-105.
[2]	MA Weiwei, CHEN Yue, WANG Yongmei. Recognition of Sugarcane Leaf Diseases in Complex Backgrounds Based on Deep Network Ensembles [J]. Smart Agriculture, 2025, 7(1): 136-145.
[3]	ZHOU Xiushan, WEN Luting, JIE Baifei, ZHENG Haifeng, WU Qiqi, LI Kene, LIANG Junneng, LI Yijian, WEN Jiayan, JIANG Linyuan. Real-time Detection Algorithm of Expanded Feed Image on the Water Surface Based on Improved YOLOv11 [J]. Smart Agriculture, 2024, 6(6): 155-167.
[4]	LIU Chang, SUN Yu, YANG Jing, WANG Fengchao, CHEN Jin. Grape Recognition and Localization Method Based on 3C-YOLOv8n and Depth Camera [J]. Smart Agriculture, 2024, 6(6): 121-131.
[5]	YE Dapeng, JING Jun, ZHANG Zhide, LI Huihuang, WU Haoyu, XIE Limin. MSH-YOLOv8: Mushroom Small Object Detection Method with Scale Reconstruction and Fusion [J]. Smart Agriculture, 2024, 6(5): 139-152.
[6]	SHANG Fengnan, ZHOU Xuecheng, LIANG Yingkai, XIAO Mingwei, CHEN Qiao, LUO Chendi. Detection Method for Dragon Fruit in Natural Environment Based on Improved YOLOX [J]. Smart Agriculture, 2022, 4(3): 120-131.
[7]	GUO Xiuming, ZHU Yeping, LI Shijuan, ZHANG Jie, LYU Chunyang, LIU Shengping. Scale Adaptive Small Objects Detection Method in Complex Agricultural Environment: Taking Bees as Research Object [J]. Smart Agriculture, 2022, 4(1): 140-149.
[8]	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.
[9]	LI Yang, ZHANG Ping, YUAN Jin, LIU Xuemei. Visual Positioning and Harvesting Path Optimization of White Asparagus Harvesting Robot [J]. Smart Agriculture, 2020, 2(4): 65-78.