Design and experiment of cracked egg sorting robot based on machine vision and YOLO v5

doi:10.12398/j.issn.2096-7217.2025.01.004

Abstract

Abstract:

With the improvement of our national living standard， consumers have higher requirements for the quality of eggs. The detection of cracked eggs is an important step before packing eggs. In order to solve the problems of high labor intensity and heavy workload in manual sorting of cracked eggs， a cracked egg sorting robot based on machine vision and YOLO v5 was designed. Firstly， the rotation Angle of each steering gear of the manipulator was obtained by inverse kinematics and converted into PWM duty cycle to realize the control of the three-axis series manipulator. Secondly， a high-definition 120° wide-angle camera was used as the image acquisition core to quickly acquire the image information of egg surface and label the 1 000 images collected. YOLO v5 models with different gradient descent batch sizes were then trained respectively，among which the model with gradient descent batch size of 8 had the highest mAP with a value of 98.92%. Finally， the model was called on the main board of the upper computer of the mechanical arm， and after the recognition and judgment of normal eggs and cracked eggs， the sorting stroke of the mechanical arm was started. In addition， the end pickup mechanism of the egg sorting robot was a pneumatic suction cup， which was mounted on the mechanical arm to achieve non-destructive absorption of eggs. The test results showed that the robot can identify the two kinds of eggs with the accuracy of 93.33% and 99.17% respectively， the average success rate of sorting is 94.34%， and the average sorting rate was 7.55 s/egg， which basically met the requirements. The research results can provide technical support for the screening of cracked eggs， and provide solutions for crack detection and sorting of eggs， which has high practical significance.

Key words: cracked egg, YOLO v5, machine vision, three-axis mechanical arm, intelligent sorting

CLC Number:

CAI Jiayi, LIU Shiwei, SHAN Longxiang, LIU Yong, SHEN Hongyi, WANG Qiaohua. Design and experiment of cracked egg sorting robot based on machine vision and YOLO v5[J]. Journal of Intelligent Agricultural Mechanization, 2025, 6(1): 41-50.

Figures/Tables 21

Figure 1 Structure of the egg sorting robot system

Figure 2 Block diagram of the structure of the egg sorting robot

Table 1 Parameters of bus motor HTS-35H

指标名称	参数
工作电压/V	9~12.6
转动速度/（sec/60°）	0.18(11.1 V)
堵转扭矩/(kg·cm)	35(11.1 V)
转动范围/(°)	0~240
舵机精度/(°)	0.2
控制方式	UART串口指令

Figure 3 Schematic diagram of the left side structure of the robot

Figure 4 Simplified diagram of inverse kinematic geometric analysis

Figure 5 Schematic diagram of the right side structure of the robot arm

Figure 6 Schematic diagram of geometric analysis of the robot arm

Figure 7 Structure diagram of the robot arm overlooking

Figure 8 Flow chart of egg target detection

Figure 9 YOLO v5 model structure

Figure 10 Label interface of Labelimg

Figure 11 Flow chart of model training

Table 2 mAPs of different gradient descent Batch_size models

Batch_size	2	4	8
mAP/%	95.68	98.64	98.92

Figure 12 Confusion matrix

Figure 13 F1 score curves

Figure 14 Diagram of various loss functions

Figure 15 Device of identification accuracy test experiment

Figure 16 A comparison of egg recognition results

Table 3 Test results of cracked eggs recognition accuracy

蛋品种类	环境光强度	数量 $×$ 试验次数	正确识别数/个	识别准确率/%	漏检率/%	误检率/%
裂纹蛋	较强	15×8	107	89.17	1.67	9.17
裂纹蛋	较弱	15×8	112	93.33	1.67	5.00
正常蛋	较强	15×8	115	95.83	3.33	0.83
正常蛋	较弱	15×8	119	99.17	0.83	0

Table 3 Test results of cracked eggs recognition accuracy

蛋品种类	环境光强度	数量 $×$ 试验次数	正确识别数/个	识别准确率/%	漏检率/%	误检率/%
裂纹蛋	较强	15×8	107	89.17	1.67	9.17
裂纹蛋	较弱	15×8	112	93.33	1.67	5.00
正常蛋	较强	15×8	115	95.83	3.33	0.83
正常蛋	较弱	15×8	119	99.17	0.83	0

Figure 17 Test environment of sorting performance

Table 4 Test results of cracked eggs sorting performance

蛋品种类

数量 $×$

试验次数

Table 4 Test results of cracked eggs sorting performance

蛋品种类

数量 $×$

试验次数

蛋品分拣速率/(s·个^-1)

分拣成功率/%

蛋品漏拣率/%

蛋品误拣率/%

裂纹蛋

15×10

7.65

92.00

3.33

4.67

正常蛋

15×10

7.45

96.67

2.00

1.33

总平均值

7.55

94.34

2.67

3.00

References 20

1	赵芳，郝莉花，吕尊严，等. 2019—2021年全国食品安全监督抽检中鲜蛋抽检情况及分析［J］. 保鲜与加工， 2023， 23（6）： 67-72.
	ZHAO Fang， HAO Lihua， Zunyan LÜ， et al. Sampling inspection and analysis of fresh eggs in the national food safety supervision and sampling inspection from 2019 to 2021 ［J］. Storage and Process， 2023， 23（6）： 67-72.
2	关心慧，郑君杰，方芳，等. 基于抽检数据的北京市鸡蛋安全状况分析与对策探析［J］. 中国畜牧杂志， 2023， 59（6）： 321-325.
3	梁丹，李平，梁冬泰，等. 基于机器视觉的鸡蛋内外品质一体化检测与分级系统［J］. 中国食品学报， 2020， 20（11）： 247-254.
	LIANG Dan， LI Ping， LIANG Dongtai， et al. Integrated inspecting and grading system of the egg quality based on machine vision ［J］. Journal of Chinese Institute of Food Science and Technology， 2020， 20（11）： 247-254.
4	熊利荣，王树才. 基于机器视觉的鸭蛋裂纹自动检测［J］. 湖北农业科学， 2009， 48（12）： 3152-3154.
	XIONG Lirong， WANG Shucai. Automatic detection of duck egg cracks based on machine vision ［J］. Hubei Agricultural Sciences， 2009， 48（12）： 3152-3154.
5	潘磊庆，屠康，苏子鹏，等. 基于计算机视觉和神经网络检测鸡蛋裂纹的研究［J］. 农业工程学报， 2007， 23（5）： 154-158.
	PAN Leiqing， TU Kang， SU Zipeng， et al. Crack detection in eggs using computer vision and BP neural network ［J］. Transactions of the Chinese Society of Agricultural Engineering， 2007， 23（5）： 154-158.
6	李竞. 基于机器视觉的鸭蛋蛋壳检测系统［D］. 长沙：中南大学， 2013.
	LI Jing. System for duck-egg shell crack detection based on machine vision ［D］. Changsha： Central South University， 2013.
7	肖旺. 基于深度学习的禽蛋缺陷识别与检测研究［D］. 广州：广东工业大学， 2022.
	XIAO Wang. Research on egg defect recognition and detection based on deep learning ［D］.Guangzhou： Guangdong University of Technology， 2022.
8	ELSTER R T， GOODRUM J W. Detection of cracks in eggs using machine vision ［J］. Acta Crystallographica， 1991， 34（1）： 307-312.
9	OMID M， SOLTANI M， DEHROUYEH M H， et al. An expert egg grading system based on machine vision and artificial intelligence techniques ［J］. Journal of Food Engineering， 2013， 118（1）： 70-77.
10	NASIRI A， OMID M， TAHERI-GARAVAND A. An automatic sorting system for unwashed eggs using deep learning ［J］. Journal of Food Engineering， 2020， 283（1）： 110036.
11	POURREZA H R， POURREZA R S， FAZELI S， et al. Automatic detection of eggshell defects based on machine vision ［J］. Journal of Animal and Veterinary Advances， 2008， 7（10）： 1200-1203.
12	ARIVAZHAGAN S， SHEBIAH R N， SUDHARSAN H， et al. External and internal defect detection of egg using machine vision ［J］. Journal of Emerging Trends in Computing and Information Sciences， 2013， 4（3）： 257-262.
13	TURKOGLU M. Defective egg detection based on deep features and Bidirectional Long-Short-Term-Memory ［J］. Computers and Electronics in Agriculture， 2021， 185： 106152.
14	陈明猷，罗陆锋，刘威，等. 采摘机器人全果园视觉感知及自主作业综述［J］. 智慧农业（中英文）， 2024， 6（5）： 20-39.
	CHEN Mingyou， LUO Lufeng， LIU Wei， et al. Orchard-wide visual perception and autonomous operation of fruit picking robots： A review ［J］. Smart Agriculture， 2024， 6（5）： 20-39.
15	李培艺，汪小旵，王延鑫，等. 基于LiDAR多维点云优化的垄作菊花采摘机器人自主导航方法研究［J］. 南京农业大学学报， 2024， 47（4）： 809-822.
	LI Peiyi， WANG Xiaochan， WANG Yanxin， et al. Autonomous navigation system for chrysanthemum picking robot based on LiDAR multidimensional point cloud optimization ［J］． Journal of Nanjing Agricultural University， 2024， 47（4）： 809-822．
16	冷舒，吴克，居鹤华. 机械臂运动学建模及解算方法综述［J］. 宇航学报， 2019， 40（11）： 1262-1273.
	LENG Shu， WU Ke， JU Hehua. Overview of manipulator kinematics modeling and solving method ［J］. Journal of Astronautics， 2019， 40（11）： 1262-1273.
17	赵一名，沈明霞，刘龙申，等. 基于改进YOLO v5s和图像融合的笼养鸡死鸡检测方法研究［J］. 南京农业大学学报， 2024， 47（2）： 369-382.
	ZHAO Yiming， SHEN Mingxia， LIU Longshen， et al. study on the method of detecting dead chickens in caged chicken based on improved YOLO v5s and image fusion ［J］. Journal of Nanjing Agricultural University， 2024， 47（2）： 369-382.
18	REDMON J， DIVVALA S， GIRSHICK R， et al. You only look once： Unified， real-time object detection ［C］∥IEEE Conference on Computer Vision and Patten Recognition. Las Vegas： IEEE， 2016： 779-788.
19	商枫楠，周学成，梁英凯，等. 基于改进YOLOX的自然环境中火龙果检测方法［J］. 智慧农业（中英文）， 2022， 4（3）： 120-131.
	SHANG Fengnan， ZHOU Xuecheng， LIANG Yingkai， et al. Detection method for dragon fruit in natural environment based on improved YOLOX ［J］. Smart Agriculture， 2022， 4（3）： 120-131.
20	祝志慧，何昱廷，李沃霖，等. 基于改进YOLOv7模型的复杂环境下鸭蛋识别定位［J］. 农业工程学报， 2023， 39（11）： 274-285.
	ZHU Zhihui， HE Yuting， LI Wolin， et al. Improved YOLOv7 model for duck egg recognition and localization in complex environments ［J］. Transactions of the Chinese Society of Agricultural Engineering， 2023， 39（11）： 274-285.

[1]	CHEN Xu, HE Xun, ZHU Chenhui, WANG Baoshan, WANG Wanzhang. Research status and development trend of feed quantity detection technology for grain combine harvester [J]. Journal of Intelligent Agricultural Mechanization, 2025, 6(1): 71-80.
[2]	BAI Zhenwei, YAN Fuwei, YUAN Peihai, ZHANG Haoming, YANG Runshu. Design and experiment of fruit sorting robot based on embedded machine vision [J]. Journal of Intelligent Agricultural Mechanization, 2023, 4(3): 61-70.
[3]	CHENG Zhen, SONG Jingling, XU Hongcen, ZHANG Mingliang. Overview of intelligent orientation technology of garlic seed based on machine vision [J]. Journal of Intelligent Agricultural Mechanization, 2023, 4(2): 63-70.
[4]	Zhijun Meng, Bingxin Yan, Yanxin Yin, Qiao Wang, Hui Liu, Lin Ling. Machine vision obstacle detection method in #br# unstructured farmland environment [J]. Journal of Intelligent Agricultural Mechanization, 2021, 2(2): 1-6.