基于机器视觉与BA-BP的苹果分级系统研究

doi:10.13304/j.nykjdb.2023.0506

摘要/Abstract

摘要：

为实现水果的精确分级，以苹果为分拣对象，设计了基于机器视觉与BA-BP的苹果分级系统。首先，对实时采集的苹果图像进行预处理，得到轮廓图像，采用改进的Canny边缘检测算法提取苹果轮廓，使用最小外接圆法、颜色模型转换和灰度共生矩阵等方法提取苹果果径、色泽度、圆形度和纹理特征；其次，对采集的训练组数据进行滤波和归一化处理，将处理好的数据输入到BP神经网络模型中，再利用蝙蝠算法对BP网络模型进行优化，完成网络模型的训练；最后，将测试组数据分别输入到完成训练的BA-BP神经网络系统和BP神经网络系统中。结果表明，BA-BP神经网络系统识别准确率达到96%，性能明显优于BP神经网络系统，平均分级时间在1.25 s以内。因此，该系统满足实际生产中对于苹果分级的需求，有助于实现对于苹果品级的准确识别。

关键词: 水果分级, 机器视觉, 蝙蝠算法, BP神经网络

Abstract:

In order to achieve accurate fruit classification， an apple classification system based on machine vision and BA-BP was designed with apple as the sorting object. Firstly， the apple image acquired in real time was preprocessed to obtain the apple contour image. The improved Canny edge detection algorithm was used to extract the apple contour， and methods such as minimum circumferential circle method， color model conversion and gray scale co-existence matrix were used to extract the apple diameter， color， roundness and texture features. Secondly， the collected training group data was filtered and normalized， and the processed data was input into the BP neural network model. Then the bat algorithm was used to optimize the BP network model and complete the network model training. Finally， the data of the test group were input into the BA-BP neural network system and the BP neural network system. The experimental results showed that the recognition accuracy of BA-BP neural network system was 96%， the performance was obviously better than BP neural network system， and the average classification time was less than 1.25 s. Therefore， the system could meet the demand of apple grading in actual production and help to realize the accurate identification of apple grade.

Key words: fruit grading, machine vision, bat algorithm, BP neural network

中图分类号:

TP273

刘佳浩, 高军伟. 基于机器视觉与BA-BP的苹果分级系统研究[J]. 中国农业科技导报, 2024, 26(11): 117-125.

Jiahao LIU, Junwei GAO. Research on Apple Grading System Based on Machine Vision and BA-BP[J]. Journal of Agricultural Science and Technology, 2024, 26(11): 117-125.

图/表 12

图1 系统硬件结构

Fig. 1 System hardware structure

图2 图像采集系统流程

Fig. 2 Flow of image acquisition system

表1 苹果品质分级标准

Table 1 Apple quality grading standards

参数

Parameter

一等果

First class fruit

二等果

Second class fruit

三等果

Third class fruit

等外果

Off-grade fruit

果径

Fruit diameter/mm

色泽度

tincture

圆形度

circularity

图3 图像预处理

Fig. 3 Image preprocessing

图4 苹果边缘检测结果

Fig. 4 Apple edge detection result

图5 苹果的最小外接圆图像

Fig. 5 Smallest circumscribed circle image of an apple

图6 苹果红色区域的分割效果图像

Fig. 6 Segmentation effect image of apple red area

图7 样本圆度值

Fig. 7 Roundness value of sample

表2 苹果纹理特征统计结果

Table 2 Statistical results of apple texture characteristics

样本编号 Sample number	角二阶矩 ASM	对比度 CON	熵 ENT	反差分矩阵 IDM
1	0.70	0.47	9.64	0.91
2	0.68	0.46	9.01	0.86
3	0.68	0.41	8.21	0.84
4	0.60	0.42	7.32	0.76
5	0.56	0.38	6.54	0.77
6	0.50	0.33	5.21	0.64
7	0.49	0.26	4.32	0.62
8	0.32	0.05	2.11	0.55

图8 网络误差曲线

Fig. 8 Network error curve

图9 BP神经网络分类结果

Fig. 9 Classification results of BP neural network

图10 BA-BP神经网络分类结果

Fig. 10 BA-BP neural network classification result

参考文献 16

1	李龙,彭彦昆,李永玉,等.基于纹理和梯度特征的苹果伤痕与果梗/花萼在线识别[J].农业机械学报,2018,49(11):328-335.
	LI L, PENG Y K, LI Y Y, et al.. Online identification of apple scarring and stems /calyxes based on texture and edge gradient features [J]. Trans. Chin. Soc. Agric. Mach., 2018,49(11):328-335.
2	ABDULLAH M Z, MOHAMAD-SALEH J, FATHINUL-SYAHIR A S, et al.. Discrimination and classification of fresh-cut starfruits (Averrhoa carambola L.) using automated machine vision system [J]. J. Food Eng., 2005, 76(4):506-523.
3	唐晓彬,张瑞,刘立新.基于蝙蝠算法SVR模型的北京市二手房价预测研究[J].统计研究, 2018, 35(11):71-81.
	TANG X B, ZHANG R, LIU L X. Research on forecast of second-hand house price in Beijing based on SVR model of bat algorithm [J]. Stat. Res., 2018, 35 (11): 71-81.
4	黎成.新型元启发式蝙蝠算法[J].电脑知识与技术,2010,6(23):6569-6572.
	LI C. A new metaheuristic bat-inspired algorithm [J]. Computer Knowledge Technol., 2010,6(23):6569-6572.
5	韩天雨,路长厚,李建美,等.利用图像处理技术测量丝杠螺距的机器视觉系统[J].山东大学学报(工学版), 2022, 52(3):80-85.
	HAN T Y, LU C H, LI J M, et al.. A machine vision system for measuring screw pitch with image processing techniques [J] J. Shandong Univ. (Eng.), 2022,52 (3): 80-85.
6	孙晓杰,梁义.基于双阈值的图像边缘实时检测系统[J].仪表技术与传感器,2019, 439(8):83-86, 116.
	SUN X J, LIANG Y. Real time image edge detection system based on double threshold [J]. Instrument Technol. Sens., 2019, 439 (8): 83-86, 116.
7	刘新庭,匡迎春,陈熵,等.基于最小外接圆直径的苹果分级研究[J].中国农学通报, 2015, 31(20): 239-243.
	LIU X T, KUANG Y C, CHEN S, et al.. Research on apple grading based on the minimum circumscribed circle diameter [J] China Agric. Sci. Bull., 2015, 31 (20): 239-243.
8	石瑞瑶,田有文,赖兴涛,等.基于机器视觉的苹果品质在线分级检测[J].中国农业科技导报,2018,20(3):80-86.
	SHI R Y, TIAN Y W, LAI X T, et al.. Development of apple intelligent on-line inspection and classification system based on machine vision [J]. J. Agric. Sci. Technol., 2018,20 (3): 80-86.
9	谢忠红,郭小清,姬长英.基于颜色模型和纹理特征的彩色水果图像分割方法[J].计算机工程与科学, 2010, 32(1): 64-66, 100.
	XIE Z H, GUO X Q, JI C Y. A method of color fruit image segmentation based on the color model and the texture feature [J]. Computer Eng. Sci., 2010, 32 (1): 64-66, 100.
10	罗昔联,汪怡珂,韩霁昌,等.实验年限对砒砂岩改良风沙土微观结构的影响[J].西安交通大学学报, 2021, 55(12):172-179.
	LUO X L, WANG Y K, HAN J C, et al.. Influence of planting years on microstructure of sandy soil amended by soft rock [J]. J. Xi’an Jiaotong Univ., 2021, 55 (12): 172-179.
11	毕明德,孙志刚,李叶松.基于机器视觉的布匹疵点检测系统[J].仪表技术与传感器,2012, 359(12):37-39, 125.
	BI M D, SUN Z G, LI Y S. Fabric defect detection system base on machine vision [J]. Instrument Technol. Sens., 2012, 359 (12): 37-39, 125.
12	刘国欣.Kolmogorov三级数定理的推广[J].数学物理学报,1995(3):311-314.
	LIU G X. A generalization of Kolmogorov’s third-order number theorem [J]. Acta Mathemat. Sci.,1995(3):311-314.
13	焦敬品,李勇强,吴斌,等.基于BP神经网络的管道泄漏声信号识别方法研究[J].仪器仪表学报,2016,37(11):2588-2596.
	JIAO J P, LI Y Q, WU B, et al.. Research on acoustic signal recognition method for pipeline leakage with BP neural network [J]. Chin. J. Sci. Instrument, 2016,37(11):2588-2596.
14	王宇轲.基于BA-BP算法的汽车配件需求预测系统研究与实现[D].成都:西南交通大学, 2017.
	WANG Y K.Research and implementation of automotive parts demand prediction system based on BA-BP algorithm [D]. Chengdu: Southwest Jiaotong University, 2017.
15	周磊,张杰.基于BA-BP开关磁阻电机电流分段建模研究[J].中国农机化学报, 2016, 37(8):136-138, 154.
	ZHOU L, ZHANG J. Segmentation current modeling of switched reluctance motor based on BP neural network optimized by bat algorithm [J]. J. Chin. Agric. Mechanization, 2016, 37 (8): 136-138, 154.
16	路永华. LIBS结合GA-BP神经网络与GA-SVM对塑料分类研究[J]. 激光与红外, 2022, 52(2):273-279.
	LU Y H. Plastic classification research by LIBS combined with GA-BP-NN and GA-SVM [J]. Laser Infrared, 2022, 52 (2): 273-279.

[1]	胡国玉, 董娅兰, 古丽巴哈尔·托乎提, 刘广, 周建平. 基于机器视觉的葡萄藤结构分割方法研究[J]. 中国农业科技导报, 2024, 26(9): 105-111.
[2]	李锐风, 杨云福, 杨永发, 于永顺. 基于机器视觉的玫瑰花检测与特征提取[J]. 中国农业科技导报, 2024, 26(4): 106-113.
[3]	王辉, 付虹雨, 岳云开, 崔国贤, 佘玮. 基于气候变量的苎麻产量SSA-BP预测模型[J]. 中国农业科技导报, 2024, 26(1): 110-118.
[4]	刘朝阳, 王永强, 周聪玲, 强斯祺. 室外池塘养殖投饵系统智能控制方法研究[J]. 中国农业科技导报, 2023, 25(5): 123-130.
[5]	许鑫, 马兆务, 熊淑萍, 马新明, 程涛, 李海洋, 赵锦鹏. 基于气候年型的河南省冬小麦产量预测[J]. 中国农业科技导报, 2022, 24(2): 136-144.
[6]	黄昊, 谢圣桥, 陈度, 王恒. 深度学习在苹果产业链中的应用与研究进展[J]. 中国农业科技导报, 2022, 24(10): 79-89.
[7]	王传旭, 王康, 陈林, 李学, 张红伟. 基于Kriging插值和BP神经网络结合的粮仓温度场预测模型研究及实现[J]. 中国农业科技导报, 2021, 23(9): 96-102.
[8]	陈昊宇, 杨光, 韩雪莹, 刘昕, 刘峰, 王宁. 基于连续小波变换的土壤有机质含量高光谱反演[J]. 中国农业科技导报, 2021, 23(5): 132-142.
[9]	于洋, 张珏, 田海清, 王迪, 王轲, 张红旗. 基于高光谱成像技术的冷鲜羊肉嫩度检测方法研究[J]. 中国农业科技导报, 2021, 23(12): 101-108.
[10]	姚应方, 刘峰, 张海东, 李超, 周海俊, 王满. 基于机器视觉的八角颜色与果形识别研究[J]. 中国农业科技导报, 2021, 23(11): 110-120.
[11]	罗建军1,杨红云2*,路艳1,易文龙2,孙爱珍1. 基于遗传算法优化的BP神经网络进行水稻氮素营养诊断[J]. 中国农业科技导报, 2020, 22(8): 83-92.
[12]	陈玉玲1,吴保国1*,崔岩2,魏彦军3. 基于BP神经网络的华北落叶松小班蓄积预估模型研究与应用[J]. 中国农业科技导报, 2019, 21(7): 82-93.
[13]	刘瑾蓉,林剑辉*,李婷婷. 基于卷积神经网络的银杏叶片患病程度识别[J]. 中国农业科技导报, 2018, 20(6): 55-61.
[14]	石瑞瑶,田有文*,赖兴涛,古文君. 基于机器视觉的苹果品质在线分级检测[J]. 中国农业科技导报, 2018, 20(3): 80-86.
[15]	王诣,闫志勇*. 基于图像处理的青冈栎叶绿素含量检测系统研究[J]. 中国农业科技导报, 2017, 19(4): 59-64.