基于透射光和反射光图像同位分割的小麦品种识别方法研究

doi:10.3969/j.issn.1004-1524.2022.03.20

浙江农业学报 ›› 2022, Vol. 34 ›› Issue (3): 590-598.DOI: 10.3969/j.issn.1004-1524.2022.03.20

基于透射光和反射光图像同位分割的小麦品种识别方法研究

闫宁¹^,²(), 张晗²^,^*(), 董宏图³, 康凯³, 罗斌²

1.中北大学机械工程学院,山西太原 030051
2.北京农业智能装备技术研究中心,北京 100094
3.北京农业信息技术研究中心,北京 100094

收稿日期:2021-07-19 出版日期:2022-03-25 发布日期:2022-03-30
作者简介:*张晗,E-mail: zhangha@nercita.org.cn
闫宁(1996—),男,山西吕梁人,硕士研究生,主要从事基于机器视觉的种子无损检测方法研究。E-mail: yanning96815@163.com
通讯作者: 张晗
基金资助:
国家重点研发计划(2017YFD0701205);北京市农林科学院青年基金(QNJJ202104);北京市农林科学院2021年度科研创新平台建设(PT2021-04)

Wheat variety recognition method based on same position segmentation of transmitted light and reflected light images

YAN Ning¹^,²(), ZHANG Han²^,^*(), DONG Hongtu³, KANG Kai³, LUO Bin²

1. School of Mechanical Engineering, North University of China, Taiyuan 030051, China
2. Beijing Research Center for Intelligent Agricultural Equipment, Beijing 100094, China
3. Beijing Agricultural Information Technology Research Center, Beijing 100094, China

Received:2021-07-19 Online:2022-03-25 Published:2022-03-30
Contact: ZHANG Han

摘要/Abstract

摘要：

为提高基于机器视觉的小麦品种识别准确性,本文通过透射光和反射光同位图像分割对种子颜色特征参数进行了优化提取。采用透射光图像辅助反射光图像分割的方式从种子图像中分割出胚部区域,并分别提取小麦整粒、种胚、胚乳区域的颜色特征参数。以济麦22、济麦44、京麦9、京麦11共4个品种种子作为研究对象,利用HALCON机器视觉软件获取种子的颜色特征参数,通过偏最小二乘法判别分析法(partial least squares discriminant analysis,PLS-DA)建立分类模型。结果表明:通过透射光图像辅助反射光图像分割后,融合的更多种子颜色参数信息,使得4个品种的小麦种子识别正确率均获得提升。济麦和京麦间混杂识别正确率从种粒反射光颜色特征的95%提升到融合了透射光、胚和胚乳颜色特征的99%以上,济麦22和济麦44混杂识别正确率从73.28%提高到84.60%,京麦9和京麦11混杂识别正确率从74.15%提高到83.73%。通过透射光特征进一步融合分析种子胚和胚乳图像所包含的颜色特征可有效提高小麦品种识别正确率。

关键词: 机器视觉, 小麦, 品种识别, 透射光, 胚部, 胚乳, 颜色特征

Abstract:

A segmentation method of embryo and endosperm image based on transmitted light and reflected light at the same position of wheat seeds was proposed in order to improve the recognition accuracy of wheat varieties. In this paper, according to the characteristics of great difference in transmissibility between embryo and endosperm of wheat seed, color feature information was extracted from reflected light images and transmitted light images for modeling and analysis, the influence of transmission light characteristics and color characteristics of seed embryo and endosperm on wheat seed variety recognition were studied. Taking the four varieties of seeds of Jimai 22, Jimai 44, Jingmai 9, and Jingmai 11 as the materials, the color characteristic parameters of the seeds were obtained by using the HALCON machine vision software, and through partial least squares discriminant analysis (PLS-DA) method to establish a classification model. The results showed that after the transmission light image was used to assist the reflected light image segmentation, more seed color parameter information was fused, so that the accuracy of the four varieties of wheat seed recognition was improved. The correct rate of hybrid recognition between Jimai and Jingmai had increased from 95% of the color characteristics of seed reflected light to more than 99% of the color characteristics of transmitted light, embryo and endosperm. The hybrid recognition accuracy rate of Jimai 22 and Jimai 44 increased from 73.28% to 84.60%, and the hybrid recognition accuracy rate of Jingmai 9 and Jingmai 11 increased from 74.15% to 83.73%. The further fusion analysis of the color features contained in the image of seed embryo and endosperm through the transmitted light features can effectively improve the accuracy of wheat variety recognition.

Key words: machine vision, wheat, species identification, transmitted light, embryo, endosperm, color features

中图分类号:

TP391.4

闫宁, 张晗, 董宏图, 康凯, 罗斌. 基于透射光和反射光图像同位分割的小麦品种识别方法研究[J]. 浙江农业学报, 2022, 34(3): 590-598.

YAN Ning, ZHANG Han, DONG Hongtu, KANG Kai, LUO Bin. Wheat variety recognition method based on same position segmentation of transmitted light and reflected light images[J]. Acta Agriculturae Zhejiangensis, 2022, 34(3): 590-598.

图/表 9

图1 纯度检测装置 1,暗室;2,工业相机;3,显微镜;4,主动环形光源;5,背部光源板;6,种子样本;7,载物台;8,小麦种子;9,计算机。

Fig.1 Purity detection device 1, Darkroom; 2, Industrial camera; 3, Microscope; 4, Active ring light source; 5, Back light source plate; 6, Seed samples; 7, Stage; 8, Wheat seeds; 9, Computer.

图2 小麦种子反射光图像和透射光图像 a,京麦9;b,京麦11;c,济麦22;d,济麦44。

Fig.2 Active light and backlight images of wheat seeds a, Jingmai 9; b, Jingmai 11; c, Jimai 22; d, Jimai 44.

图3 背景分割 a,透射光B通道;b,二值化;c,透射光图像;d,反射光图像。

Fig.3 Background segmentation a, Transmitted light channel B; b, Binarization; c, Transmitted light image; d, Reflected light image.

图4 透射光R分量图像增强 a,未增强图像;b,增强后图像。

Fig.4 Image enhancement of R component of transmitted light a, Unenhanced image; b, Enhanced image.

图5 胚部区域形态学处理 a,胚部二值化;b,分割连通域;c,胚部轮廓选择;d,区域闭运算。

Fig.5 Morphological treatment of embryonic region a, Embryo binarization; b, Split connected domain; c, Embryo contour selection; d, Region closed operation.

图6 胚部、胚乳部分割图像 a,胚部反射光图像;b,胚部透射光图像;c,胚乳反射光图像;d,胚乳透射光图像。

Fig.6 Images of embryo and endosperm sections a, Reflected light image of embryo; b, Transmitted light image of embryo; c, Reflected light image of endosperm; d, Transmitted light image of endosperm.

图7 特征参数分布图

Fig.7 Distribution of characteristic parameters

表1 按单个特征划分阈值区间识别结果

Table 1 Identification results based on threshold interval according to a single feature

特征分类 Feature classification	提取特征 Extract features	分选区间 Sorting interval	京麦9 Jingmai 9	京麦11 Jingmai 11	济麦22 Jimai 22	济麦44 Jimai 44	准确率 Accuracy/%
整粒Seeds	B_tm	<84.215	3	4	49	47	94.5
		≥84.215	47	46	1	3
	R_td	<35.09	2	4	49	43	93.0
		≥35.09	48	46	1	7
	G_td	<33.395	4	4	49	43	92.0
		≥33.395	46	46	1	7
	V_td	<35.22	2	4	49	43	93.0
		≥35.22	48	46	1	7
胚部Seed embryo	B_tm	<59.365	3	2	46	44	92.5
		≥59.365	47	48	4	6
胚乳部Seed endosperm	B_tm	<89.78	4	3	49	46	94.0
		≥89.78	46	47	1	4
	R_td	<30.405	5	5	48	43	90.5
		≥30.405	45	45	2	7
	V_td	<30.52	5	5	48	43	90.5
		≥30.52	45	45	2	7

表2 PLS-DA种粒分类正确率

Table 2 Classification accuracy of PLS-DA seeds analysis %

二分类小麦品种 Binary classification of wheat varieties	整粒透射光 Seeds transmitted light	整粒反射光 Seeds reflected light	整粒全光 Seeds full light	分割透射光 Transmitted light after seeds segmentation	分割反射光 Reflected light after seeds segmentation	分割全光 Full light after seeds segmentation	全部特征 All features
京麦9、济麦22	98.88	95.50	99.43	99.43	96.38	99.60	99.98
Jingmai 9、Jimai 22
京麦9、济麦44	98.90	95.55	99.10	99.45	97.63	99.53	99.90
Jingmai 9、Jimai 44
京麦11、济麦22	98.68	95.25	99.00	99.10	95.18	99.18	99.58
Jingmai 11、Jimai 22
京麦11、济麦44	98.43	95.08	99.10	99.33	94.53	99.60	99.68
Jingmai 11、Jimai 44
京麦组、济麦组	99.38	95.78	99.69	99.61	97.58	99.78	99.84
Jingmai Group、Jimai Group
济麦22、济麦44	69.48	73.28	83.00	81.60	76.88	83.75	84.60
Jimai 22、Jimai 44
京麦9、京麦11	59.98	74.15	75.28	72.45	76.20	81.83	83.73
Jingmai 9、Jingmai 11

参考文献 30

[1]	何中虎, 庄巧生, 程顺和, 等. 中国小麦产业发展与科技进步[J]. 农学学报, 2018, 8(1): 99-106.
	HE Z H, ZHUANG Q S, CHENG S H, et al. Wheat production and technology improvement in China[J]. Journal of Agriculture, 2018, 8(1): 99-106. (in Chinese with English abstract)
[2]	颜启传. 种子检验原理和技术[M]. 杭州: 浙江大学出版社, 2001.
[3]	王婷, 张进生, 戴钢, 等. 电泳检测小麦、玉米种子纯度存在的问题探析[J]. 河南农业科学, 2005, 34(12): 41-42.
	WANG T, ZHANG J S, DAI G, et al. Analysis on the problems existing in the detection of wheat and corn seed purity by electrophoresis[J]. Journal of Henan Agricultural Sciences, 2005, 34(12): 41-42. (in Chinese)
[4]	刘燕德, 应义斌, 成芳. 机器视觉技术在种子纯度检验中的应用[J]. 农业机械学报, 2003, 34(5): 161-163.
	LIU Y D, YING Y B, CHENG F. Research of machine vision in purity inspection of seed[J]. Transactions of the Chinese Society of Agricultural Machinery, 2003, 34(5): 161-163. (in Chinese with English abstract)
[5]	张谊寒. 杂种小麦种子纯度鉴定技术的研究[D]. 杨凌: 西北农林科技大学, 2005.
	ZHANG Y H. Study on purity identification of seeds in hybrid wheat[D]. Yangling: Northwest A & F University, 2005. (in Chinese with English abstract)
[6]	王芳. 种子纯度鉴定方法及其评述[J]. 中国种业, 2008(10): 62-63.
	WANG F. Seed purity identification methods and their comments[J]. China Seed Industry, 2008(10): 62-63. (in Chinese)
[7]	章志兴, 徐开盛, 陈俊涛, 等. 叶色标记技术在杂交水稻种子生产中的应用[J]. 种子科技, 2001, 19(1): 33-34.
	ZHANG Z X, XU K S, CHEN J T, et al. Application of leaf color marker technology in hybrid rice seed production[J]. Seed Science & Technology, 2001, 19(1): 33-34. (in Chinese)
[8]	YANG X L, HONG H M, YOU Z H, et al. Spectral and image integrated analysis of hyperspectral data for waxy corn seed variety classification[J]. Sensors (Basel, Switzerland), 2015, 15(7): 15578-15594. DOI URL
[9]	BAO Y D, MI C X, WU N, et al. Rapid classification of wheat grain varieties using hyperspectral imaging and chemometrics[J]. Applied Sciences, 2019, 9(19): 4119. DOI URL
[10]	LIU S X, ZHANG H J, WANG Z, et al. Determination of maize seed purity based on multi-step clustering[J]. Applied Engineering in Agriculture, 2018, 34(4): 659-665. DOI URL
[11]	HUANG M, TANG J Y, YANG B, et al. Classification of maize seeds of different years based on hyperspectral imaging and model updating[J]. Computers and Electronics in Agriculture, 2016, 122: 139-145. DOI URL
[12]	ZHANG J, DAI L M, CHENG F. Corn seed variety classification based on hyperspectral reflectance imaging and deep convolutional neural network[J]. Journal of Food Measurement and Characterization, 2021, 15(1): 484-494. DOI URL
[13]	宋韬, 曾德超. 基于人工神经网络的玉米籽粒形态识别方法的研究[J]. 农业工程学报, 1996, 12(1): 177-181.
	SONG T, ZENG D C. An investigation on morphological discrimination of corn kernels based on neural network[J]. Transactions of the Chinese Society of Agricultural Engineering, 1996, 12(1): 177-181. (in Chinese with English abstract)
[14]	MAJUMDAR S, JAYAS D S. Classification of cereal grains using machine vision i: morphology models[J]. Transactions of the ASAE, 2000, 43(6): 1669-1675. DOI URL
[15]	MAJUMDAR S, JAYAS D S. Classification of cereal grains using machine vision ii: color models[J]. Transactions of the ASAE, 2000, 43(6): 1677-1680. DOI URL
[16]	MAJUMDAR S, JAYAS D S. Classification of cereal grains using machine vision iii: texture models[J]. Transactions of the ASAE, 2000, 43(6): 1681-1687. DOI URL
[17]	刘双喜, 张宏建, 王金星, 等. 基于可见光波段的色彩概率聚类模型的玉米杂交种子识别[J]. 光谱学与光谱分析, 2018, 38(8): 2516-2523.
	LIU S X, ZHANG H J, WANG J X, et al. Hybrid seed recognition of maize based on probability clustering model using visible light color features[J]. Spectroscopy and Spectral Analysis, 2018, 38(8): 2516-2523. (in Chinese with English abstract)
[18]	程洪, 李江涛, 史智兴, 等. 玉米籽粒的特征选择算法:基于支持向量机与遗传算法[J]. 农机化研究, 2009, 31(2):30-33.
	CHENG H, LI J T, SHI Z X, et al. Feature selection of corn seed based on genetic algorithm and support vector machine[J]. Journal of Agricultural Mechanization Research, 2009, 31(2):30-33. (in Chinese with English abstract)
[19]	宁纪锋, 何东健, 杨蜀秦. 玉米籽粒的尖端和胚部的计算机视觉识别[J]. 农业工程学报, 2004, 20(3): 117-119.
	NING J F, HE D J, YANG S Q. Identification of tip cap and germ surface of corn kernel using computer vision[J]. Transactions of the Chinese Society of Agricultural Engineering, 2004, 20(3): 117-119. (in Chinese with English abstract)
[20]	韩仲志, 赵友刚, 杨锦忠. 基于籽粒RGB图像独立分量的玉米胚部特征检测[J]. 农业工程学报, 2010, 26(3): 222-226, 389.
	HAN Z Z, ZHAO Y G, YANG J Z. Detection of embryo based on independent components for kernel RGB images in maize[J]. Transactions of the Chinese Society of Agricultural Engineering, 2010, 26(3): 222-226, 389. (in Chinese with English abstract)
[21]	董高, 郭建, 王成, 等. 基于近红外高光谱成像及信息融合的小麦品种分类研究[J]. 光谱学与光谱分析, 2015, 35(12): 3369-3374.
	DONG G, GUO J, WANG C, et al. The classification of wheat varieties based on near infrared hyperspectral imaging and information fusion[J]. Spectroscopy and Spectral Analysis, 2015, 35(12): 3369-3374. (in Chinese with English abstract)
[22]	王炳强, 程洪, 刘冲. 基于双边滤波的RSG玉米籽粒胚部提取研究[J]. 食品与机械, 2017, 33(2): 36-38.
	WANG B Q, CHENG H, LIU C. Study on extraction of corn kernel embryo by RSG base on bilateral filtering[J]. Food & Machinery, 2017, 33(2): 36-38. (in Chinese with English abstract)
[23]	张俊雄, 武占元, 宋鹏, 等. 玉米单倍体种子胚部特征提取及动态识别方法[J]. 农业工程学报, 2013, 29(4): 199-203.
	ZHANG J X, WU Z Y, SONG P, et al. Embryo feature extraction and dynamic recognition method for maize haploid seeds[J]. Transactions of the Chinese Society of Agricultural Engineering, 2013, 29(4): 199-203. (in Chinese with English abstract)
[24]	LIU J, PAULSEN M R. Corn whiteness measurement and classification using machine vision[J]. Transactions of the ASAE, 2000, 43(3): 757-763. DOI URL
[25]	史智兴, 程洪, 李江涛, 等. 图像处理识别玉米品种的特征参数研究[J]. 农业工程学报, 2008, 24(6): 193-195.
	SHI Z X, CHENG H, LI J T, et al. Characteristic parameters to identify varieties of corn seeds by image processing[J]. Transactions of the Chinese Society of Agricultural Engineering, 2008, 24(6): 193-195. (in Chinese with English abstract)
[26]	程洪, 史智兴, 尹辉娟, 等. 基于机器视觉的多个玉米籽粒胚部特征检测[J]. 农业工程学报, 2013, 29(19): 145-151.
	CHENG H, SHI Z X, YIN H J, et al. Detection of multi-corn kernel embryos characteristic using machine vision[J]. Transactions of the Chinese Society of Agricultural Engineering, 2013, 29(19): 145-151. (in Chinese with English abstract)
[27]	程洪, 史智兴, 冯娟, 等. 基于玉米胚部特征参数优化的玉米品种识别研究[J]. 中国粮油学报, 2014, 29(6): 22-26.
	CHENG H, SHI Z X, FENG J, et al. Corn embryo parameters optimization and varieties identification research[J]. Journal of the Chinese Cereals and Oils Association, 2014, 29(6): 22-26. (in Chinese with English abstract)
[28]	杨嘉能, 李华, 田宸玮, 等. 基于自适应校正的动态直方图均衡算法[J]. 计算机工程与设计, 2021, 42(5): 1264-1270.
	YANG J N, LI H, TIAN C W, et al. Adaptive correction based dynamic histogram equalization[J]. Computer Engineering and Design, 2021, 42(5): 1264-1270. (in Chinese with English abstract)
[29]	朱海洋, 徐根玖, 李元晨, 等. 基于图像分块的局部区域动态阈值选取方法[J]. 计算机与现代化, 2016(11): 53-57.
	ZHU H Y, XU G J, LI Y C, et al. Dynamic threshold selection of local region based on image partition[J]. Computer and Modernization, 2016(11): 53-57. (in Chinese with English abstract)
[30]	阿基业. 代谢组学数据处理方法: 主成分分析[J]. 中国临床药理学与治疗学, 2010, 15(5): 481-489.
	AA J Y. Analysis of metabolomic data: principal component analysis[J]. Chinese Journal of Clinical Pharmacology and Therapeutics, 2010, 15(5): 481-489. (in Chinese with English abstract)

基于透射光和反射光图像同位分割的小麦品种识别方法研究

Wheat variety recognition method based on same position segmentation of transmitted light and reflected light images

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 30

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张均, 张博, 胡碧博, 刘京亮, 张晓宇, 李春阳, 熊盛婷, 郭彬彬, 王秀存, 马超. 小麦SWEET和SUT家族基因鉴定与表达分析[J]. 浙江农业学报, 2025, 37(9): 1825-1839.
[2]	吕胤春, 段恩泽, 朱一星, 郑霞, 柏宗春. 基于YOLOv8-Swin Transformer模型的翻覆肉鸭实时检测[J]. 浙江农业学报, 2025, 37(7): 1556-1566.
[3]	刘胜男, 朱建义, 李明, 赵浩宇, 熊涛, 汤永禄, 周小刚, 李朝苏. 稻茬免耕带旋播种小麦的田间杂草防除效果与小麦产量[J]. 浙江农业学报, 2025, 37(10): 2129-2137.
[4]	李萌民, 刘朔, 欧阳宇, 张鹏. 基于YOLOv8n的高效轻量化柑橘叶片病害检测模型[J]. 浙江农业学报, 2025, 37(10): 2198-2208.
[5]	杨晓雨, 马指挥, 魏青, 牛志鹏, 陈安琪, 胡正冲, 王林生. 一个小麦芒长基因的初步定位及候选基因预测[J]. 浙江农业学报, 2025, 37(1): 14-23.
[6]	沈峥嵘, 戴远兴, 郭留明, 汪芷瑶, 张恒木. 中国小麦花叶病毒(CWMV)外壳蛋白(CP)特异性抗体的制备与应用[J]. 浙江农业学报, 2024, 36(9): 2042-2050.
[7]	郭秀明, 王大伟, 刘升平, 诸叶平, 刘晓辉, 林克剑, 王佳宇, 李非. 基于深度学习的近地面草原鼠洞识别计数关键问题研究与应用[J]. 浙江农业学报, 2024, 36(9): 2146-2154.
[8]	齐学礼, 李莹, 段俊枝. 耐盐基因在小麦耐盐基因工程中的应用[J]. 浙江农业学报, 2024, 36(6): 1447-1457.
[9]	李晶晶, 李闯, 路亚南, 郑文明. 小麦类硫素基因家族鉴定及表达分析[J]. 浙江农业学报, 2024, 36(4): 729-737.
[10]	张永彬, 李想, 满卫东, 刘明月, 樊继好, 胡皓然, 宋利杰, 刘玮佳. 融合Sentinel-1/2数据和机器学习算法的冬小麦产量估算方法研究[J]. 浙江农业学报, 2024, 36(12): 2812-2822.
[11]	刘永安, 黄业昌, 岳高红, 高锡腾, 邓立章, 潘彬荣. 优质小麦品种温麦10号籽粒蛋白质组学分析[J]. 浙江农业学报, 2024, 36(11): 2437-2446.
[12]	娄渊根, 李闯, 李晶晶, 邢国珍, 路亚南, 郑文明. 小麦HP基因家族鉴定和分析[J]. 浙江农业学报, 2023, 35(9): 2023-2032.
[13]	宁文楷, 李静, 沈晓东, 吴鑫, 李臻锋. 南瓜干燥过程中β-胡萝卜素的多源融合预测[J]. 浙江农业学报, 2023, 35(8): 1876-1887.
[14]	余桂红, 宋桂成, 张鹏, 王化敦, 范祥云. 十八个小麦品种(系)拔节期耐渍性的综合评价[J]. 浙江农业学报, 2023, 35(6): 1235-1242.
[15]	杨凯, 陈凯, 李红梅, 赵忠娟, 扈进冬, 李纪顺, 杨合同. 哈茨木霉LTR-2与产脲节杆菌DnL1-1协同对小麦茎基腐病的防治效果与机理[J]. 浙江农业学报, 2023, 35(6): 1385-1395.