基于高光谱成像的核桃仁品质检测与分类方法

引用本文: 马文强, 张漫, 李源, 李民赞, 杨莉玲, 朱占江, 崔宽波. 基于高光谱成像的核桃仁品质检测与分类方法[J]. 分析化学, 2020, 48(12): 1737-1746. doi: 10.19756/j.issn.0253-3820.191289 shu

Citation: MA Wen-Qiang, ZHANG Man, LI Yuan, LI Min-Zan, YANG Li-Ling, ZHU Zhan-Jiang, CUI Kuan-Bo. Detection and Grading Method of Walnut Kernel Quality Based on Hyperspectral Image[J]. Chinese Journal of Analytical Chemistry, 2020, 48(12): 1737-1746. doi: 10.19756/j.issn.0253-3820.191289 shu

基于高光谱成像的核桃仁品质检测与分类方法

马文强 ^1,2, ,
张漫 ^{1,
,} ,
李源 ^3, ,
李民赞 ^1, ,
杨莉玲 ^2, ,
朱占江 ^2, ,
崔宽波 ^2,

1.
中国农业大学现代精细农业系统集成研究教育部重点实验室, 北京 100083;
2.
新疆农业科学院农业机械化研究所, 乌鲁木齐 830091;
3.
新疆农业科学院土壤肥料与农业节水研究所, 乌鲁木齐 830091

通讯作者: 张漫, cauzm@cau.edu.cn

基金项目:
本文系新疆农业科学院青年科技骨干创新能力培养项目（No.xjnkq-2019007）、国家自然科学基金项目（No.31701319）和新疆农业科学院科技创新重点培育专项（No.xjkcpy-004）资助

摘要: 采用光谱与图像相结合，实现了核桃仁蛋白质和脂肪含量预测及基于完整度和色泽的核桃仁外观品质分级。选用新疆"温185"核桃仁，采集了862.9~1704.02 nm和382.19~1026.66 nm范围高光谱图像。采用多元散射校正（MSE）和标准正态化（SNV）方法进行预处理后，通过竞争性自适应重加权采样算法与相关系数法，对核桃仁样品的蛋白质含量、脂肪含量、总色差3个参数进行了特征波段筛选。通过偏最小二乘回归（PLSR）算法建立了全光谱波段与特征光谱波段的蛋白质和脂肪含量预测模型，与全光谱波段相比，蛋白质含量特征波段预测模型的验证集决定系数（R²）由0.66增长到0.91，均方根误差（RMSEP）由1.37%下降到0.78%；脂肪含量特征波段预测模型的验证集R²由0.83增长到0.93，RMSEP由0.98%下降到0.47%。在外观品质方面，采用全光谱波段、RGB光谱波段、总色差特征光谱波段为输入，采用决策树、K近邻和支持向量机算法建立了核桃仁外观品质分类模型。通过对比发现，采用总色差特征波段建模，可大幅减低冗余信息的干扰，同时分类准确率也高于RGB波段；在光谱信息的基础上加入图像统计特征参数信息，能够进一步提升分类的准确率，当采用决策数算法建立的色泽分类模型时，模型具有最高的分类准确率（98.6%）；分类算法方面，当输入变量数目较少时，决策树算法在分类准确率和速度方面都具有明显的优势。利用高光谱技术可以实现核桃仁内部品质检测与外观分级，为核桃仁品质无损检测的提供了新的理论依据。

关键词:

English

Detection and Grading Method of Walnut Kernel Quality Based on Hyperspectral Image

1.
Key Laboratory of Modern Precision Agriculture System Integration Research, Ministry of Education, College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China;
2.
Agricultural Mechanization Institute, Xinjiang Academy of Agricultural Sciences, Urumqi 830091, China;
3.
Soil Fertilizer and Agricultural Water Saving Research Institute, Xinjiang Academy of Agricultural Sciences, Urumqi 830091, China

Corresponding author: ZHANG Man, cauzm@cau.edu.cn

Received Date: 16 August 2019
Revised Date: 14 October 2020
Fund Project:
This work was supported by the Project of Renovation Capacity Building for the Young Sci-Tech Talents Sponsored by Xinjiang Academy of Agricultural Sciences (No. xjnkq-2019007), the National Natural Science Foundation of China (No.31701319) and the Science and Technology Innovation Key Cultivation Special Project of Xinjiang Academy of Agricultural Sciences (No.xjkcpy-004).

Abstract: Hyperspectral imaging technology enables rapid non-destructive inspection and grading of various agricultural products. In this work, the research on the quality detection method of walnut kernel based on hyperspectral image was carried out. The combination of spectrum and image information was used to realize the prediction of protein and fat content and the classification of the integrity and color of walnut kernel. The "Wen 185" walnuts, which were produced from Xinjiang, were shelled and prepared by different grades of integrity and color. Then the hyperspectral image of each sample was measured in the range of 862.9-1704.02 nm and 382.19-1026.66 nm by Gaia hyperspectral imager. After that, the color difference, fat content and protein content of samples were measured. Multivariate scatter correction and standard normalized variate were used to pre-processing the original spectral information. And the feature bands were screened by the method, which combined competitive adaptive re-weighting sampling (CARS) and correlation coefficient method (CCM ) algorithm, for the three parameters of protein content, fat content and total color difference of walnut kernel samples. Six feature bands related to protein content and 7 feature bands related to fat content were screened out. The internal quality parameter prediction model of the full spectrum band and the characteristic spectrum band were established by partial least squares regression (PLSR) algorithm. Compared with the full-spectrum band, the verification set coefficient (R²) of the feature band protein content prediction model increased from 0.66 to 0.91, and the mean square error (RMSEP) decreased from 1.37% to 0.78%. The verification set coefficient (R²) of the feature band fat content prediction model increased from 0.83 to 0.93, and the RMSEP decreased from 0.98% to 0.47%. It showed that the selected characteristic bands effectively reduced the complexity of the full spectrum information and improved the quality of modeling. In terms of appearance quality, the feature bands associating with the color difference were selected to be 402.5 and 689.2 nm. The full-spectral spectrum, RGB spectrum, characteristic spectrum and the combination of spectral and image information were used to establish the walnut appearance quality classification model by decision tree, K-nearest neighbor and support vector machine algorithm. It showed that the feature bands modeling greatly reduced the interference of redundant information, improved the modeling efficiency, and the classification accuracy were also significantly higher than the RGB bands by comparison. The adding image statistical feature parameter to the feature bands and RGB bands could further improve the accuracy of classification model which had the highest classification accuracy rate reached to 98.6% by decision tree algorithm. In terms of classification algorithm, the decision tree algorithm had obvious advantages in classification accuracy and calculation speed when the number of input variables was less. The used of hyperspectral technology could realize the internal quality detection and appearance classification of walnut kernels, which provided a new theoretical basis for the application of non-destructive testing of walnut kernel quality.

Key words:

1. [1]
  CAO Ya-Long, JIE Bo, WANG Lei, MENG Hai-Jun, WU Guo-Liang. J. Fruit Sci.,2018,35(9):1128-1140 曹亚龙, 揭波, 王磊, 孟海军, 吴国良.果树学报,2018,35(9):1128-1140
2. [2]
  LYT 1922-2010, Walnut Kernel. Forestry Industry Standards of the People's Republic of China核桃仁.中华人民共和国林业行业标准. LYT 1922-2010
3. [3]
  GB/T 20398-2006, Walnut Quality Grade. National Standards of the People's Republic of China核桃坚果质量等级.中华人民共和国国家标准. GB/T 20398-2006
4. [4]
  SHEN Ai-Min, HUO Xiao-Jing, WANG Wen-Di, WANG Hong-Xia. Jiangsu Agric. Sci., 2016,44(9):383-385 申爱敏, 霍晓静, 王文娣, 王红霞.江苏农业科学,2016,44(9):383-385
5. [5]
  CAI Jian, ZHOU Jun, SHI Jian-Xin, WANG Yong, GUO Zheng, LIU Hang. Jiangsu Agric. Sci., 2018,46(5):158-161 蔡建, 周军, 史建新, 王勇, 郭政, 刘航.江苏农业科学,2018,46(5):158-161
6. [6]
  ZHOU Jun, CAI Jian, GUO Jun-Xian, ZHANG Jing, SHI Jian-Xin, ALI Hasmu, JIANG Yan-Wu. Jiangsu Agric. Sci., 2018,46(11):175-179 周军, 蔡建, 郭俊先, 张静, 史建新, 艾力·哈斯木, 姜彦武.江苏农业科学, 2018,46(11):175-179
7. [7]
  7 HAO Zhong-Cheng, PENG Yun-Fa, ZHANG Hong, ZHANG Rui, ZHANG Rui-Li. J. Anhui Agric. Sci., 2014,42(21):7191-7193,7233 郝中诚, 彭云发, 张宏, 张锐, 张锐利.安徽农业科学,2014,42(21):7191-7193,7233
8. [8]
  MA Wen-Qiang, ZHANG Man, LI Zhong-Xin. Transact. Chin. Soc. Agric. Machin., 2015,46(S1):128-133 马文强, 张漫, 李忠新.农业机械学报,2015,46(S1):128-133
9. [9]
  MA Wen-Qiang, ZHANG Man, LI Zhong-Xin, YANG Li-Ling. Transacte. Chin. Soc. Agric. Machin., 2017, 48(S1):407-411 马文强, 张漫, 李忠新, 杨莉玲.农业机械学报,2017,48(S1):407-411
10. [10]
  Kandpal L M, Park E, Tewari J, Cho B. J. Biosyst. Engineer., 2015,40(2):124-136
11. [11]
  Kandpal L M, Cho B. J. Biosyst. Engineer., 2014,39(3):215-226
12. [12]
  Mishra P, Nordon A, Asaari M S M, Lian G P, Redfern S. J. Food Engineer.,2019,249(5):40-47
13. [13]
  Su W H, Bakalis S, Sun D W. Biosyst. Engineer., 2019,180(4):70-86
14. [14]
  LI Xiao-Yu, XU Sen-Miao, FENG Yao-Ze, HUANG Tao, DING Chong-Yi. Transact. Chin. Soc. Agric. Machin.,2016, 47(1):221-226 李小昱, 徐森淼, 冯耀泽, 黄涛, 丁崇毅.农业机械学报,2016,47(1):221-226
15. [15]
  Cheng J H, Jin H L, Xu Z Y, Zheng F P. Anal. Methods, 2017,43(9):6148-6154
16. [16]
  Wei Y Z, Wu F Y, Xu J,Sha J J, Zhao Z F, He Y, Li X L. J. Food Engineer.,2019,248:98-96
17. [17]
  Zhao H T, Feng Y Z, Chen W, Jia G F. Meat Sci., 2019,151:75-81
18. [18]
  Fan S X, Li C Y, Huang W Q, Chen L P. Sensors,2018,18(12):1-19
19. [19]
  ZHANG Hang, YAO Chuan-An, JIANG Meng-Meng, JI Yu-Hang, LI Hua-Jie. J. Triticeae Crops, 2019, 39(1):96-104 张航, 姚传安, 蒋梦梦, 姬豫航, 李华杰.麦类作物学报,2019,39(1):96-104
20. [20]
  CHEN Cai-Hong, ZHANG Shu-Juan, SUN Hai-Xia, LI Cheng-Ji, XING Shu-Hai. J. Shanxi Univ. (Nat. Sci. Edit.), 2018, 38(11):27-32 陈彩虹, 张淑娟, 孙海霞, 李成吉, 邢书海.山西农业大学学报(自然科学版),2018,38(11):27-32

扫一扫看文章

计量

PDF下载量: 21
文章访问数: 1182
HTML全文浏览量: 176

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

基于高光谱成像的核桃仁品质检测与分类方法

通讯作者: 张漫, cauzm@cau.edu.cn

English

Detection and Grading Method of Walnut Kernel Quality Based on Hyperspectral Image

Corresponding author: ZHANG Man, cauzm@cau.edu.cn

CCS Chemistry：嵌段共聚物自组装成 “三明治”二维有机材料，实现纳米级压力传感功能

CCS Chemistry：颜徐州、鲍哲南：你的皮肤可能是一片SPMs

CCS Chemistry：工作高效不疲劳，只须让催化剂动起来

CCS Chemistry：静电组装导向聚合- 聚电解质纳米凝胶量化制备新策略

CCS Chemistry：金黄色葡萄球菌醛基脱氢酶是如何识别特异性底物的？

计量

通讯作者: 陈斌, bchen63@163.com

中国化学会秘书处

基于高光谱成像的核桃仁品质检测与分类方法

通讯作者: 张漫, cauzm@cau.edu.cn

English

Detection and Grading Method of Walnut Kernel Quality Based on Hyperspectral Image

Corresponding author: ZHANG Man, cauzm@cau.edu.cn

CCS Chemistry：嵌段共聚物自组装成 “三明治”二维有机材料，实现纳米级压力传感功能

CCS Chemistry：颜徐州、鲍哲南： 你的皮肤可能是一片SPMs

CCS Chemistry：工作高效不疲劳，只须让催化剂动起来

CCS Chemistry：静电组装导向聚合- 聚电解质纳米凝胶量化制备新策略

CCS Chemistry：金黄色葡萄球菌醛基脱氢酶是如何识别特异性底物的？

计量

文章相关

通讯作者: 陈斌, bchen63@163.com

中国化学会秘书处

CCS Chemistry：颜徐州、鲍哲南：你的皮肤可能是一片SPMs