基于高光谱图像的贝类重金属污染快速无损检测方法研究

熊建芳; 乔付; 刘忠艳; 刘瑶; 郝博麟; 姜薇; 许乐乐; 卢利琼

doi:10.13205/j.hjgc.202210019

基于高光谱图像的贝类重金属污染快速无损检测方法研究

doi: 10.13205/j.hjgc.202210019

1. 岭南师范学院计算机与智能教育学院, 广东湛江 524048;
2. 岭南师范学院电子与电气工程学院, 广东湛江 524048;
3. 岭南师范学院生命科学与技术学院, 广东湛江 524048

基金项目:

广东省教育厅普通高校特色创新项目"自然场景图像中多尺度多形态文本检测与评价方法研究"(2021KTSCX065)

国家自然科学基金青年科学基金项目"贝类重金属污染的高光谱无损检测机理、模型及综合性能评估研究"(62005109)

广东省普通高校特色创新项目"基于粗糙集的高光谱特征波段选择方法研究及其在贝类重金属无损检测中的应用"(2019KTSCX)

广东省自然科学基金面上项目"贝类毒素的高光谱快速检测机理及方法研究"(2020A515011368)

详细信息

作者简介:
熊建芳(1980-),女,讲师,主要研究方向为高光谱图像处理及数据分析。xjf181@163.com

计量
- 文章访问数: 46
- HTML全文浏览量: 9
- PDF下载量: 2
- 被引次数: 0
出版历程
- 收稿日期: 2022-01-23

RAPID AND NON-DESTRUCTIVE DETECTION FOR SHELLFISH CONTAMINATED BY HEAVY METAL BASED ON HYPERSPECTRAL IMAGES

1. School of Computer and Intelligent Education, Lingnan Normal University, Zhanjiang 524048, China;
2. School of Electronic and Electrical Engineering, Lingnan Normal University, Zhanjiang 524048, China;
3. School of Life Sciences and Technology, Lingnan Normal University, Zhanjiang 524048, China

摘要

摘要: 基于高光谱图像技术和机器学习算法,提出了一种对重金属污染蛤仔进行快速无损检测的新方法。该方法分为3步:采集蛤仔样本高光谱图像并使用3种方法进行预处理;采用线性判别分析(linear discriminant analysis,LDA)对高光谱数据降维;应用支持向量机(support vector machine, SVM)实现重金属污染蛤仔分类检测。对于以单类重金属污染样本和健康样本为样本集的二分类检测,LDA-SVM模型检测重金属污染样本的准确率可达到99.33%以上。对于以Cd、Cu、Pb、Zn 4类重金属污染样本和健康样本为样本集的五分类检测,检测准确率可达到93.33%。结果表明:LDA-SVM模型能够实现对蛤仔重金属污染快速无损检测,且该模型性能基本不受预处理方法和模型参数的影响,鲁棒性强。
- 重金属 /
- 菲律宾蛤仔 /
- 高光谱 /
- SVM /
- LDA-SVM
Abstract: A new method for rapid and non-destructive detection of Ruditapes Philippinarum contaminated by heavy metals has been proposed in this paper, which is based on hyperspectral image technology and machine learning algorithm. Firstly, the hyperspectral images of Ruditapes Philippinarum were collected and preprocessed by three preprocessing methods. Then the dimension of hyperspectral image data was reduced by Linear Discriminant Analysis (LDA). Finally, heavy metal-contaminated Ruditapes Philippinarum was detected by the Support Vector Machine (SVM). For binary classification of single heavy metal-contaminated samples and healthy samples, the accuracy of LDA-SVM model for detecting heavy metal-contaminated samples was higher than 99.33%. For multi-classification of Cd, Cu, Pb and Zn-contaminated samples and healthy samples, the accuracy was more than 93.33%. The experimental results showed that LDA-SVM model could realize rapid and non-destructive detection of Ruditapes Philippinarum contaminated by heavy metals. Moreover, the performance of LDA-SVM model was not affected by preprocessing methods and model parameters, and the model had strong robustness.
- heavy metal /
- Ruditapes Philippinarum /
- hyperspectral /
- SVM /
- LDA-SVM

HTML全文

参考文献(29)

[1]	HU M H, CHEN X J, YE P C, et al. Combination of multiple model population analysis and mid-infrared technology for the estimation of copper content in Tegillarca granosa[J].Infrared Physics & Technology, 2016, 79:198-204.
[2]	YUAN L M, CHEN X J, LAI Y J, et al. A novel strategy of clustering informative variables for quantitative analysis of potential toxics element in Tegillarca granosa using laser-induced breakdown spectroscopy[J]. Food Analytical Methods, 2018, 11(5):1405-1416.
[3]	张民,李银花,袁晴春,等.近红外光谱对鲜茶叶茶多酣和氨基酸总量检测的研究[J].上海农业学报,2015,31(6):36-40.
[4]	林楠,刘翰霖,孟祥发,等.基于高光谱的黑土区土壤重金属含量估测[J].农业机械学报,2021,52(3):218-225.
[5]	李勋兰,易时来,何绍兰,等. 高光谱成像技术的柚类品种鉴别研究[J].光谱学与光谱分析,2015,35(9):2639-2643.
[6]	SUN J, ZHOU X, WU X H,et al.Research and analysis of cadmium residue in tomato leaves based on WT-LSSVR and Vis-NIR hyperspectral imaging[J]. Spectrochimica Acta Part A:Molecular and Biomolecular Spectroscopy, 2019,212:215-221.
[7]	张初,刘飞,章海亮,等. 近地高光谱成像技术对黑豆品种无损鉴别[J].光谱学与光谱分析,2014,34(3):746-749.
[8]	姜洪喆,蒋雪松,杨一,等. 肉类掺杂掺假的高光谱成像检测研究进展[J].食品与发酵工业,2021,47(6):300-305.
[9]	孙宗保,梁黎明,李君奎,等.高光谱成像的冰鲜与冻融三文鱼鉴别研究[J].光谱学与光谱分析,2020,40(11):3530-3536.
[10]	LIU Y, QIAN F, WANG S W, et al.Application of hyperspectral imaging technology for rapid identifification of Ruditapes philippinarum contaminated by heavy metals[J]. RSC Advances, 2021, 11:33939-33951.
[11]	GARCÍA-MARTÍN J F, BADARÓ A T, BARBIN D F,et al. Identification of copper in stems and roots of Jatropha curcas L. by hyperspectral imaging[J]. Processes, 2020,8(7):823.
[12]	SHI T Z, CHEN Y Y, LIU Y L, et al. Visible and near-infrared reflectance spectroscopy:an alternative for monitoring soil contamination by heavy metals[J]. Journal of Hazardous Materials, 2014, 265:166-176.
[13]	GAGNÉ F, GAGNON C, TURCOTTE P, et al. Changes in metallothionein levels in freshwater mussels exposed tourban wastewaters:effects from exposure to heavy metals[J]. Biomarker Insights, 2007,2:107-116.
[14]	PAWAR N, GIREESH-BABU P, SABNIS S, et al. Development of a fluorescent transgenic zebrafish biosensor for sensing aquatic heavy metal pollution[J]. Transgenic Research, 2016,25(5):617-627.
[15]	QIN J W, VASEFI F, HELLBERG R S, et al. Detection of fish fillet substitution and mislabeling using multimode hyperspectral imaging techniques[J]. Food Control, 2020, 114:107234.
[16]	CHENG J H, SUN D W. Hyperspectral imaging as an effective tool for quality analysis and control of fish and other seafoods:current research and potential applications[J]. Trends in Food Science & Technology, 2014, 37(2):78-91.
[17]	CHEN X, YUAN L M, CHEN X J, et al. A strategy for rapid identification of healthy Tegillarca granosa from among those contaminated with unspecified heavy metals using infrared spectroscopy[J]. Analytical Methods, 2017,49(30):4447-4454.
[18]	CHEN X J, LIU K, CAI J B, et al. Identification of heavy metal-contaminated Tegillarca granosa using infrared spectroscopy[J]. Analytical Methods, 2015,7(5):2172-2181.
[19]	陈新伟,王俊,沈睿谦,基于GPRS的远程检测无线电子鼻系统[J].农业机械学报, 2015,46(4):238-245.
[20]	LUO F L, DU B, ZHANG L P, et al. Feature learning using spatial-spectral hypergraph discriminant analysis for hyperspectral Image[J].IEEE Transactions on Cybernetics,2019(7):2406-2419.
[21]	JUAN F G, AMANDA T B, DOUGLAS F B, et al. Identification of copper in stems and roots of Jatropha curcas L. by hyperspectral imaging[J].Processes, 2020, 8(7):823.
[22]	XIA C, YANG S, HUANG M, et al. Maize seed classification using hyperspectral image coupled with multi-linear discriminant analysis[J]. Infrared Physics & Technology,2019, 103(2):103077.
[23]	杨明莉,范玉刚,李宝芸.基于LDA和ELM的高光谱图像降维与分类方法研究[J].电子测量与仪器学报,2020,34(5):190-196.
[24]	张静,刘忠宝,宋文爱,等. 基于多类支持向量机的恒星光谱分类方法[J].光谱学与光谱分析,2018,38(7):2307-2310.
[25]	褚小立,许育鹏,陆婉珍.支持向量回归建立成品汽油通用近红外校正模型的研究[J].分析测试学报,2008,27(6):619-622.
[26]	XU L, YAN S M, CAI C B, et al. N ondestructive discrimination of lead (Pb) in preserved eggs (Pidan) by near-infrared spectroscopy and chemometrics[J]. Journal of Spectroscopy, 2014(253143).
[27]	王书越,杨玉柱,何伟文,等.基于高光谱的黑色签字笔墨水种类鉴别方法研究[J].分析测试学报,2021,40(10):1489-1496.
[28]	CHEN X J, LIU K, CAI J B, et al. Identification of heavy metal-contaminated Tegillarca granosa using infrared spectroscopy[J].Anall Methods,2015,7:2172.
[29]	LI X Y, ZHANG L F, YOU J. Locally weighted discriminant analysis for hyperspectral image classification[J].Remote Sensing, 2019, 11(2):109.