食品科学 ›› 2018, Vol. 39 ›› Issue (16): 321-327.doi: 10.7506/spkx1002-6630-201816047

• 安全检测 • 上一篇    下一篇

基于近红外光谱技术对多年际建三江、五常大米产地溯源

钱丽丽1,2,宋雪健1,张东杰1,2,*,张丽媛1,2,阮长青1,2,鹿保鑫1,2   

  1. (1.黑龙江八一农垦大学食品学院,黑龙江?大庆 163319;2.黑龙江八一农垦大学 黑龙江省农产品加工与质量安全重点实验室,黑龙江?大庆 163319)
  • 出版日期:2018-08-25 发布日期:2018-08-17
  • 基金资助:
    黑龙江省水稻现代农业科技协同创新体系-加工技术攻关实验室主任专家项目(2017); “十三五”国家重点研发计划重点专项(2017YFD0401203); 黑龙江省垦总局“十三五”重点科技攻关项目(HNK135-06-06);黑龙江省应用技术研究与开发计划项目(GA14B104)

Tracing the Geographical Origin of Sanjiang and Wuchang Rice Grown in Different Years by Near Infrared Spectroscopy

QIAN Lili1,2, SONG Xuejian1, ZHANG Dongjie1,2,*, ZHANG Liyuan1,2, RUAN Changqing1,2, LU Baoxin1,2   

  1. (1. College of Food Science, Heilongjiang Bayi Agricultural University, Daqing 163319, China; 2. Key Laboratory of Agro-Products Processing and Quality Safety of Heilongjiang Province, Heilongjiang Bayi Agricultural University, Daqing 163319, China)
  • Online:2018-08-25 Published:2018-08-17

摘要: 研究多年际地理标志大米产地溯源判别技术,保护地理标志大米的品牌效益。利用近红外漫反射光谱技术对试验田样品进行与产地有关的特征波段筛选,并在其范围内对2013—2015年来自建三江地区及五常地区的291?份大米样品进行产地溯源研究。结果表明,试验田样品在波段为5?136~5?501?cm-1范围内产地因素差异显著。在此范围内利用因子化法建立的定性分析模型及聚类分析模型对建三江大米及五常大米的正确判别率均高于97.00%。利用偏最小二乘法建立的定量分析模对两地区大米的正确判别率分别为95.83%、94.00%。因此,应用该技术对多年份大米的产地溯源进行判别具有一定的可行性。

关键词: 大米, 产地溯源, 近红外漫反射光谱技术, 因子化法, 偏最小二乘法

Abstract: This study aimed to develop a technique to trace the geographical origin of protected geographical indication (PGI) rice grown in different years for the purpose of protecting the brand benefits of PGI rice. Near-infrared diffuse reflectance spectroscopy was used to detect rice samples from experimental fields and the characteristic bands related to the growing area were selected. The geographical origin of 291 rice samples from Sanjiang and Wuchang areas, collected during 2013 to 2015 was analyzed using the characteristic bands. There were significant differences among the experimental filed grow rice samples in terms of geographical origin in the range of 5 136–5 501 cm-1. In this range, the qualitative analysis model and the cluster analysis model established by the factorization method showed a correct recognition rate of higher than 97.00% for Sanjiang and Wuchang rice; the correct recognition rates of the quantitative analysis model established by partial least squares (PLS) were 95.83% and 94.00%, respectively. Therefore, this method was proved feasible for tracing the geographical origin of rice harvested in different years.

Key words: rice, geographical origin traceability, near infrared diffuse reflectance spectroscopy, factorization method, partial least squares

中图分类号: