浙江农业学报 ›› 2020, Vol. 32 ›› Issue (3): 527-533.DOI: 10.3969/j.issn.1004-1524.2020.03.19

• 生物系统工程 • 上一篇    下一篇

基于高光谱成像技术的泾源黄牛肉色度PLSR预测模型构建

禹文杰, 王彩霞, 乔芦, 王松磊, 贺晓光*   

  1. 宁夏大学 农学院,宁夏 银川 750021
  • 收稿日期:2019-09-03 出版日期:2020-03-25 发布日期:2020-04-03
  • 通讯作者: *贺晓光,E-mail:875949696@qq.com
  • 作者简介:禹文杰(1992—),男,宁夏银川人,硕士研究生,主要从事农产品无损检测和自动控制研究。E-mail:yu_wj5954@163.com
  • 基金资助:
    宁夏高校科研基金(NGY2016018); 中央财政支持地方高校改革发展资金—食品学科建设项目(2017); 宁夏回族自治区重点项目

Construction of PLSR prediction model for detecting color of Jingyuan yellow beef by hyperspectral technique

YU Wenjie, WANG Caixia, QIAO Lu, WANG Songlei, HE Xiaoguang*   

  1. School of Agriculture, Ningxia University, Ningxia 750021, China
  • Received:2019-09-03 Online:2020-03-25 Published:2020-04-03

摘要: 利用高光谱成像技术对泾源黄牛肉色度的PLSR预测模型进行构建。通过可见近红外高光谱成像系统获取牛肉样本的高光谱图像,提取感兴趣区域(ROI)的光谱信息并计算平均光谱,采用蒙特卡洛法剔除异常样本后进行样本集划分,并对划分后的样本数据进行预处理。其中,亮度(L*)经Deresolve法预处理的模型结果最好,其RC2为0.979 0,预测集相关系数RP2为0.976 6;红度(a*)经卷积平滑法预处理的模型结果最好,其RC2RP2分别为0.807 0、0.915 5;黄度(b*)经卷积平滑法预处理的模型结果最好,其RC2RP2分别为0.931 1、0.950 6。分别利用竞争性自适应重加权法(CARS)、连续投影算法(SPA)和无信息变量消除算法(UVE)进行特征波长提取,建立基于特征波段的偏最小二乘回归(PLSR)模型,进而优选出最佳预测模型,结合视觉的空间深度、立体程度,对样本的形态和色觉感知进行提取与辨别。结果表明,利用高光谱成像技术所构建的色度PLSR模型均是可行的,研究结果为牛肉品质在线快速检测提供了理论依据。

关键词: 泾源黄牛, 高光谱成像, 色度值, 偏最小二乘回归

Abstract: The PLSR prediction model for beef color of Jingyuan yellow cattle was constructed by hyperspectral image. The hyperspectral images of samples were obtained by visible near-infrared hyperspectral imaging system, the spectral information of the interest regions were extracted, and the average spectrums were calculated. The Monte Carlo method was used to eliminate the abnormal samples, then the sample set was divided and the sample data was preprocessed. Results showed that model with the lightness (L*) pretreated by the Deresolve method performed best while RC2 was 0.979 0 and RP2 was 0.976 6. Model with the redness (a*) pretreated by convolution smoothing (Smoothing-SG) method performed best while RC2 was 0.807 0 and RP2 was 0.915 5. Model with the yellowness (b*) pretreated by convolution smoothing (Smoothing-SG) method performed best while RC2 was 0.931 1 and RP2 was 0.950 6. Characteristic wavelengths were extracted by using competitive adaptive re-weighting method (CARS), continuous projection algorithm (SPA) and non-information variable elimination algorithm (UVE) respectively. The partial least squares regression (PLSR) model based on characteristic band was achieved and further optimized. The best prediction model was combined with the spatial depth and stereoscopic degree of the vision to extract and distinguish form and color perception of the sample. Therefore, it was feasible to construct the PLSR model by using hyperspectral imaging based on values of L*, a* and b*, and the theoretical basis for online rapid detection of beef quality was provided.

Key words: Jingyuan yellow cattle, hyperspectral imaging technology, chromaticity values, partial least squares regression

中图分类号: