• Spectroscopy and Spectral Analysis
  • Vol. 42, Issue 4, 1222 (2022)
Ling-kai MA*, Shi-ping ZHU*;, Yu-jie MIAO, Xiao WEI, Song LI, You-lie JIANG, and Jia-xin ZHUO
Author Affiliations
  • College of Engineering and Technology, Southwest University, Chongqing 400716, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2022)04-1222-07 Cite this Article
    Ling-kai MA, Shi-ping ZHU, Yu-jie MIAO, Xiao WEI, Song LI, You-lie JIANG, Jia-xin ZHUO. The Discrimination of Organic and Conventional Eggs Based on Hyperspectral Technology[J]. Spectroscopy and Spectral Analysis, 2022, 42(4): 1222 Copy Citation Text show less
    Hyperspectral system in laboratory(a): Physical image of system; (b): Architecture drawing of system
    Fig. 1. Hyperspectral system in laboratory
    (a): Physical image of system; (b): Architecture drawing of system
    Hyperspectral image of sample(a): Raw hyperspectral image; (b): Linear stretched image;(c): The selected region of interest
    Fig. 2. Hyperspectral image of sample
    (a): Raw hyperspectral image; (b): Linear stretched image;(c): The selected region of interest
    The spectral curves and average spectral curves of all samples in different ROI(a): The spectral curves of all samples in albumen ROI; (b): The spectral curves of all samples in yolk ROI;(c): The average spectral curves of all samples in albumen ROI; (d): The average spectral curves of all samples in yolk ROI
    Fig. 3. The spectral curves and average spectral curves of all samples in different ROI
    (a): The spectral curves of all samples in albumen ROI; (b): The spectral curves of all samples in yolk ROI;(c): The average spectral curves of all samples in albumen ROI; (d): The average spectral curves of all samples in yolk ROI
    The effect of the number of bands used in modeling on model accuracy(a): The changing trend of number of bands; (b): The changing trend of RMSECV of model
    Fig. 4. The effect of the number of bands used in modeling on model accuracy
    (a): The changing trend of number of bands; (b): The changing trend of RMSECV of model
    The selected bands through SPA and CARS respectively
    Fig. 5. The selected bands through SPA and CARS respectively
    区域鉴别模型训练集预测集
    有机普通准确率/%有机普通准确率/%
    蛋清偏最小二乘判别分析24625099.221225092.4
    支持向量机25025010022024993.8
    蛋黄偏最小二乘判别分析24524798.423323092.6
    支持向量机24224797.823423493.6
    Table 1. The identification results of category of eggs using original hyperspectral data of different ROI
    区域降维算法鉴别模型训练集预测集
    准确率
    /%
    准确率
    /%
    蛋黄连续投影
    算法
    偏最小二乘判别分析94.892.2
    支持向量机96.694.2
    竞争性自适应
    重加权算法
    偏最小二乘判别分析96.893.6
    支持向量机99.292.6
    Table 2. The identification results of category of eggs using dimension hyperspectral data in yolk ROI
    Ling-kai MA, Shi-ping ZHU, Yu-jie MIAO, Xiao WEI, Song LI, You-lie JIANG, Jia-xin ZHUO. The Discrimination of Organic and Conventional Eggs Based on Hyperspectral Technology[J]. Spectroscopy and Spectral Analysis, 2022, 42(4): 1222
    Download Citation