• Spectroscopy and Spectral Analysis
  • Vol. 42, Issue 5, 1567 (2022)
Ruo-chen DAI1、*, Huan TANG2、2; *;, Bin TANG1、1; *;, Ming-fu ZHAO1、1;, Li-yong DAI1、1;, Ya ZHAO3、3;, Zou-rong LONG1、1;, and Nian-bing ZHONG1、1;
Author Affiliations
  • 11. Chongqing Key Laboratory of Fiber Optic Sensor and Photodetector, Chongqing University of Technology, Chongqing 400054, China
  • 22. Key Scientific Research Base of Pest and Mold Control of Museum Collections of National Cultural Heritage Administration, Chongqing China Three Gorges Museum, Chongqing 400015, China
  • 33. Chongqing University of Education, Chongqing 400065, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2022)05-1567-05 Cite this Article
    Ruo-chen DAI, Huan TANG, Bin TANG, Ming-fu ZHAO, Li-yong DAI, Ya ZHAO, Zou-rong LONG, Nian-bing ZHONG. Study on Detection Method of Foxing on Paper Artifacts Based on Hyperspectral Imaging Technology[J]. Spectroscopy and Spectral Analysis, 2022, 42(5): 1567 Copy Citation Text show less
    Paper cultural relics foxing detection system
    Fig. 1. Paper cultural relics foxing detection system
    (a)Part of foxing sample; (b), (c) Super depth of field micrograph
    Fig. 2. (a)Part of foxing sample; (b), (c) Super depth of field micrograph
    Paper cultural relics reflection spectrum and average reflection spectrum
    Fig. 3. Paper cultural relics reflection spectrum and average reflection spectrum
    Paper cultural relics feature band image(a): 464 nm band image; (b): 787 nm band image
    Fig. 4. Paper cultural relics feature band image
    (a): 464 nm band image; (b): 787 nm band image
    Calculation results of band math of paper cultural relics(a): 464~767 nm image; (b): 464/767 nm image; (c): 767~464 nm image; (d): 767~464 nm image
    Fig. 5. Calculation results of band math of paper cultural relics
    (a): 464~767 nm image; (b): 464/767 nm image; (c): 767~464 nm image; (d): 767~464 nm image
    The images of the simulating sample after MNF transformation(a): MNF-band 1; (b): MNF-band 2; (c): MNF-band 7; (d): RGB image after MNF transformation
    Fig. 6. The images of the simulating sample after MNF transformation
    (a): MNF-band 1; (b): MNF-band 2; (c): MNF-band 7; (d): RGB image after MNF transformation
    类型KNNBP神经网络
    训练集
    判别率/%
    测试集
    判别率/%
    训练集
    判别率/%
    测试集
    判别率/%
    染病部分92.279.195.283.8
    纸张部分65.257.163.263.6
    墨迹部分706083.383.3
    总体83.373.388.385
    Table 1. Results of the KNN and BP models for different parts of simulating sample
    Ruo-chen DAI, Huan TANG, Bin TANG, Ming-fu ZHAO, Li-yong DAI, Ya ZHAO, Zou-rong LONG, Nian-bing ZHONG. Study on Detection Method of Foxing on Paper Artifacts Based on Hyperspectral Imaging Technology[J]. Spectroscopy and Spectral Analysis, 2022, 42(5): 1567
    Download Citation