• Spectroscopy and Spectral Analysis
  • Vol. 40, Issue 11, 3508 (2020)
Meng-qi LÜ1、1、*, Yu-jie SONG1、1, Hai-yong WENG1、1, Da-wei SUN1、1, Xiao-ya DONG1、1, Hui FANG1、1, and Hai-yan CEN1、1
Author Affiliations
  • 1[in Chinese]
  • 11. College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2020)11-3508-07 Cite this Article
    Meng-qi LÜ, Yu-jie SONG, Hai-yong WENG, Da-wei SUN, Xiao-ya DONG, Hui FANG, Hai-yan CEN. Effect of Near Infrared Hyperspectral Imaging Scanning Speed on Prediction of Water Content in Arabidopsis[J]. Spectroscopy and Spectral Analysis, 2020, 40(11): 3508 Copy Citation Text show less
    RGB images of Arabidopsis phenotypes(a)—(f): Col-0 ecotype on 0~5 days of drought stress;(g)—(l): OSCA1 mutant genotypes on 0~5 days of drought stress
    Fig. 1. RGB images of Arabidopsis phenotypes
    (a)—(f): Col-0 ecotype on 0~5 days of drought stress;(g)—(l): OSCA1 mutant genotypes on 0~5 days of drought stress
    The relationship between water content of Arabidopsis thaliana and the number of days post drought stressGray color represents Col-1 and white color represents OSCA1; Changes are represented by the mean±standard deviation; * and ** represent the degree of significant differences based on Duncan test (* represents 0.03pp<0.03)
    Fig. 2. The relationship between water content of Arabidopsis thaliana and the number of days post drought stress
    Gray color represents Col-1 and white color represents OSCA1; Changes are represented by the mean±standard deviation; * and ** represent the degree of significant differences based on Duncan test (* represents 0.03<p<0.05; ** represents 0.01<p<0.03)
    Mean spectra of all Arabidopsis canopy (20 mm·s-1)
    Fig. 3. Mean spectra of all Arabidopsis canopy (20 mm·s-1)
    PLSR model results for different moving speeds with full spectra(a): 20 mm·s-1; (b): 30 mm·s-1; (c): 40 mm·s-1
    Fig. 4. PLSR model results for different moving speeds with full spectra
    (a): 20 mm·s-1; (b): 30 mm·s-1; (c): 40 mm·s-1
    The variation of spatial resolution caused by different scanning speed(a): 20 mm·s-1; (b): 30 mm·s-1; (c): 40 mm·s-1
    Fig. 5. The variation of spatial resolution caused by different scanning speed
    (a): 20 mm·s-1; (b): 30 mm·s-1; (c): 40 mm·s-1
    The performance of PLSR models of different scanning speed based on optimal wavelengths(a): 20 mm·s-1; (b): 30 mm·s-1; (c): 40 mm·s-1
    Fig. 6. The performance of PLSR models of different scanning speed based on optimal wavelengths
    (a): 20 mm·s-1; (b): 30 mm·s-1; (c): 40 mm·s-1
    预处理方法NoneMSCSGSG+log
    RC20.9200.9210.9150.914
    RMSEC0.0030.0030.0030.003
    RP20.9050.9070.9020.897
    RMSEP0.0030.0030.0030.003
    RPD3.053.283.193.12
    Table 1. The performance of PLSR model based on different spectral pretreatment methods (20 mm·s-1)
    扫描速度
    /(mm·s-1)
    最优特征波长/nm
    20901.23, 904.58, 907.93, 917.99, 928.05,
    944.82, 951.53, 975.01, 1 042.16, 1 237.22,
    1 338.32, 1 372.05, 1 402.42, 1 429.42,
    1 476.71, 1 567.98, 1 652.59, 1 669.52,
    1 679.68, 1 686.46, 1 689.85, 1 700.01,
    1 703.40
    30901.22, 904.58, 928.05, 968.30, 1 190.09,
    1 334.95, 1 453.06, 1 503.74, 1 564.60,
    1 652.59, 1 669.52, 1 693.24
    40901.22, 911.28, 917.99, 928.05, 1 200.19,
    1 240.59, 1 338.32, 1 372.05, 1 409.17,
    1 513.88, 1 574.74, 1 659.36, 1 672.91,
    1 683.07, 1 700.01
    Table 2. The summary of optimal wavelengths based on Successive projection algorithm (SPA)
    Meng-qi LÜ, Yu-jie SONG, Hai-yong WENG, Da-wei SUN, Xiao-ya DONG, Hui FANG, Hai-yan CEN. Effect of Near Infrared Hyperspectral Imaging Scanning Speed on Prediction of Water Content in Arabidopsis[J]. Spectroscopy and Spectral Analysis, 2020, 40(11): 3508
    Download Citation