Laser & Optoelectronics Progress, Volume. 59, Issue 5, 0530001(2022)
Estimation of Chlorophyll Content of Long-Staple Cotton Based on Canopy Spectrum Characteristics
Figures & Tables(13)
Fig. 3. Spectral characteristics of cotton leaves. (a) Spectral reflectance of long-staple cotton with different chlorophyll contents; (b) spectral reflectance of different chlorophyll contents in the visible light range
Fig. 4. Changes of spectrum after mathematical transformation. (a) R; (b) R′; (c) Rcr
Fig. 5. Correlation between chlorophyll contents and reflectances. (a) R; (b) R′; (c) Rcr
Fig. 7. Estimation R2 of SPAD using different SVM kernels. (a) Calibration dataset; (b) validation dataset
Fig. 8. SVM regression model's fitting analysis result diagrams of measured and predicted values
Table 1. Index calculation formulas
View tableTable 1. Index calculation formulas
Index type Formula Reference VI SSR=R750/R710 ［ 29 ］NNDVI=（R750- R705）/ R750+R705） ［ 30 ］MMTCI=（R750-R710）/（R710-R680） ［ 31 ］CCARI=［（R700-R670）-0.2×（R700-R550）］ ［ 32 ］OSI RRSI（λ1，λ2）=Rλ1/Rλ2 DDSI（λ1，λ2）=Rλ1-Rλ2 ［ 33 ］NNDSI（λ1，λ2）=（Rλ1-Rλ2）/（Rλ1+Rλ2）
Table 2. Cotton spectral characteristic bands and correlation coefficients
View tableTable 2. Cotton spectral characteristic bands and correlation coefficients
Algorithm Characteristic band /nm Correlation coefficient Mean R 712， 713， 714， 715， 716， ‒0.43， 0.43， ‒0.43， ‒0.43， ‒0.43， 0.43 717， 718， 719， 720， 721 ‒0.43， ‒0.43， ‒0.43， ‒0.43， ‒0.43 R′ 513， 524， 527， 572， 589， ‒0.53， ‒0.56， ‒0.58， 0.55， 0.53， 0.54 691， 692， 697， 698， 699 ‒0.53， ‒0.51， ‒0.54， ‒0.54， ‒0.53 Rcr 761， 762， 763， 764， 817， ‒0.50， ‒0.49， ‒0.46， ‒0.46， ‒0.47， 0.47 818， 823， 829， 830， 831 ‒0.47， ‒0.46， ‒0.48， ‒0.47， ‒0.46
Table 3. Correlation coefficients between vegetation index and chlorophyll content
View tableTable 3. Correlation coefficients between vegetation index and chlorophyll content
Vegetation index R R′ Rcr SR 0.35** 0.41** 0.35** NDVI 0.27* 0.46** 0.28* MTCI 0.49** 0.46** 0.49** CARI -0.52** -0.51** -0.42**
Table 4. Correlation between chlorophyll contents and optimal spectral indices
View tableTable 4. Correlation between chlorophyll contents and optimal spectral indices
Optimized spectral index Band combination Correlation coefficient Mean RSI 942 945， 941 947，950 956， 1028 1051，928 1051 0.48， 0.43，0.42， 0.39，0.39 0.422 DSI 716 719， 693 936，565 738， 610 740，636 741 0.77， 0.76，0.75， 0.74，0.73 0.75 NDSI 941 952， 941 947，956 950， 928 1051，1028 1051 0.46， 0.43，0.42， 0.39，0.38 0.418
Table 5. Comparison of modeling results of SVM regression model RBF kernel type
View tableTable 5. Comparison of modeling results of SVM regression model RBF kernel type
Model Item Calibration Validation R2 RRMSE RRE R2 RRMSE RRE Algorithm R 0.41 2.88 1.26 0.29 1.53 0.43 R′ 0.72 1.99 0.72 0.84 0.72 0.14 Rcr 0.39 2.91 1.1 0.83 0.69 0.16 VI R4 0.51 2.61 1.1 0.64 1.11 0.31 R′4 0.49 2.74 1.18 0.43 1.37 0.39 Rcr4 0.38 2.96 1.36 0.50 1.29 0.35 OSI DDSI 0.49 3.02 1.34 0.23 1.6 0.52 NNDSI 0.45 2.82 1.09 0.58 1.19 0.25 RRSI 0.53 2.62 0.99 0.58 1.18 0.24
Tools
Get Citation
Copy Citation Text
Arkin Ansardin, Sawut Mamat, Jinzhao Li. Estimation of Chlorophyll Content of Long-Staple Cotton Based on Canopy Spectrum Characteristics[J]. Laser & Optoelectronics Progress, 2022, 59(5): 0530001
Paper Information
Category: Spectroscopy
Received: Apr. 20, 2021
Accepted: May. 27, 2021
Published Online: Feb. 22, 2022
The Author Email: Sawut Mamat (korxat@xju.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20