Combined Estimation of Chlorophyll Content in Cotton Canopy Based on Hyperspectral Parameters and Back Propagation Neural Network

Ablet Ershat; Maimaitiaili Baidengsha; Sawut Mamat; Shenqun An

doi:10.3788/AOS201939.0930003

Acta Optica Sinica, Volume. 39, Issue 9, 0930003(2019)

Combined Estimation of Chlorophyll Content in Cotton Canopy Based on Hyperspectral Parameters and Back Propagation Neural Network

Ablet Ershat^1,2, Maimaitiaili Baidengsha⁴, Sawut Mamat^1,2,3、*, and Shenqun An⁵

Author Affiliations

¹ College of Resource and Environment Sciences, Xinjiang University, Urumqi, Xinjiang 830046, China

² Key Laboratory of Oasis Ecology of Ministry of Education, Urumqi, Xinjiang 830046, China

³ Key Laboratory of Xinjiang General Institutions of Higher Learning for Smart City and Environment Modeling, Urumqi, Xinjiang 830046, China

⁴ Institute of Nuclear and Biotechnologies, Xinjiang Academy of Agricultural Sciences, Urumqi, Xinjiang 830046, China

⁵ College of Resources and Environmental Sciences, China Agricultural University, Beijing 100193, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(7)

Fig. 1. Spectral reflectance of canopy at different conditions. (a) Spectral reflectance of cotton with different health conditions; (b) spectral reflectance of cotton with different phosphorus treatments; (c) spectral reflectance of different cotton cultivars

Download full size

Fig. 2. Correlation between transformation spectral curves and chlorophyll content in cotton. (a) Original spectrum; (b) continuum-removal spectrum; (c) first-order differential spectrum

Download full size

Fig. 3. Visual representation of autocorrelation matrix between spectral bands. (a) Original spectra; (b) continuum-removal spectra; (c) first-order differential spectra

Download full size

Fig. 4. 1∶1 fitting results between measured values and predicted values by BP neural network models. (a) BP neural network model based on REP parameters; (b) BP neural network model based on FDR-VI; (c) BP neural network model based on R-VI; (d) BP neural network model based on CR-VI

Download full size

Table 1. Hyperspectral parameters

View table

Table 1. Hyperspectral parameters

Transformation	Parameter	Full name	Ref.
Red edge parameter	REP	Red edge position	[7]
	D_λ_red	Red edge amplitude	[10]
	D_λ_min	Minimum amplitude	[11]
	S_red	Red edge area	[11]
	λ_min	Red gully	[10]
FDR-VI	BmSR	First derivative mSR	[12]
FDR-VI	BmND	First derivative mND	[12]
R-VI and CR-VI	mSR	Modified simple ratio index	[13]
	mND	Modified normalized difference index	[13]
	NDI	Red-edge normalized difference index	[14]
	DD	Double difference index	[15]
	MCARI	Modified chlorophyll absorption ratio index	[16]
	BGI	Blue/green pigment index	[13]
	TCARI	Transformed chlorophyll absorption in reflectance index	[16]
	CARI	Chlorophyll absorption ratio index	[17]

Table 2. Correlation between hyperspectral parameters and chlorophyll contents

View table

Table 2. Correlation between hyperspectral parameters and chlorophyll contents

Parameter	Correlation coefficient
CARI	-0.255(R-VI), -0.570(CR-VI)
MCARI	-0.036(R-VI),-0.284(CR-VI)
mSR	0.836(R-VI),0.782(CR-VI)
mND	0.836(R-VI),0.782(CR-VI)
NDI	0.868(R-VI),0.804(CR-VI)
DD	0.809(R-VI),0.765(CR-VI)
BGI	-0.408(R-VI),-0.229(CR-VI)
TCARI	-0.109(R-VI),-0.397(CR-VI)
REP	0.342
D_λ_red	0.528
D_λ_min	-0.01
S_red	0.543
λ_min	-0.09
BmSR	0.759
BmND	0.781

Table 3. Comparison of modeling results by different models
View table
Table 3. Comparison of modeling results by different models
Model Calibration Validation
R² RMSE RE /% R² RMSE RE /%
Red edge parameter 0.35 2.42 3.48 0.27 2.77 4.03
FDR-VI 0.53 2.11 2.54 0.56 2.19 2.89
R-VI 0.69 1.69 1.65 0.82 1.46 2.08
CR-VI 0.66 1.70 1.70 0.85 1.37 1.97

Tools

Get Citation

Copy Citation Text

Ablet Ershat, Maimaitiaili Baidengsha, Sawut Mamat, Shenqun An. Combined Estimation of Chlorophyll Content in Cotton Canopy Based on Hyperspectral Parameters and Back Propagation Neural Network[J]. Acta Optica Sinica, 2019, 39(9): 0930003

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites