Nondestructive detection of anthocyanin content in fresh leaves of
purple maize using hyperspectral data

Xin Yang; Shichen Gao; Xiaohe Gu; Chao Zhang; Chao Zhang; Chao Zhang; Qian Sun; Zhonghui Wei; Xueqian Hu; Xuzhou Qu

doi:10.1364/AO.458787

Applied Optics
Vol. 61,
Issue 21,
pp. 6213-6222
(2022)
•https://doi.org/10.1364/AO.458787

Nondestructive detection of anthocyanin content in fresh leaves of purple maize using hyperspectral data

Xin Yang, Shichen Gao, Xiaohe Gu, Chao Zhang, Qian Sun, Zhonghui Wei, Xueqian Hu, and Xuzhou Qu

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Xin Yang, Shichen Gao, Xiaohe Gu, Chao Zhang, Qian Sun, Zhonghui Wei, Xueqian Hu, and Xuzhou Qu, "Nondestructive detection of anthocyanin content in fresh leaves of purple maize using hyperspectral data," Appl. Opt. 61, 6213-6222 (2022)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Check for updates

Abstract

Anthocyanins are widely used in the food industry as an additive, improving antioxidant capacity and strengthening the human immune system. However, rapid and nondestructive detection methods are lacking. This study aimed to develop a rapid and nondestructive method to detect anthocyanin content in fresh purple maize leaves using hyperspectral reflectance. Sensitivity bands were screened by analyzing the correlation between the spectrum and anthocyanin, chlorophyll, and moisture content in maize leaves with models constructed. Through a combination of the sensitivity bands of the three components, the interference of chlorophyll and moisture on the spectral detection of anthocyanin in fresh leaves was analyzed. The results showed that the anthocyanin sensitivity band was approximately 550 nm. The determination coefficient and root mean square error of the optimal hyperspectral model were 0.766 and 4.215 mg/g, respectively. After excluding chlorophyll and moisture interference, the anthocyanin content detection accuracy was improved by only 2% compared to that of the original. These results indicate that hyperspectral technology can be used to nondestructively detect anthocyanin content in fresh purple maize leaves with good accuracy. Chlorophyll and moisture in the leaves did not significantly influence anthocyanin content.

Full Article | PDF Article

More Like This

Nondestructive nitrogen content estimation in tomato plant leaves by Vis-NIR hyperspectral imaging and regression data models

Razieh Pourdarbani, Sajad Sabzi, Mohammad H. Rohban, Ginés García-Mateos, and Juan I. Arribas
Appl. Opt. 60(30) 9560-9569 (2021)

Indirect quantitative analysis of soluble solid content in citrus by the leaves using hyperspectral imaging combined with machine learning

Lianbo Guo, Weihua Hunag, Weinan Zheng, Feng Chen, Weiliang Wang, Deng Zhang, Zhenlin Hu, and Yanwu Chu
Appl. Opt. 61(2) 491-497 (2022)

Nondestructive measurement of chlorophyll pigment content in plant leaves from three-color reflectance and transmittance

Norihide Yamada and Sadao Fujimura
Appl. Opt. 30(27) 3964-3973 (1991)

Previous Article Next Article

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (8)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (6)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (10)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Function	Expression
Linear	$Y = a X + b$
Quadratic	$Y = a X^{2} + b X + c$
Power	$Y = a X^{b}$
Exponential	$Y = a e^{b x}$
Logarithmic	$Y = a \ln (X) + b$

Components	Minimum	Maximum	Mean	Median	Deviation
Foliar					Standard
Anthocyanin	3.62	55.86	27.61	25.74	10.35
Chlorophyll	2.93	5.21	4.03	3.99	0.53
Moisture	637.79	773.45	725.99	726.71	26.29

	Anthocyanin		Chlorophyll		Moisture
Transformation	Sensitivity Band (nm)	Correlation Coefficient	Sensitivity Band (nm)	Correlation Coefficient	Sensitivity Band (nm)	Correlation Coefficient
P	548	−0.56	549	−0.52	975	−0.49
1/P	544	0.60	552	0.55	978	0.52
lg P	551	−0.62	563	−0.50	975	−0.48
1/lg P	546	0.58	549	0.53	978	0.54
$P^{'}$	550	0.68	746	0.63	947	−0.64
$(1 / P)^{'}$	550	−0.81	748	−0.70	947	0.72
$(l g P)^{'}$	552	0.77	750	0.64	947	−0.69
$(1 / l g P)^{'}$	558	−0.75	741	−0.68	949	0.66

Transformation	Sensitivity Band (nm)	Hyperspectral Model	$R^{2}$ (training sets)	$R^{2}$ (validating sets)
P	548	$\begin{array}{l} y = - 0.696 \times W_{548} + 0.213 \\ y = 22.147 \times {W_{548}}^{2} - 5.66 \times W_{548} + 0.274 \\ y = - 0.102 \times \ln (W_{548}) - 0.216 \\ y = 0.632 \times e^{- 32.911 \times W_{548}} \\ y = 2.528 \times 1 0^{- 5} \times {W_{548}}^{- 3.188} \end{array}$	0.311	0.306
			0.430	0.423
			0.329	0.308
			0.403	0.389
			0.426	0.410
1/P	544	$\begin{array}{l} y = 0.005 \times W_{544} - 0.049 \\ y = 0.001 \times {W_{544}}^{2} - 0.033 \times W_{544} + 0.264 \\ y = 0.088 \times \ln (W_{544}) - 0.097 \\ y = 0.002 \times e^{0.284 \times W_{544}} \\ y = 2.163 \times 1 0^{- 5} \times {W_{544}}^{3.152} \end{array}$	0.354	0.344
			0.452	0.413
			0.387	0.358
			0.461	0.453
			0.448	0.430
lg P	551	$\begin{array}{l} y = - 0.181 \times W_{551} - 0.155 \\ y = 1.044 \times {W_{551}}^{2} + 1.972 \times W_{551} + 0.948 \\ y = 0.182 \times \ln (W_{551}) + 0.027 \\ y = 2.064 \times 1 0^{- 5} \times e^{- 6.924 \times W_{551}} \end{array}$	0.352	0.344
			0.418	0.395
			0.340	0.310
			0.426	0.412
1/lg P	546	$\begin{array}{l} y = 0.181 \times W_{546} + 0.209 \\ y = 1.246 \times {W_{546}}^{2} + 2.641 \times W_{546} + 1.414 \\ y = - 0.182 \times \ln (W_{546}) + 0.027 \\ y = 54.395 \times e^{7.839 \times W_{546}} \end{array}$	0.329	0.322
			0.405	0.411
			0.340	0.319
			0.453	0.446
P^′	550	$\begin{array}{l} y = 550.666 \times W_{550} + 0.015 \\ y = 8823776 \times {W_{550}}^{2} - 86.835 \times W_{550} + 0.015 \\ y = 0.009 \times \ln (W_{550}) + 0.125 \\ y = 0.014 \times e^{18788.04 \times W_{550}} \\ y = 5382.409 \times {W_{550}}^{1.194} \end{array}$	0.511	0.497
			0.697	0.652
			0.361	0.339
			0.685	0.681
			0.596	0.602
$(1 / P)^{'}$	550	$\begin{array}{l} y = - 3.997 \times W_{550} + 0.015 \\ y = 297.008 \times {W_{550}}^{2} + 0.085 W_{550} + 0.018 \\ y = 0.009 \times \ln (W_{550}) + 0.083 \\ y = 0.016 \times e^{- 106.381 \times W_{550}} \\ y = 2.322 \times {W_{550}}^{0.8} \end{array}$	0.593	0.567
			0.716	0.719
			0.498	0.562
			0.772	0.766
			0.568	0.518
$(l g P)^{'}$	552	$\begin{array}{l} y = 114.560 \times W_{552} + 0.015 \\ y = 279210.7 \times {W_{552}}^{2} - 4.837 \times W_{552} + 0.017 \\ y = 0.009 \times \ln (W_{552}) + 0.114 \\ y = 0.016 \times e^{3265.226 \times W_{552}} \\ y = 203.058 \times {W_{552}}^{1.025} \end{array}$	0.575	0.553
			0.718	0.709
			0.316	0.277
			0.711	0.715
			0.616	0.610
$(1 / l g P)^{'}$	558	$\begin{array}{l} y = - 122.638 \times W_{558} + 0.020 \\ y = 451821.2 \times {W_{558}}^{2} + 14.711 \times W_{558} + 0.026 \\ y = 0.006 \times \ln (W_{558}) + 0.281 \\ y = 0.018 \times e^{- 4251.622 \times W_{558}} \\ y = 931.912 \times {W_{558}}^{1.207} \end{array}$	0.501	0.488
			0.718	0.713
			0.510	0.481
			0.717	0.698
			0.678	0.640

Transformation	Integration	Hyperspectral Model	$R^{2}$ (training sets)	$R^{2}$ (validating sets)
P	$P_{a n t} \times P_{c h l}$	$y = 2.632 \times 1 0^{- 5} \times {W_{548 \cdot 549}}^{- 1.322}$	0.424	0.381
1/P	$P_{a n t} - P_{c h l}$	$y = 1.671 \times 10^{- 6} \times {W_{544 \cdot 552}}^{2} - 0.0003 \times W_{544 \cdot 552} + 0.019$	0.443	0.426
lg P	$P_{a n t} / P_{c h l}$	$y = 0.082 \times {W_{551 \cdot 563}}^{80.119}$	0.421	0.397
1/lg P	$P_{a n t} + P_{c h l}$	$y = 3.518 \times {W_{546 \cdot 549}}^{- 7.352}$	0.447	0.415
P^′	$P_{a n t} - P_{c h l}$	$y = 0.016 \times e^{7273571 \times W_{550 \cdot 746}}$	0.709	0.691
$(1 / P)^{'}$	$P_{a n t} / P_{c h l}$	$y = 0.018 \times e^{2.017 \times W_{550 \cdot 748}}$	0.788	0.780
$(l g P)^{'}$	$P_{a n t} / P_{c h l}$	$y = 0.017 \times e^{1167405 \times W_{552 \cdot 750}}$	0.722	0.696
$(1 / l g P)^{'}$	$P_{a n t} / P_{c h l}$	$y = - 0.032 \times {W_{558 \cdot 741}}^{2} - 0.094 \times W_{558 \cdot 741} + 0.013$	0.719	0.711

Transformation	Integration	Hyperspectral Model	$R^{2}$ (training sets)	$R^{2}$ (validating sets)
P	$P_{a n t} / P_{m o i}$	$y = 5.612 \times {W_{548 \cdot 975}}^{2} - 2.421 \times W_{548 \cdot 975} + 0.409$	0.427	0.406
1/P	$P_{a n t} / P_{m o i}$	$y = 0.006 \times {W_{544 \cdot 978}}^{2} - 0.052 \times W_{544 \cdot 978} + 0.206$	0.459	0.425
lg P	$P_{a n t} / P_{m o i}$	$y = 0.195 \times {W_{551 \cdot 975}}^{2} - 0.929 \times W_{551 \cdot 975} + 1.512$	0.431	0.425
1/lg P	$P_{a n t} + P_{m o i}$	$y = 5769.827 \times {W_{546 \cdot 978}}^{- 9.359}$	0.450	0.420
$P^{'}$	$P_{a n t} - P_{m o i}$	$y = 0.014 \times e^{- 93414112 \times W_{550 \cdot 947}}$	0.698	0.656
$(1 / P)^{'}$	$P_{a n t} / P_{m o i}$	$y = 0.016 \times e^{- 94.750 \times W_{550.947}}$	0.778	0.765
$(l g P)^{'}$	$P_{a n t} / P_{m o i}$	$y = 0.009 \times e^{3160.344 \times W_{552.947}}$	0.724	0.701
$(1 / l g P)^{'}$	$P_{a n t} - P_{m o i}$	$y = 0.018 \times e^{- 6.274 \times W_{558.949}}$	0.721	0.717

Abstract

Data availability

Cited By

Figures (8)

Tables (6)

Equations (10)

Applied Optics