Laser & Optoelectronics Progress, Volume. 57, Issue 9, 093006(2020)
Detection of Anthracnose in Camellia Oleifera Based on Laser-Induced Breakdown Spectroscopy
Figures & Tables(14)
Fig. 1. Sample of camellia oleifera leaves. (a) Healthy camellia oleifera leaves; (b) infected anthracnose camellia oleifera leaves
Fig. 4. Working curve of standard solution. (a) Healthy camellia oleifera leaves; (b) infected anthracnose camellia oleifera leaves
Fig. 7. PLS model and prediction model of Mn element after 7-point data smoothing and first derivative de-noising. (a) PLS model; (b)prediction model
Fig. 8. Best sub-interval selected by the iPLS model. (a) Spectral graph of the 24th interval with RMSECV; (b) spectral graph corresponding to the 6th sub-interval
Table 1. Determination conditions of Mn element
View tableTable 1. Determination conditions of Mn element
Test condition Wavelength /nm Lamp current /mA Acetylene flow rate /(L·min-1) Airflow rate /(L·min-1) Slit width /nm Parameter 279.5 3 1.3 7.5 0.2
Table 2. Division of samples
View tableTable 2. Division of samples
Sample Category Number of samples Range value /(mg·mg-1) Average value /(mg·mg-1) Healthy camellia oleifera leaves calibration set 181 1.0600-4.2570 2.2919 prediction set 59 1.4610-3.7240 2.2767 Camellia oleifera leaves with anthracnose calibration set 157 0.7990-3.3290 1.4476 prediction set 51 1.0680-2.7880 1.3581
Table 3. PLS model results of smoothed data processed by different preprocessing methods
View tableTable 3. PLS model results of smoothed data processed by different preprocessing methods
Spectral pretreatment method Evaluation index Before smoothing 5 points smoothing 7 points smoothing 9 points smoothing RC 0.8461 0.8621 0.8956 0.8760 Original RMSECV /(μg·mg-1) 0.2575 0.2499 0.2204 0.2433 RP 0.8195 0.8212 0.8540 0.8315 RMSEP /(μg·mg-1) 0.2769 0.2658 0.2548 0.2591 RC 0.8534 0.8562 0.8648 0.8605 Denoising RMSECV /(μg·mg-1) 0.2434 0.2375 0.2403 0.2482 RP 0.8162 0.8196 0.8329 0.8142 RMSEP /(μg·mg-1) 0.2553 0.2469 0.2373 0.2499 RC 0.8345 0.8440 0.8840 0.8396 Baseline correction RMSECV /(μg·mg-1) 0.2875 0.2570 0.1770 0.2764 RP /(μg·mg-1) 0.8010 0.8219 0.8333 0.8019 RMSEP /(μg·mg-1) 0.3169 0.2722 0.2524 0.2911 RC 0.8584 0.8947 0.9025 0.8892 First derivative de-noising RMSECV /(μg·mg-1) 0.2414 0.2206 0.2192 0.2227 RP 0.8215 0.8519 0.8882 0.8352 RMSEP /(μg·mg-1) 0.2692 0.2474 0.2356 0.2339 RC 0.8523 0.8521 0.8507 0.8619 Second derivative de-noising RMSECV /(μg·mg-1) 0.2528 0.2532 0.2569 0.2354 RP 0.8190 0.8150 0.8377 0.8323 RMSEP /(μg·mg-1) 0.2784 0.2791 0.2774 0.2802 RC 0.8323 0.8389 0.8429 0.8413 Normalization RMSECV /(μg·mg-1) 0.2829 0.2705 0.2711 0.2514 RP 0.8189 0.8106 0.8254 0.8177 RMSEP /(μg·mg-1) 0.3062 0.3098 0.2955 0.3016
Table 4. iPLS modeling analysis results
View tableTable 4. iPLS modeling analysis results
Interval number Optimum principal component RMSECV /(μg·mg-1) R Optimum interval 1 7 0.2770 0.8304 1 2 9 0.2580 0.8548 1 3 10 0.2360 0.8807 1 4 12 0.2200 0.8966 1 5 6 0.2510 0.8632 2 6 6 0.2490 0.8660 2 7 6 0.2560 0.8577 2 8 6 0.2550 0.8589 2 9 7 0.2330 0.8843 3 10 8 0.2350 0.8832 3 11 7 0.2380 0.8797 3 12 7 0.2400 0.8769 3 13 7 0.2200 0.8966 4 14 6 0.2260 0.8918 4 15 7 0.2240 0.8931 4 16 7 0.2270 0.8899 4 17 6 0.2220 0.8945 5 18 6 0.2210 0.8961 5 19 7 0.2170 0.8999 5 20 8 0.2200 0.8979 5 21 6 0.2350 0.8816 6 22 8 0.2160 0.9016 6 23 8 0.2160 0.9013 6 24 8 0.2090 0.9076 6 25 6 0.2280 0.8889 7 26 7 0.2100 0.9068 7 27 6 0.2140 0.9027 7 28 6 0.2090 0.9067 7 29 8 0.2700 0.8414 7 30 8 0.2190 0.8996 8
Tools
Get Citation
Copy Citation Text
Yande Liu, Xue Gao, Mengjie Cheng, Zhaoguo Hou, Xiaodong Lin, Jia Xu. Detection of Anthracnose in Camellia Oleifera Based on Laser-Induced Breakdown Spectroscopy[J]. Laser & Optoelectronics Progress, 2020, 57(9): 093006
Paper Information
Category: Spectroscopy
Received: Sep. 5, 2019
Accepted: Sep. 26, 2019
Published Online: May. 6, 2020
The Author Email: Yande Liu (jxliuyd@163.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20