Water Sample Classification and Fluorescence Component Identification Based on Fluorescence Spectrum

Qing Chen; Bin Tang; Junfeng Miao; Yan Zhou; Zourong Long; Jinfu Zhang; Jianxu Wang; Mi Zhou; Binqiang Ye; Mingfu Zhao; Nianbing Zhong

doi:10.3788/AOS221518

Acta Optica Sinica, Volume. 43, Issue 6, 0630001(2023)

Water Sample Classification and Fluorescence Component Identification Based on Fluorescence Spectrum

Qing Chen¹, Bin Tang^1、*, Junfeng Miao¹, Yan Zhou³, Zourong Long^1、**, Jinfu Zhang¹, Jianxu Wang¹, Mi Zhou¹, Binqiang Ye^1,2, Mingfu Zhao¹, and Nianbing Zhong¹

Author Affiliations

¹Chongqing Key Laboratory of Fiber Optic Sensor and Photodetector, Chongqing University of Technology, Chongqing 400054, China

²School of Microelectronics and Communication Engineering, Chongqing University, Chongqing 400044, China

³Tongliang District Environmental Protection Bureau of Chongqing, Chongqing 402560, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(11)

Fig. 1. Overall flow of PARAFAC method

Download full size

Fig. 2. CNN fast classification and recognition model

Download full size

Fig. 3. 3D-EEM classification network based on MobileNetV2

Download full size

Fig. 4. 3D-EEM fitting network based on CF-VGG11. (a) Structural diagram of single-layer CF-VGG11 network; (b) structural diagram of whole fitting network

Download full size

Fig. 5. Analysis process of PARAFAC method

Download full size

Fig. 6. Analysis results of DB obtained by PAFARAC. (a) Component maps; (b) results of split half verification corresponding to loading component maps

Download full size

Fig. 7. PARAFAC result of 3D-EEM spectrum and fitting component map obtained by CNN model. (a) WS water sample; (b) XCYY water sample

Download full size

Table 1. Water sample collection
View table
Table 1. Water sample collection
Type Label Number
Surface water DB 51
Treatment water of industrial wastewater FS 127
Inlet and outlet water of sewage treatment plant WS 37
Rural drinking water XCYY 58

Table 2. Spectral characteristics of EEM seen from comparison results of PARAFAC components and OpenFluor database

View table

Table 2. Spectral characteristics of EEM seen from comparison results of PARAFAC components and OpenFluor database

Type	Component	$λ_{e x}$ /nm	$λ_{e m}$ /nm	Fluorescent substance	Number of OpenFluor matches
DB	C1	370	464	Humic acid^［23］	11
	C2	315	384	Microbial humus^［24］	21
	C3	290，395	460	Terrestrial humus^［25］	4
	C4	360，395	526	Soil fulvic acid^［26］	1
FS	C1	350	428	Waste water collection tracer^［27］	17
	C2	275，400	480	Terrestrial humus^［25］	1
	C3	360，395	526	Soil fulvic acid^［26］	1
	C4	315	384	Microbial humus^［28］	7
WS	C1	<270，365	465	Terrestrial humus^［29］	2
	C2	345	409	Humus like substance^［30］	1
	C3	295	369	Microbial humus^［31］	8
	C4	275，420	488	Microbial humus^［32］	3
XCYY	C1	345	429	Anthropogenic humus^［33］	9
	C2	390	454	Fulvic acid and humus^［34］	1
	C3	<270，365	465	Terrestrial humus^［35］	4
	C4	360，395	526	Soil fulvic acid^［26］	1

Table 3. Results of model training

View table

Table 3. Results of model training

Network model	Accuracy of training set	Accuracy of test set	Loss value of training set	Loss value of test set
MobileNetV2_1	99.65	98.61	0.2500	2.0000
MobileNetV2_2	99.30	95.83	2.3800	12.6000
CF-VGG11	99.60	98.11	0.0879	0.0455

Table 4. Comparison between PARAFAC method and proposed model
View table
Table 4. Comparison between PARAFAC method and proposed model
Model Data quantity requirement Operation environment Time cost Analysis process
PARAFAC ≥20 MATLAB High Complex
MobileNetV2+CF-VGG11 ≥1 Python Low Simple

Tools

Get Citation

Copy Citation Text

Qing Chen, Bin Tang, Junfeng Miao, Yan Zhou, Zourong Long, Jinfu Zhang, Jianxu Wang, Mi Zhou, Binqiang Ye, Mingfu Zhao, Nianbing Zhong. Water Sample Classification and Fluorescence Component Identification Based on Fluorescence Spectrum[J]. Acta Optica Sinica, 2023, 43(6): 0630001

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites