Paper
3 June 2024 Laser-induced fluorescence detection of oil film thickness in water based on convolutional neural network algorithm
Xiaofen Li, Lanjun Sun, Zhaojuan Zhang, Zhenyu Xing, Zhijian Liu
Author Affiliations +
Proceedings Volume 13182, 2024 International Conference on Optoelectronic Information and Optical Engineering (OIOE 2024); 131821Y (2024) https://doi.org/10.1117/12.3030731
Event: 2024 International Conference on Optoelectronic Information and Optical Engineering (OIOE 2024), 2024, Kunming, China
Abstract
Laser-induced fluorescence spectroscopy plays an important role in rapid detection and identification of oil spills at sea. The fluorescence spectra of different thicknesses of oil films of the same type of oil are very similar, which makes it difficult for the method to detect and recognize the thickness of oil films. In this paper, a laser-induced fluorescence detection method based on convolutional neural network algorithm is proposed to measure the thickness of oil films. Experimental studies of laser-induced fluorescence were conducted for different oil spill film thicknesses, and fluorescence spectral data were obtained for oil-free water and for 10 different oil film thicknesses. The recognition of 11 samples with different oil film thicknesses was realized by Matlab software using convolutional neural network algorithm. It was verified that the method can completely recognize different thicknesses of oil film with high accuracy and low training set error. Based on the convolutional neural network algorithm, the oil spill volume can be calculated by detecting the oil spill thickness, which has certain application value for the protection of marine environment.
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Xiaofen Li, Lanjun Sun, Zhaojuan Zhang, Zhenyu Xing, and Zhijian Liu "Laser-induced fluorescence detection of oil film thickness in water based on convolutional neural network algorithm", Proc. SPIE 13182, 2024 International Conference on Optoelectronic Information and Optical Engineering (OIOE 2024), 131821Y (3 June 2024); https://doi.org/10.1117/12.3030731
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Film thickness

Laser induced fluorescence

Detection and tracking algorithms

Fluorescence

Fluorescence spectroscopy

Convolutional neural networks

Education and training

Back to Top