The identification of tea varieties and producing areas has an increasingly important market value as the continuous development of the tea industry. A method combining laser induced breakdown spectroscopy (LIBS) and neural network algorithm is proposed in order to identify tea varieties and producing areas rapidly and accurately. In this paper, LIBS spectra of six major tea varieties and eight green tea samples from different producing areas in the range of 200-430 nm are collected, elemental analysis and spectra pre-processing are performed, principal component analysis (PCA) method is applied to select the features, and error back propagation (BP) neural network is used to model the tea classification problem. Key issues such as feature dimensions, network parameters, network structure, and the size of dataset are discussed. The result shows that the best model accuracy reaches 100%, indicating that this method can be used to establish a practicable tea classification model as long as sufficient data are obtained, which is feasible for solving the problems of the identification of tea varieties and producing areas.
This paper studied the data process methods of Laser-Induced Breakdown Spectroscopy (LIBS) and proposed a spectral intensity correction method utilizing laser energy monitoring and plasma morphology imaging, combined with neural network algorithm to improve the spectral stability. We set up a LIBS system, which had a laser beam sampling module to monitor laser output energy and a CMOS camera imaging module to capture plasma flame outlines. A back propagation neural network (BPNN) model was designed for standardizing the spectrum to a lower relative standard deviation (RSD) of emission line intensities, in which training spectrum, sample energy and image parameters were inputs, and the average of all the training spectrums were outputs. This method was applied in both aluminum and soil LIBS tests, and got effective results in reducing spectral intensity fluctuations. This data processing method not only provided a practical access to acquire stable spectrum information for both qualitative and quantitative LIBS analysis but also showed a bright future of combining LIBS data processing with machine learning methods.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.