A qualitative recognition method based on Karhunen–Loeve decomposition for near-infrared (NIR) and mid infrared (MIR) spectroscopy analysis

Zhao, Zhentang; Li, Bin; Liu, Wei; Yuan, Hairong; Song, Chunfeng

doi:10.1039/c9ay01969c

Cited by 4 publications

(1 citation statement)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These techniques often require specialist knowledge and involve environmentally harmful, time-consuming, and labour-intensive wet chemistry, which is not conducive to maintaining sample integrity. On the other hand, infrared(IR) spectroscopy is more conducive to non-destructive detection and identi cation of ink samples as a rapid and low-cost detection method [7].…”

Section: Introductionmentioning

confidence: 99%

Accurate felt-tip pen brands classification based on a convolutional neural network using data augmentation

Huang

Wang

Zhang

et al. 2023

Preprint

View full text Add to dashboard Cite

Ink analysis played an important role in document examination, but the limited dataset made it difficult for many algorithms to distinguish inks accurately. This paper aimed to evaluate the feasibility of two data augmentation(DA) methods, Gaussian noise data augmentation (GNDA) and extended multiplicative signal augmentation (EMSA), for the classification of felt-tip pen ink brands. Four brands of felt-tip pens were analysed using FTIR spectroscopy. Five classification models were used, convolutional neural network (CNN), K-nearest neighbor (KNN), support vector machine (SVM), random forest (RF), and partial least squares discriminant analysis (PLS-DA). The results showed that the datasets generated by GNDA and EMSA are similar to the original datasets and have some diversity. The EMSA method had optimal classification results when combined with CNN, with classification accuracy (ACC), precision (PRE), recall (REC) and F1 score reaching 99.86%, 99.87%, 99.86%, 99.86%, and 99.86%, compared with GNDA-CNN method(ACC = 80.90%, PRE = 87.34%, REC = 81.62%, F1 score = 79.23%). This study shows that when raw spectral data is small, DA methods can be combined with neural network models to identify ink brands effectively.

show abstract

Section: Introductionmentioning

confidence: 99%

Accurate felt-tip pen brands classification based on a convolutional neural network using data augmentation

Huang

Wang

Zhang

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

An Overview of Univariate and Multivariate Karhunen Loève Expansions in Statistics

Daw

Simpson

Wikle

et al. 2022

J Indian Soc Probab Stat

View full text Add to dashboard Cite

In-process Estimation of Ascorbic Acid Assay Using Machine Learning Algorithms

Felipe

Baldovino

2020

IOP Conf. Ser.: Mater. Sci. Eng.

View full text Add to dashboard Cite

Demand for greater volume of pediatric vitamin C syrups become higher every year whilst maintaining an expected high quality Current laboratory techniques employ costly time-consuming and destructive testing which inhibit manufacturers to attain higher process efficiency In this study machine learning methods: random forest RF support vector machine SVM artificial neural network ANN and adaptive neuro-fuzzy inference system ANFIS are used for the estimation of percentage ascorbic acid %AA assay Likewise these methods are compared to conventional linear regression LR model and to each other pH specific gravity viscosity and %AA assay measurements were used for the training of the network Preprocessing technique involving data smoothing was employed on each nonlinear main effect relationship to reduce the noise and achieve better prediction accuracy Upon training it was found the ANN coupled with Bayeasian Regularization exhibited 02314 MSE a higher accuracy among other algorithms Furthermore it generally pushed ANN to be relatively more accurate than ANFIS with a minimal MSE of 030810

show abstract

A qualitative recognition method based on Karhunen–Loeve decomposition for near-infrared (NIR) and mid infrared (MIR) spectroscopy analysis

Abstract: Qualitative recognition is an important research area of NIR and MIR spectroscopy analyses.

Cited by 4 publications

References 23 publications

Accurate felt-tip pen brands classification based on a convolutional neural network using data augmentation

Accurate felt-tip pen brands classification based on a convolutional neural network using data augmentation

An Overview of Univariate and Multivariate Karhunen Loève Expansions in Statistics

In-process Estimation of Ascorbic Acid Assay Using Machine Learning Algorithms

Contact Info

Product

Resources

About