scholarly journals Mid-Infrared Spectroscopy for Coffee Variety Identification: Comparison of Pattern Recognition Methods

2016 ◽  
Vol 2016 ◽  
pp. 1-7 ◽  
Author(s):  
Chu Zhang ◽  
Chang Wang ◽  
Fei Liu ◽  
Yong He
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Yunnan Province ◽  
Variety Identification ◽  
Support Vector ◽  
Hainan Province ◽  
Mid Infrared ◽  
Arabica Coffee ◽  
Robusta Coffee ◽  
Pattern Recognition Methods

The potential of using mid-infrared transmittance spectroscopy combined with pattern recognition algorithm to identify coffee variety was investigated. Four coffee varieties in China were studied, including Typica Arabica coffee from Yunnan Province, Catimor Arabica coffee from Yunnan Province, Fushan Robusta coffee from Hainan Province, and Xinglong Robusta coffee from Hainan Province. Ten different pattern recognition methods were applied on the optimal wavenumbers selected by principal component analysis loadings. These methods were classified as highly effective methods (soft independent modelling of class analogy, support vector machine, back propagation neural network, radial basis function neural network, extreme learning machine, and relevance vector machine), methods of medium effectiveness (partial least squares-discrimination analysis,Knearest neighbors, and random forest), and methods of low effectiveness (Naive Bayes classifier) according to the classification accuracy for coffee variety identification.

scholarly journals Electronic Nose-Based Technique for Rapid Detection and Recognition of Moldy Apples

2019 ◽  
Author(s):  
Wenshen Jia ◽  
Gang Liang ◽  
Hui Tian ◽  
Jing Sun ◽  
Cihui Wan
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Electronic Nose ◽  
Gas Sensors ◽  
Penicillium Expansum ◽  
Support Vector ◽  
Linear Discriminant ◽  
Pattern Recognition Methods ◽  
Group A ◽  
Group B

In this paper, PEN3 electronic nose was used to detect and recognize fresh and moldy apples (inoculated with Penicillium expansum and Aspergillusniger) taken Golden Delicious apples as model subject. Firstly, the apples were divided into two groups: apples only inoculated with different molds (Group A) and mixed apples of inoculated apples with fresh apples (Group B). Then the characteristic gas sensors of the PEN3 electronic nose that were most closely correlated with the flavor information of the moldy apples were optimized and determined, which can simplify the analysis process and improve the accuracy of results. Four pattern recognition methods, including linear discriminant analysis (LDA), backpropagation neural network (BPNN), support vector machines (SVM) and radial basis function neural network (RBFNN), were then applied to analyze the data obtained from the characteristic sensors, respectively, aiming at establishing the prediction model of flavor information and fresh/moldy apples. The results showed that only the gas sensors of W1S, W2S, W5S, W1W and W2W in the PEN3 electronic nose exhibited strong signal response to the flavor information, indicating were most closely correlated with the characteristic flavor of apples and thus the data obtained from these characteristic sensors was used for modeling. The results of the four pattern recognition methods showed that BPNN presented the best prediction performance for the training and validation sets for both the Group A and Group B, with prediction accuracies of 96.29% and 90.00% (Group A), 77.70% and 72.00% (Group B), respectively. Therefore, it first demonstrated that PEN3 electronic nose can not only effectively detect and recognize the fresh and moldy apples, but also can distinguish apples inoculated with different molds.

scholarly journals Electronic Nose-Based Technique for Rapid Detection and Recognition of Moldy Apples

Sensors ◽  
2019 ◽  
Vol 19 (7) ◽  
pp. 1526 ◽  
Author(s):  
Wenshen Jia ◽  
Gang Liang ◽  
Hui Tian ◽  
Jing Sun ◽  
Cihui Wan
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Electronic Nose ◽  
Gas Sensors ◽  
Penicillium Expansum ◽  
Support Vector ◽  
Linear Discriminant ◽  
Pattern Recognition Methods ◽  
Group A ◽  
Group B

In this study, the PEN3 electronic nose was used to detect and recognize fresh and moldy apples inoculated with Penicillium expansum and Aspergillus niger, taking Golden Delicious apples as the model subject. Firstly, the apples were divided into two groups: individual apple inoculated only with/without different molds (Group A) and mixed apples of inoculated apples with fresh apples (Group B). Then, the characteristic gas sensors of the PEN3 electronic nose that were most closely correlated with the flavor information of the moldy apples were optimized and determined to simplify the analysis process and improve the accuracy of the results. Four pattern recognition methods, including linear discriminant analysis (LDA), backpropagation neural network (BPNN), support vector machines (SVM), and radial basis function neural network (RBFNN), were applied to analyze the data obtained from the characteristic sensors, aiming at establishing the prediction model of the flavor information and fresh/moldy apples. The results showed that only the gas sensors of W1S, W2S, W5S, W1W, and W2W in the PEN3 electronic nose exhibited a strong signal response to the flavor information, indicating most were closely correlated with the characteristic flavor of apples and thus the data obtained from these characteristic sensors were used for modeling. The results of the four pattern recognition methods showed that BPNN had the best prediction performance for the training and testing sets for both Groups A and B, with prediction accuracies of 96.3% and 90.0% (Group A), 77.7% and 72.0% (Group B), respectively. Therefore, we demonstrate that the PEN3 electronic nose not only effectively detects and recognizes fresh and moldy apples, but also can distinguish apples inoculated with different molds.

Classification and Prediction of Goldfish Population and Water Quality Using a Potentiometric E-Tongue

2017 ◽  
Vol 60 (4) ◽  
pp. 1037-1044
Author(s):  
Zhenbo Wei ◽  
Yu Zhao ◽  
Jun Wang
Keyword(s):  
Neural Network ◽  
Water Quality ◽  
Pattern Recognition ◽  
Comprehensive Evaluation ◽  
Back Propagation ◽  
Correlation Coefficients ◽  
Principal Component ◽  
Quality Parameters ◽  
Water Quality Parameters ◽  
Pattern Recognition Methods

Abstract. In this study, a potentiometric E-tongue was employed for comprehensive evaluation of water quality and goldfish population with the help of pattern recognition methods. Four water quality parameters, i.e., pH and concentrations of dissolved oxygen (DO), nitrite (NO2-N), and ammonium (NH3-N), were tested by conventional analysis methods. The differences in water quality parameters between samples were revealed by two-way analysis of variance (ANOVA). The cultivation days and goldfish population were classified well by principal component analysis (PCA) and canonical discriminant analysis (CDA), and the distribution of each sample was clearer in CDA score plots than in PCA score plots. The cultivation days, goldfish population, and water parameters were predicted by a T-S fuzzy neural network (TSFNN) and back-propagation artificial neural network (BPANN). BPANN performed better than TSFNN in the prediction, and all fitting correlation coefficients were >0.90. The results indicated that the potentiometric E-tongue coupled with pattern recognition methods could be applied as a rapid method for the determination and evaluation of water quality and goldfish population. Keywords: Classify, E-tongue, Goldfish water, Prediction.

scholarly journals Eggshell crack detection based on acoustic impulse response and supervised pattern recognition

2009 ◽  
Vol 27 (No. 6) ◽  
pp. 393-402 ◽  
Author(s):  
H. Lin ◽  
J. Zhao ◽  
Q. Chen ◽  
J. Cai ◽  
P. Zhou
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Artificial Neural Network ◽  
Support Vector Machine ◽  
Crack Detection ◽  
Support Vector Machine Model ◽  
Acoustic Resonance ◽  
Support Vector ◽  
Machine Model ◽  
Nearest Neighbours

A system based on acoustic resonance was developed for eggshell crack detection. It was achieved by the analysis of the measured frequency response of eggshell excited with a light mechanism. The response signal was processed by recursive least squares adaptive filter, which resulted in the signal-to-noise ratio of the acoustic impulse response reing remarkably enhanced. Five features variables were exacted from the response frequency signals. To develop a robust discrimination model, three pattern recognition algorithms (i.e. K-nearest neighbours, artificial neural network, and support vector machine) were examined comparatively in this work. Some parameters of the model were optimised by cross-validation in the building model. The experimental results showed that the performance of the support vector machine model is the best in comparison to k-nearest neighbours and artificial neural network models. The optimal support vector machine model was obtained with the identification rates of 95.1% in the calibration set, and 97.1% in the prediction set, respectively. Based on the results, it was concluded that the acoustic resonance system combined with the supervised pattern recognition has a significant potential for the cracked eggs detection.

scholarly journals The application of pattern recognition techniques in metabolite fingerprinting of six differentPhyllanthusspp.

2011 ◽  
Vol 26 (1) ◽  
pp. 69-78 ◽  
Author(s):  
Saravanan Dharmaraj ◽  
Lay-Harn Gam ◽  
Shaida Fariza Sulaiman ◽  
Sharif Mahsufi Mansor ◽  
Zhari Ismail
Keyword(s):  
Pattern Recognition ◽  
Principal Component ◽  
Canonical Variate ◽  
Batch Mode ◽  
Mid Infrared ◽  
Linear Discriminant ◽  
Pattern Recognition Techniques ◽  
Pattern Recognition Methods ◽  
Ftir Spectrometer ◽  
Component Models

FTIR spectroscopy was used together with multivariate analysis to distinguish six different species ofPhyllanthus. Among these speciesP. niruri,P. debilisandP. urinariaare morphologically similar whereasP. acidus,P. emblicaandP. myrtifoliusare different. The FTIR spectrometer was used to obtain the mid-infrared spectra of the dried powdered leaves in the region of 400–4000 cm−1. The region of 400–2000 cm−1was analyzed with four different pattern recognition methods. Initially, principal component analysis (PCA) was used to reduce the spectra to six principal components and these variables were used for linear discriminant analysis (LDA). The second technique used LDA on most discriminating wavenumber variables as searched by genetic algorithm using canonical variate approach for either 30 or 60 generations. SIMCA, which consisted of constructing an enclosure for each species using separate principal component models, was the third technique. Finally, multi-layer neural network with batch mode of backpropagation learning was used to classify the samples. The best results were obtained with GA of 60 gens. When LDA was run with the six wavenumbers chosen (1151, 1578, 1134, 609, 876 and 1227), 100% of the calibration spectra and 96.3% of the validation spectra were correctly assigned.

ELECTROCARDIOGRAM ARRHYTHMIA PATTERN RECOGNITION BASED ON AN IMPROVED WAVELET NEURAL NETWORK

2013 ◽  
Vol 13 (01) ◽  
pp. 1350018 ◽  
Author(s):  
GUANGYING YANG
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Pattern Recognition ◽  
Wavelet Neural Network ◽  
The Body ◽  
Support Vector ◽  
Ventricular Premature Beat ◽  
Ecg Signal ◽  
Data Set ◽  
Bundle Branch Block

Electrocardiography (ECG) is a transthoracic interpretation of the electrical activity of the heart over a period of time, as detected by electrodes attached to the outer surface of the skin and recorded by a device external to the body. ECG signal classification is very important for the clinical detection of arrhythmia. This paper presents an application of an improved wavelet neural network structure to the classification of the ECG beats, because of the high precision and fast learning rate. Feature extraction method in this paper is wavelet transform. Our experimental data set is taken from the MIT-BIH arrhythmia database. The correct detection rate of QRS wave is 95% by testing the data of MIT-BIH database. The proposed methods are applied to a large number of ECG signals consisting of 600 training samples and 120 test samples from the MIT-BIH database. The samples equally represent six different ECG signal types, including normal beat, atrial premature beat, ventricular premature beat, left bundle branch block, right bundle branch block and paced beat. In comparison with pattern recognition methods of BP neural networks, RBF neural networks and Support Vector Machines (SVM), the results in this experiment prove that the wavelet neural network method has a better recognition rate when classifying electrocardiogram signals. The experimental results prove that supposed method in this paper is effective for arrhythmia pattern recognition field.

DMiner-I: A software tool of data mining and its applications

Robotica ◽  
2002 ◽  
Vol 20 (5) ◽  
pp. 499-508
Author(s):  
Jie Yang ◽  
Chenzhou Ye ◽  
Nianyi Chen
Keyword(s):  
Neural Network ◽  
Data Mining ◽  
Genetic Algorithm ◽  
Pattern Recognition ◽  
Support Vector Machine ◽  
Knowledge Representation ◽  
Decision Trees ◽  
Software Tool ◽  
Support Vector ◽  
Function Models

SummaryA software tool for data mining (DMiner-I) is introduced, which integrates pattern recognition (PCA, Fisher, clustering, HyperEnvelop, regression), artificial intelligence (knowledge representation, decision trees), statistical learning (rough set, support vector machine), and computational intelligence (neural network, genetic algorithm, fuzzy systems). It consists of nine function models: pattern recognition, decision trees, association rule, fuzzy rule, neural network, genetic algorithm, HyperEnvelop, support vector machine and visualization. The principle, algorithms and knowledge representation of some function models of data mining are described. Nonmonotony in data mining is dealt with by concept hierarchy and layered mining. The software tool of data mining is realized byVisual C++under Windows 2000. The software tool of data mining has been satisfactorily applied in the prediction of regularities of the formation of ternary intermetallic compounds in alloy systems, and diagnosis of brain glioma.

scholarly journals Identifying Health Status in Grazing Dairy Cows from Milk Mid-Infrared Spectroscopy by Using Machine Learning Methods

Animals ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. 2154
Author(s):  
Brenda Contla Hernández ◽  
Nicolas Lopez-Villalobos ◽  
Matthieu Vignes
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Milk Sample ◽  
Time Window ◽  
Early Stage ◽  
Machine Learning Techniques ◽  
Cow Milk ◽  
Economic Losses ◽  
Support Vector ◽  
Mid Infrared

The early detection of health problems in dairy cattle is crucial to reduce economic losses. Mid-infrared (MIR) spectrometry has been used for identifying the composition of cow milk in routine tests. As such, it is a potential tool to detect diseases at an early stage. Partial least squares discriminant analysis (PLS-DA) has been widely applied to identify illness such as lameness by using MIR spectrometry data. However, this method suffers some limitations. In this study, a series of machine learning techniques—random forest, support vector machine, neural network (NN), convolutional neural network and ensemble models—were used to test the feasibility of identifying cow sickness from 1909 milk sample MIR spectra from Holstein-Friesian, Jersey and crossbreed cows under grazing conditions. PLS-DA was also performed to compare the results. The sick cow records had a time window of 21 days before and 7 days after the milk sample was analysed. NN showed a sensitivity of 61.74%, specificity of 97% and positive predicted value (PPV) of nearly 60%. Although the sensitivity of the PLS-DA was slightly higher than NN (65.6%), the specificity and PPV were lower (79.59% and 15.25%, respectively). This indicates that by using NN, it is possible to identify a health problem with a reasonable level of accuracy.

scholarly journals Detection and Classification of Breast Cancer in Mammography Images Using Pattern Recognition Methods

2019 ◽  
Vol 3 (4) ◽  
pp. 13-24 ◽  
Author(s):  
Naser Safdarian ◽  
Mohammadreza Hedyezadeh
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Clustering Algorithm ◽  
Probabilistic Neural Network ◽  
Screening Mammography ◽  
Fuzzy Classification ◽  
Support Vector ◽  
Svm Classifier ◽  
Final Decision ◽  
Pattern Recognition Methods

Introduction: In this paper, a method is presented to classify the breast cancer masses according to new geometric features. Methods: After obtaining digital breast mammogram images from the digital database for screening mammography (DDSM), image preprocessing was performed. Then, by using image processing methods, an algorithm was developed for automatic extracting of masses from other normal parts of the breast image. In this study, 19 final different features of each image were extracted to generate the feature vector for classifier input. The proposed method not only determined the boundary of masses but also classified the type of masses such as benign and malignant ones. The neural network classification methods such as the radial basis function (RBF), probabilistic neural network (PNN), and multi-layer perceptron (MLP) as well as the Takagi-Sugeno-Kang (TSK) fuzzy classification, the binary statistic classifier, and the k-nearest neighbors (KNN) clustering algorithm were used for the final decision of mass class. Results: The best results of the proposed method for accuracy, sensitivity, and specificity metrics were obtained 97%±4.36, 100%±0 and 96%±5.81, respectively for support vector machine (SVM) classifier. Conclusions: By comparing the results of the proposed method with the results of the other previous methods, the efficiency of the proposed algorithm was reported.

Digital Auscultation System of Traditional Chinese Medicine and Its Signals Acquisition

2011 ◽  
pp. 183-193
Author(s):  
Fanpeng Zhou ◽  
Jianjun Yan ◽  
Yiqin Wang ◽  
Fufeng Li ◽  
Chunming Xia ◽  
...  
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Signal Processing ◽  
Chinese Medicine ◽  
Traditional Chinese Medicine ◽  
New Technology ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Pattern Recognition Methods ◽  
Linear And Nonlinear

Digital auscultation of Traditional Chinese Medicine (TCM) is a relatively new technology which has been developed for several years. This system makes diagnoses by analyzing sound signals of patients using signal processing and pattern recognition. The paper discusses TCM auscultation in both traditional and current digital auscultation methods. First, this article discusses demerits of traditional TCM auscultation methods. It is through these demerits that a conclusion is drawn that digital auscultation of TCM is indispensable. Then this article makes an introduction to voice analysis methods from linear and nonlinear analysis aspects to pattern recognition methods in common use. Finally this article establishes a new TCM digital auscultation system based on wavelet analysis and Back-propagation neural network (BPNN).

Sign in / Sign up

Export Citation Format

Share Document