scholarly journals Identifying the Faces from Poor Quality Image / Video

2019 ◽  
Vol 8 (12) ◽  
pp. 1346-1353
Keyword(s):  
Median Filter ◽  
Light Condition ◽  
Principal Component ◽  
Motion Blur ◽  
K Nearest Neighbor ◽  
Angle Variation ◽  
Linear Discriminant ◽  
Illumination Problem ◽  
Wide Range ◽  
Pose Variation

Face biometric is becoming more popular because of its wide range of applications in authorizing the person either from an image or from the video sequence. The bottleneck in face recognition is Pose angle variation, varying light condition, Partial Occlusion, Blur in the image or Noise. The proposed method first removes the noise from the image using Adaptive Median Filter (AMF) then Discrete Cosine Transform(DCT) is applied to normalize the illumination problem. The algorithm is further used to remove the motion blur using Lucy Richardson’s method by calculating the Point Spread Function (PSF). The Pose variation problem is then addressed with Global Linear Regression(GLR). Then the Principal Component Analysis(PCA) and Linear Discriminant Analysis(LDA) are applied to the normalized image to get the feature vector. This combined feature score is used to recognize the image using K-Nearest Neighbor (K-NN). The result shows a maximum accuracy of 92% and 87.5% with Pose angle variation between (0°, 22°) and (22°, 40°) respectively. The pose variation greater than this shows an average accuracy of 77.5%. The result also shows a maximum computation speed of 0.018 Seconds.

A Self-Selected Movement Classification Method for Forearm Via sEMG and Attitude Sensor

2020 ◽  
Author(s):  
Lei Zhang
Keyword(s):  
Dimensional Space ◽  
Angular Acceleration ◽  
Principal Component ◽  
Classification Method ◽  
K Nearest Neighbor ◽  
External Interference ◽  
Linear Discriminant ◽  
Rehabilitation Training ◽  
Movement Intention ◽  
Attitude Sensor

Abstract Objective: The use of surface electromyography (sEMG) to realize the recognition of the movement intention can realize the control of the artificial hand or the robot, and can help the rehabilitation training for hemiplegia or muscle weakness. However, the sEMG are weak and susceptible to external interference, so the current research focuses on identifying certain types of movements. But once the subjects are changed, the recognition accuracy will greatly reduce. This study proposes a classification method which the subject could choose optional movements of forearm.Methods: Two sEMG sensors were used, and a 9-axis attitude sensor was added to the wrist. 8 different subjects participated in the experiment, and everyone selected 5 movements. The sEMG sensors were attached to the extensor pollicis brevis and the extensor digitorum. The sEMG features were: Standard Deviation (SD), Power Spectrum Density (PSD); attitude sensor features were: angle and angular acceleration in three dimensional space, and integral of angular acceleration. The results were classified and identified using Linear Discriminant Analysis (LDA), K-Nearest Neighbor (KNN), Decision Tree (DT) and Ensembles (En) algorithms. The results of using the sEMG, using the attitude sensor signals and combining the two were compared. Analysis of variance was conducted on the average accuracy. Features were reduced the dimension by the Principal Component Analysis (PCA), and the results of using PCA and not were compared. Results: The results showed that the combination of the two types of sensors could improve the recognition effect compared to the using sEMG sensor or the attitude sensor alone. The final recognition result was that KNN performed best, reaching 95.0%. The results of using PCA were more stable.Conclusion: The method could be used between different subjects, and the user could select the movements autonomously.Significance: This method can improve the adaptability of movement intention recognition based on sEMG, and has important significance for popularizing the use of the sEMG to control the manipulator or the prosthetic and the rehabilitation training.

scholarly journals A Study on the Psychological Analysis System Using Machine Learning

2018 ◽  
Vol 7 (3.33) ◽  
pp. 128
Author(s):  
Ki Young Lee ◽  
Kyu Ho Kim ◽  
Jeong Jin Kang ◽  
Sung Jai Choi ◽  
Yong Soon Im ◽  
...  
Keyword(s):  
Machine Learning ◽  
Nearest Neighbor ◽  
Smart Phone ◽  
Principal Component ◽  
Digital Data ◽  
Expression Recognition ◽  
K Nearest Neighbor ◽  
Linear Discriminant ◽  
Psychological Analysis ◽  
Analysis System

Real-time facial expression recognition and analysis technology is recently drawing attention in areas of computer vision, computer graphics, and HCI. Recognition of user’s emotion on the basis of video and voice is drawing particular interest. The technology may help managers of households or hospitals. In the present study, video and voice were converted into digital data through MATLAB by using PCA(Principal Component Analysis), LDA(Linear Discriminant Analysis), KNN(K Nearest Neighbor) algorithms to analyze emotions through machine learning. The manager of the psychological analysis counseling system may understand a user’s emotion in an smart phone environment. This system of the present study may help the manager to have a smooth conversation or develop a smooth relationship with a user on the basis of the provided psychological analysis results. 

The Role of Multivariant Analysis on the Interpretation of FTIR and Raman Spectra

2020 ◽  
pp. 104-129 ◽  
Author(s):  
Sefa Celik ◽  
Ali Tugrul Albayrak ◽  
Sevim Akyuz ◽  
Aysen E. Ozel
Keyword(s):  
Biological Systems ◽  
Principal Component ◽  
Biological Macromolecules ◽  
Spectroscopic Techniques ◽  
Linear Discriminant ◽  
Wide Range ◽  
Vibrational Bands ◽  
Multivariant Analysis

FTIR and Raman spectroscopy are complementary spectroscopic techniques that play an important role in the analysis of molecular structure and the determination of characteristic vibrational bands. Vibrational spectroscopy has a wide range of applications including mainly in physics and biology. Its applications have gained tremendous speed in the field of biological macromolecules and biological systems, such as tissue, blood, and cells. However, the vibrational spectra obtained from the biological systems contain a large number of data and information that make the interpretation difficult. To facilitate the analysis, multivariant analysis comprising the reduction of the dimension of spectrum data and classification of them by eliminating redundancy data, which are obtained from the spectra and does not have any role, becomes critical. In this chapter, the applications of Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and their combination PCA-LDA, which are widely used among multivariant techniques on biological systems will be disclosed.

scholarly journals Identification of Mint Scents Using a QCM Based E-Nose

2020 ◽  
Author(s):  
Salih Okur ◽  
Mohammed Sarheed ◽  
Robert Huber ◽  
Zejun Zhang ◽  
Lars Heinke ◽  
...  
Keyword(s):  
Nearest Neighbor ◽  
Principal Component ◽  
K Nearest Neighbor ◽  
Efficient Manner ◽  
Mentha Spicata ◽  
Electronic Noses ◽  
Linear Discriminant ◽  
Nepeta Cataria ◽  
Metal Oxide Nanoparticle ◽  
Nanoparticle Dispersions

Mints emit diverse scents that exert specific biological functions and are relevance for applications. The current work strives to develop electronic noses that can electronically discriminate the scents emitted by different species of Mint as alternative to conventional profiling by gas chromatography. Here, 12 different sensing materials including 4 different metal oxide nanoparticle dispersions (AZO, ZnO, SnO2, ITO), one Metal-Organic Frame as Cu(BPDC), and 7 different polymer films including PVA, PEDOT: PSS, PFO, SB, SW, SG, PB were used for functionalizing of QCM sensors. The purpose was to discriminate six economically relevant Mint species (Mentha x piperita, Mentha spicata, Mentha spicata ssp. crispa, Mentha longifolia, Agastache rugosa, and Nepeta cataria). The adsorption and desorption datasets obtained from each modified QCM sensor were processed by three different classification models including Principal Component Analy-sis (PCA), Linear Discriminant Analysis (LDA), and k-Nearest Neighbor Analysis (k-NN). This allowed discriminating the different Mints with classification accuracies of 97.2% (PCA), 100% (LDA), and 99.9% (k-NN), respectively. Prediction accuracies with a repeating test measurement reached up to 90.6% for LDA, and 85.6% for k-NN. These data demonstrate that this electronic nose can discriminate different Mint scents in a reliable and efficient manner.

scholarly journals Speech emotion recognition based on SVM and KNN classifications fusion

2021 ◽  
Vol 11 (2) ◽  
pp. 1259
Author(s):  
Mohammed Jawad Al Dujaili ◽  
Abbas Ebrahimi-Moghadam ◽  
Ahmed Fatlawi
Keyword(s):  
Speech Processing ◽  
English Language ◽  
Nearest Neighbor ◽  
Principal Component ◽  
Recognition System ◽  
Speech Emotion Recognition ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
Zero Crossing ◽  
Wide Range

Recognizing the sense of speech is one of the most active research topics in speech processing and in human-computer interaction programs. Despite a wide range of studies in this scope, there is still a long gap among the natural feelings of humans and the perception of the computer. In general, a sensory recognition system from speech can be divided into three main sections: attribute extraction, feature selection, and classification. In this paper, features of fundamental frequency (FEZ) (F0), energy (E), zero-crossing rate (ZCR), fourier parameter (FP), and various combinations of them are extracted from the data vector, Then, the principal component analysis (PCA) algorithm is used to reduce the number of features. To evaluate the system performance. The fusion of each emotional state will be performed later using support vector machine (SVM), K-nearest neighbor (KNN), In terms of comparison, similar experiments have been performed on the emotional speech of the German language, English language, and significant results were obtained by these comparisons.

A Robust Neonatal Facial Pain Cues Classification

2013 ◽  
Vol 475-476 ◽  
pp. 1110-1117
Author(s):  
Muhammad Naufal Mansor ◽  
Mohd Nazri Rejab
Keyword(s):  
Nearest Neighbor ◽  
Early Stage ◽  
Principal Component ◽  
Experimental Result ◽  
K Nearest Neighbor ◽  
Feed Forward Neural Network ◽  
Linear Discriminant ◽  
Feature Extraction Method ◽  
Pain State ◽  
Infant Pain

Late of infant pain detection on the early stage may affect newborns growth. Regarding of this matter, different techniques have been proposed such as facial expressions, speech production variation, and physiological signals to detect the pain states of a person. For past 2 decades, the determination of pain state through images has been undergone substantial research and development. Various techniques are used in the literature to classify pain states on the basis of images. In this paper, a feature extraction method using Principal Component Analysis (PCA) was adopted for identifying the pain states of an infant. In this study images samples are taken from Classification of Pain Expressions (COPE) database. Fuzzy k-NN, k Nearest Neighbor (k-NN), Feed Forward Neural network (FFNN) and Linear Discriminant analysis (LDA) based classifier is used to test usefulness of suggested features. Experimental result shows that the suggested methods can be used to identify the pain states of an infant.

scholarly journals Detecting Types of Sleep Apneas Through Nonlinear Features of Electrocardiogram and Electroencephalogram Signals

2021 ◽  
Author(s):  
Shaghayegh Saghafi ◽  
Fereidoun Nowshiravan Rahatabad ◽  
Keivan Maghooli
Keyword(s):  
Sleep Apnea ◽  
Nearest Neighbor ◽  
Principal Component ◽  
Majority Voting ◽  
Support Vector ◽  
Svm Classifier ◽  
Common Disease ◽  
K Nearest Neighbor ◽  
Linear Discriminant ◽  
Nonlinear Features

Purpose: Sleep apnea is a common disease among women, and mainly men. The most dangerous complication of this disorder is heart stroke. Other complications include insufficient sleep and resulting daytime tiredness and illness that affect the individual's activities during the day, disrupt their life. Therefore, identifying this disease is important. Materials and Methods: We used Electroencephalogram (EEG) and Electrocardiogram (ECG) channels from the data of 25 patients with sleep apnea, for each type of sleep apnea, 8 nonlinear-like features, including fractal dimension, correlation dimension, certainty, recurrence rate, mean diagonal lines, the entropy of recursive quantification analysis, sample Entropy, and Shannon entropy were extracted. Then, feature matrices were sorted using principal component analysis in the order of linear combination of features, and the 20 selected features were chosen, normalized using common methods, and fed to different classifiers. Two 5-class and 2-class classification methods were assessed. In the 5-classification, three classifiers were used; the support vector machine, k-nearest neighbor, and multilayer perceptron. Results: The results showed that the highest mean validity, accuracy, sensitivity, and specificity for the SVM classifier was 88.45%, 88.35%, 88.33%, and 88.32%, respectively. In the 2-class approach, in addition to the mentioned classifiers, linear discriminant analysis, Bayes, and majority voting were used, and each class was considered against all classes. The highest average validity, average accuracy, average sensitivity, average specificity using the majority rule voting was 94.35%, 94.30%, 94.32%, and 94.15% respectively. Conclusion: When the results of classifiers are combined with the majority voting method, the validity of identifying the classes increases. The average validity for this method was obtained at 94.42%, which was higher than several other studies. It is recommended that databases with a larger sample size be used. This would lead to increased reliability of the proposed analysis method. Moreover, using novel deep-learning-based methods could help obtain better results.

scholarly journals Rapid qualitative and quantitative detection of formaldehyde in squids based on colorimetric sensor array

2021 ◽  
Vol 233 ◽  
pp. 02021
Author(s):  
Binbin Guan ◽  
Hongmei Ding ◽  
Bin Chen ◽  
Mi Zhou ◽  
Zhaoli Xue
Keyword(s):  
Sensor Array ◽  
Recognition Rate ◽  
Principal Component ◽  
Colorimetric Sensor ◽  
Quantitative Detection ◽  
Ann Model ◽  
K Nearest Neighbor ◽  
Training Set ◽  
Linear Discriminant ◽  
Colorimetric Sensor Array

The colorimetric sensor array was used to detect the volatile organic compounds (VOCs) in squids with different formaldehyde content. In order to distinguish whether the formaldehyde is artificially added in the squids, the linear discriminant analysis (LDA) and K-nearest neighbor (KNN) based on principal component analysis (PCA) were used to make qualitative judgments, the result shows that the recognition rates of the training set and prediction set of the LDA model were 95% and 85% respectively, and the recognition rates of the training set and prediction set of the KNN model were both 90%. Moreover, error back propagation artificial neural network (BP-ANN) was used to quantitatively predict the concentration of formaldehyde in squids. The result indicates that the BP-ANN model acquired a good recognition rate with the correlation coefficient (Rp) for prediction was 0.9887 when the PCs was 10. To verify accuracy and applicability of the model, paired sample t-test was used to verify the difference between the predicted value of formaldehyde in the BP-ANN model and the actual addition amount. Therefore, this approach showed well potentiality to provide a fast, accuracy, no need for a pretreatment, and low-cost technique for detecting the formaldehyde in squids.

scholarly journals Comparison of Sensory Qualities in Eggs from Three Breeds Based on Electronic Sensory Evaluations

Foods ◽  
2021 ◽  
Vol 10 (9) ◽  
pp. 1984
Author(s):  
Xiaoguang Dong ◽  
Libing Gao ◽  
Haijun Zhang ◽  
Jing Wang ◽  
Kai Qiu ◽  
...  
Keyword(s):  
Electronic Nose ◽  
Classification Accuracy ◽  
Laying Hens ◽  
Electronic Tongue ◽  
Principal Component ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
Linear Discriminant ◽  
Sensory Qualities ◽  
Egg Yolks

The present study was conducted on three commercial laying breeder strains to evaluate differences of sensory qualities, including texture, smell, and taste parameters. A total of 140 eggs for each breed were acquired from Beinong No.2 (B) laying hens, Hy-Line Brown (H) laying hens, and Wuhei (W) laying hens. Sensory qualities of egg yolks and albumen from three breeds were detected and discriminated based on different algorithms. Texture profile analysis (TPA) showed that the eggs from three breeds had no differences in hardness, adhesiveness, springiness, and chewiness other than cohesiveness. The smell profiles measured by electronic nose illustrated that differences existed in all 10 sensors for albumen and 8 sensors for yolks. The taste profiles measured by electronic tongue found that the main difference of egg yolks and albumen existed in bitterness and astringency. Principal component analysis (PCA) successfully showed grouping of three breeds based on electronic nose data and failed in grouping based on electronic tongue data. Based on electronic nose data, linear discriminant analysis (LDA), fine k-nearest neighbor (KNN) and linear support vector machine (SVM) were performed to discriminate yolks, albumen, and the whole eggs with 100% classification accuracy. While based on electronic tongue data, the best classification accuracy was 96.7% for yolks by LDA and fine tree, 88.9% for albumen by LDA, and 87.5% for the whole eggs by fine KNN. The experiment results showed that three breeds’ eggs had main differences in smells and could be successfully discriminated by LDA, fine KNN, and linear SVM algorithms based on electronic nose.

scholarly journals BLACKFACE SURVEILLANCE CAMERA DATABASE FOR EVALUATING FACE RECOGNITION IN LOW QUALITY SCENARIOS

2017 ◽  
Vol 15 (2) ◽  
pp. 13-31
Author(s):  
A ABAYOMI-ALLI ◽  
E O OMIDIORA ◽  
S O OLABIYISI ◽  
J A Ojo ◽  
A Y AKINGBOYE
Keyword(s):  
Face Recognition ◽  
Real Life ◽  
Principal Component ◽  
Performance Comparison ◽  
Night Vision ◽  
Linear Discriminant ◽  
Surveillance Camera ◽  
Lighting Conditions ◽  
Recognition Algorithms ◽  
Pose Variation

Many face recognition algorithms perform poorly in real life surveillance scenarios because they were tested with datasets that are already biased with high quality images and certain ethnic or racial types. In this paper a black face surveillance camera (BFSC) database was described, which was collected from four low quality cameras and a professional camera. There were fifty (50) random volunteers and 2,850 images were collected for the frontal mugshot, surveillance (visible light), surveillance (IR night vision), and pose variations datasets, respectively. Images were taken at distance 3.4, 2.4, and 1.4 metres from the camera, while the pose variation images were taken at nine distinct pose angles with an increment of 22.5 degrees to the left and right of the subject. Three Face Recognition Algorithms (FRA), a commercially available Luxand SDK, Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) were evaluated for performance comparison in low quality scenarios. Results obtained show that camera quality (resolution), face-to-camera distance, average recognition time, lighting conditions and pose variations all affect the performance of FRAs. Luxand SDK, PCA and LDA returned an overall accuracy of 97.5%, 93.8% and 92.9% after categorizing the BFSC images into excellent, good and acceptable quality scales. 

Sign in / Sign up

Export Citation Format

Share Document