Search

Communication of the human brain with the surroundings became reality by using Brain- Computer Interface (BCI) based mechanism. Electroencephalography (EEG) being the non-invasive method has become popular for interaction with the brain. Traditionally, the devices were used for clinical applications to detect various brain diseases but with the advancement in technologies, companies like Emotiv, NeuoSky are coming up with low cost, easily portable EEG based consumer graded devices that can be used in various application domains like gaming, education etc as these devices are comfortable to wear also. This paper reviews the fields where the EEG has shown its impact and the way it has proved useful for individuals with severe motor disorder, rehabilitation and has become a means of communication to the real world. This paper investigates the use of Cubic SVM algorithm In the EEG classification. EEG feature extraction is Implemented by maximum overlap discrete wavelet transform (MODWT) to reduce the dimensionality of data. The Sliding Window Technique is used to calculate the mean within each window samples. The feature vectors are loaded into the support vector machine (SVM) and optimize tree.

ML (Machine learning) is a subset of AI and also improved learning technique has different performance in result over the conventional ML determining the complexity in structures of dimensional data. ADHD is one of the most important neurological disorders and it is represented by different symptoms and we can extract useful the information from FMRI time series. In this paper the ADHD identification and classification is obtained by machine learning techniques. This paper explores an artificial intelligence in unsupervised learning is appropriate to learn features from raw data. The proposed system presented with two stage approaches for ADHD diagnosis which associated SoftMax Regression and SVM fine tuning approach. In the implementation part used FMRI brain images are data sets. The two stage approach shows the high accuracy in performance by using the learning techniques.

Electrodermography (EDG) / Galvanic Skin Response (GSR) indicates the psychophysiological of emotion, EDG is an emerging signal used in the field of emotion classification aside from Electroencephalography (EEG) and Electrocardiography (ECG). The Empatica E4 wearable device was used in collecting EDG signals and employed as the method in capturing the test subject’s physiological signal of their skin activity. This experiment had 10 participants that use a Virtual Reality (VR) headset for viewing video stimuli in 360 degrees while collecting the EDG signals. Python with Support Vector Machine (SVM) was used in processing the 10 subjects’ data. This paper aims to compare the accuracy of the SVM experiments with different parameters, different settings based on the data retrieved from the wearable. The emotions were classified into four distinct quadrants with inter-subject classifications yielding an accuracy of 54.3%, and intra-subject classification yielded an accuracy of 57.1% to 99.2%. The presented results show that it is possible to achieve results with higher accuracy when parameter tuning. Hence, promising results were demonstrated for emotion prediction in four quadrants using wearable EDG technology in virtual reality environments. This paper provides two contributions, the use of EDG signals in emotion prediction, and the parameter setting to increase the accuracy for SVM classification.

The use of Electrodermography (EDG) in emotion classification is emerging in recent studies, however, it is still limited when compared to the use of other physiological signals such as Electroencephalography (EEG) and Electrocardiography (ECG). Galvanic Skin Response (GSR) or EDG can be used in studies relating to the psychophysiological of emotion. This paper presents the result of an experiment conducted using EDG as the main signal for emotion classification with the use of K-Nearest Neighbor (KNN) as the classifier. In the experiment, the EDG data is acquired from 10 subjects while Virtual Reality (VR) headset is used to view 360 degrees video. Python is used as the programming language for the emotion classification with KNN as the classifier to classify intra-subject (individual) and inter-subject (overall) data. The main objective of this paper is to present the result of the experiment when using KNN as the classifier rather than using Support Vector Machine (SVM) which is synonymous with machine learning. The data were then classified into four classes of distinct emotion, inter-subject achieved an accuracy of 54%, while intra-subject classifications, two subjects achieved an accuracy of 96.9%. This result shows that KNN can provide good accuracy for emotion classification using machine learning as an alternative to SVM.

As an efficient and clean energy in China's energy structure, the proportion of natural gas in the energy structure is increasing. Because the data of gas boilers are very complex, there is no obvious relationship, and the data changes with time, so there is some difficulty in detecting data anomalies. Therefore, in this paper, isolated forest algorithm, One-Class SVM algorithm and LOF algorithm are used to build a model to detect the outlier of multidimensional time series and capture the abnormal fluctuations in the data. Through model training on the data set of gas group, it is found that the isolated forest algorithm, one-class algorithm and LOF algorithm all have good function of distinguishing outliers, but in this data set, the isolated forest algorithm has better effect, and the identification of outliers conforms to the theory of energy conservation model.

In order to effectively detect the fatigue status of equipment operators, this paper proposes a fatigue detection model that integrates multi-source physiological signals to evaluate the fatigue level of equipment operators in order to improve the efficiency of military operations and training. This method selects various physiological signals such as human heart rate, heart rate variability, body temperature, respiratory rate, blood oxygen concentration, etc. according to the working environment and working characteristics of the equipment operator, and establishes the relationship between the characteristic values of various physiological parameters and fatigue levels, And then use the optimized support vector machine to perform fusion judgment on the feature values of multi-source physiological parameters, and obtain the fatigue state of the equipment operator. Experimental results show that this method can make a more accurate judgment on the fatigue state of equipment operators, and is of great significance for ensuring the efficiency of equipment operators’ work and training.

Face recognition is one of the most active areas of research from the past two decades. Attempts are being made to understand how a human recognizes another human face. It is widely accepted that facial recognition can be based on structural information and nonstructural / spatial details. In the present study, he is applying differential observations using Eigen / docking characteristics of many built-in facial features and artificial neural networks. The proposed method aims to obtain a facial feature by reducing facial features such as eyes, nose, mouth, and face depending on the importance of facial features. The face recognition system developed in this paper will inform the human face and assess the current percentage of accuracy. Therefore, this work is for human facial recognition and includes a percentage of facial expressions. The implementation of this function also offers many applications such as photography, bio-metric in bank Lockers, etc.

In the production process of electronic components such as capacitors, bubble defects will occur inside the devices, which will lead to the loss of the functions of electronic components. Traditional methods for detecting bubble defects have many problems, such as low efficiency and low accuracy. Aiming at the bubble defects of the existing electronic components, an SVM-based detection method was proposed, which combined machine learning and image processing to complete the automatic detection of the bubble defects of electronic components. By using the feedback method, the error results obtained from the model classification are added to the training sample set, so that the problems of insufficient sample quantity and incorrect sample classification are solved. The accuracy of the proposed method was verified by capacitor component samples. The accuracy rate was 98.95%, the emission rate was 0.12%, and the error rate was 0.93%. It can be seen that the method proposed in this paper has a good effect on the defect detection of electronic components with the capacitor as an example.

In this paper, we propose a feature coding method based on shared weights support vector data description (FCM-SWSVDD). The proposed process of FCM-SWSVDD is as follows. By considering the density information of the clusters and introducing the weighting learning, we propose an improved support vector data description (SVDD), named shared weights support vector data description (SWSVDD). SWSVDD can obtain the cluster center and cluster radius more accurately. Incorporating SWSVDD and the triangle coding into the same feature coding learning process, FCM-SWSVDD is proposed. After the features of those images are extracted by using FCM-SWSVDD, a sparse representation classifier is used to classify those features. Experimental results show that the performance of the proposed method exceeds many methods.

In adaptive radiotherapy planning, the contour map of the treatment target area is still the most difficult problem. Usually this work is done by a professional radiologist oncologist, which is very time-consuming and labor-intensive. In order to solve this problem, this paper proposes an automatic contour method based on support vector machine. Experiments were conducted on the lower abdomen MR data set of eight patients. In the experiment, the simulated data was only used to train the classifier, and the treatment day images were used to evaluate the performance of the classifier. DSI was used to compare the manual contour and automatic contour of the kidney. The experiment showed that, The automatic contour method based on support vector machine has better classification performance than most classification algorithms. Among the eight sets of results, the DSI value of the six sets of results is 1, and the smallest DSI value is also greater than 9.423.

Human faces contain rich information. Recent studies found that facial features have relation with human weight or body mass index (BMI). Decoding “facial information” from the face in predicting the BMI could be linked to the various health marker. This paper proposed the classification of body mass index (BMI) based on appearance based features of facial images using empirical mode decomposition (EMD) as feature extraction technique. The facial images that describe the body mass index was extracted using EMD to obtain a set of significant features. In this framework, the facial image was decomposed using EMD to produce a small set of intrinsic mode functions (IMF) via sifting process. The IMF features which exhibit the unique pattern were used to classify the BMI. The obtained features were then fed into machine learning classifier such as k-nearest neighbour and support vector machines (SVM) to classify the three BMI classes namely normal, overweight and obese. The obtained results show that the IMF2 feature using SVM classifier achieved recognition rate of 99.12% which show promising result.

Real-time emotion detection and recognition system (REDRS) has been rapidly grown in the current era, especially in human-computer interaction and artificial intelligence field. The mutual interaction of human-computer like service-based sectors including online classrooms, e-business, banking services, robotic automation, and others to utilize the growing technology for the base analysis i.e. tracing the human face recognition for the state of the receiver to adapt the strategy that useful for growing sectors. However, the robust emotional identification of the facial emotions from the images and videos residues a big challenging task because of the accuracy of the emotional features. These features can sometimes represented in several forms, like point-based geometric, static, dynamic, or region-based appearance. Changes of facial features like feature position and shape movements are usually affected by the changes of facial elements and muscles during the expression of emotion. Trying to predict an individual’s spirit during a spoken form, usually requires decoding his/her face. Repeatedly, body language and particularly facial expressions, speak quiet words about the state of mind of a person. The proposed paper focuses on experimenting with a person’s countenance and classifying the mood of the person.

Meridians and acupoints are the basis of TCM theory and play an important role in disease diagnosis and acupuncture treatment. There are still many problems in current research on electrical signals of acupoints. On the one hand, most of the studies did not consider the integrity of the meridian, but only based on a few acupoints. On the other hand, the lack of targeted feature extraction and classification methods leads to unsatisfactory classification results. Considering the above problems, a method combining traditional features and wavelet features is proposed to classify acupoints and non-acupoints. Based on the integrity of the meridians, we first collect the body surface electrical signals of some acupoints and non-acupoints on the twelve meridians of the human body, and then extract traditional and wavelet features from the measured signals. Finally, SVM and XGBoost are used to classify acupoints and non-acupoints respectively. The experimental results show that this method can effectively improve the classification performance of acupoints and non-acupoints, and for the feature vectors constructed in this paper, XGBoost has better classification capabilities.

DNA Microarray technology forms an integral part of the bioinformatics world and provides researchers with ability to monitor a large amount of gene expressions simultaneously. The analysis of this data proves extremely beneficial in the detection of several diseases. Classification of the gene expression data obtained from microarrays is an imperative step in providing further information on the types of possible diseases present. Several classification models for microarray data have been proposed, yielding considerable results in terms of accuracy and execution time. This paper demonstrates the implementation of a microarray data classification system using Elastic Net for feature selection on two classification mechanisms. The classification of the data has been carried out using SVM and Random Forest Classifier methods. The implemented system using Elastic Net and SVM exhibits a much better performance in terms of accuracy and execution time than most of the existing systems.

Ground penetrating radar (GPR) is one of the promising non-destructive imaging tools investigations for shallow subsurface exploration such as locating and mapping the buried utilities. In practical applications, GPR images could be noisy due to the system noise, the heterogeneity of the medium, and mutual wave interactions thus, it is a complex task to recognizing the hyperbolic signature of buried objects from GPR images. Therefore, this paper aims to develop nonlinear feature extraction technique of using Empirical Mode Decomposition (EMD) in recognizing the four geometrical shapes (cubic, cylindrical, disc and spherical) from GPR images. A pre-processing step of isolating hyperbolic signature from different background was first employed by mean of Region of Interest (ROI). The hyperbolic signature that describes the shapes was extracted using EMD decomposition to obtain a set of significant features. In this framework, the hyperbolic pattern was decomposed of using EMD, to produce a small set of intrinsic mode functions (IMF) via sifting process. The IMF properties of the signature that exhibit the unique pattern was used as potential features to differentiate the geometrical shapes of buried objects. The extracted IMF features were then fed into machine learning classifier namely Support Vector Machines. To evaluate the effectiveness of the proposed method, a set data collection of GPR-images has been acquired. The experimental results show that the recognition rate of using IMF features was achieved 99.12% accuracy in recognizing the shapes of buried objects whose shows the promising result.