Train SVM Research Articles

Machine learning ellipsometry as a sensitive diagnostic tool to study reproductive biology in Zika virus infected murine models

Machine-learning spectroscopy is a recent scientific and fast-growing area where specific machine-learning solutions are used to process spectroscopic data beyond traditional physical modeling. Here, several machine learning models are optimized and trained for processing highly sensitive broadband variable angle spectroscopic ellipsometry measurements as a new tool in reproductive biology. This new method is applied to classify semen samples obtained from male offspring from ZIKV-infected and uninfected murine models. The method is label-free, cost-effective, simple, and reaches excellent performance. The combination of spectroscopic depolarization degree data with a well-optimized and trained SVM model resulted in the excellent performance of AUROC = 0.99, accuracy = 92%, specificity = 87% and sensibility = 97%. Optical modeling of ellipsometry spectra with Kramers–Kronigconsistent B-splines that takes care of incoherent reflections in the samples allowed for the extraction of the average semen sample refraction index for each class, revealing small differences that were attributed to the observed semen head defects, protamination failure, and DNA fragmentation.

Developing an ensemble machine learning study: Insights from a multi-center proof-of-concept study.

To address the numerous unmeet clinical needs, in recent years several Machine Learning models applied to medical images and clinical data have been introduced and developed. Even when they achieve encouraging results, they lack evolutionary progression, thus perpetuating their status as autonomous entities. We postulated that different algorithms which have been proposed in the literature to address the same diagnostic task, can be aggregated to enhance classification performance. We suggested a proof of concept to define an ensemble approach useful for integrating different algorithms proposed to solve the same clinical task. The proposed approach was developed starting from a public database consisting of radiomic features extracted from CT images relating to 535 patients suffering from lung cancer. Seven algorithms were trained independently by participants in the AI4MP working group on Artificial Intelligence of the Italian Association of Physics in Medicine to discriminate metastatic from non-metastatic patients. The classification scores generated by these algorithms are used to train SVM classifier. The Explainable Artificial Intelligence approach is applied to the final model. The ensemble model was validated following an 80-20 hold-out and leave-one-out scheme on the training set. Compared to individual algorithms, a more accurate result was achieved. On the independent test the ensemble model achieved an accuracy of 0.78, a F1-score of 0.57 and a log-loss of 0.49. Shapley values representing the contribution of each algorithm to the final classification result of the ensemble model were calculated. This information represents an added value for the end user useful for evaluating the appropriateness of the classification result on a particular case. It also allows us to evaluate on a global level which methodological approaches of the individual algorithms are likely to have the most impact. Our proposal represents an innovative approach useful for integrating different algorithms that populate the literature and which lays the foundations for future evaluations in broader application scenarios.

A new method of image recognition based on deep learning generates adversarial networks and integrates traditional algorithms

This paper discusses an innovative image recognition method, which combines the advanced feature learning ability of generative adversarial networks (Gan) with the robustness of traditional image recognition algorithms. Based on small-sample training, Gan can generate high-quality image samples to expand the training set. Therefore, by designing and training a generative adversarial network, the real image data and its labels can be integrated into the training set. The GAN-generated images, together with corresponding labels, are input into SVM for training, and appropriate kernel functions and parameters are selected to optimize the SVM model to maximize classification performance. GAN is good at data generation and feature learning, while SVM is good at problems with clear classification boundaries, and this model combining the two methods is used in image recognition.

SMOTE-kTLNN: A hybrid re-sampling method based on SMOTE and a two-layer nearest neighbor classifier

In recent years, class-imbalanced learning has become an important branch of machine learning. Synthetic Minority Oversampling Technique (SMOTE) is known as a benchmark method to address imbalanced learning. Although SMOTE performs well on many data, it also has the drawback of generating noisy samples. There are many SMOTE variants to solve this problem. Specifically, these methods are hybrid sampling methods, that is, carrying out an undersampling stage after SMOTE to remove noisy samples. It requires a method that can accurately identify noise to provide reliable performance. In this paper, a hybrid re-sampling method based on SMOTE and a two-layer nearest neighbor classifier (SMOTE-kTLNN) is proposed. SMOTE-kTLNN recognition noise is realized by an Iterative-Partitioning Filter (IPF). Specifically, SMOTE is performed on the original data to balance the data, then the data is divided into n equal parts, establishing kTLNN on each part to predict the whole data. And noisy samples are removed according to the majority voting rule. In the last, the balanced data sets are used to train kNN, AdaBoost, and SVM to verify whether SMOTE-kTLNN is irrelevant to the classifier. The experiment results demonstrate that SMOTE-kTLNN performs better than the comparisons in 25 binary data sets, including Recall, AUC, F1-measure, and G-mean.

Fault location in hybrid transmission lines on the basis of VMD energy entropy and SVM

Finding the head of a traveling wave is challenging because an overhead line-cable hybrid transmission line’s wave impedance is not constant. Because of this, it is difficult to measure the hybrid transmission line precisely. This leads to the proposal of a hybrid transmission line fault location algorithm, which combines VMD energy entropy for training support vector machines and traveling wave law graphs. The VMD energy entropy is first used to train SVM to identify the faulty section. The method of traveling waves with a single end is utilized to precisely fix the fault distance within the faulty segment when the type of traveling wave has been identified. The simulation results demonstrate that the method provides precise measurements and is effective for hybrid transmission lines with a range of fault distances and grounding resistances.

Design a Hybrid Approach for the Classification and Recognition of Traffic Signs Using Machine Learning

The automatic system for classifying traffic signs is a critical task of Advanced Driver Assistance Systems (ADAS) and a fundamental technique utilized as an integral part of the various vehicles. The recognizable features of a traffic image are utilized for their classification. Traffic signs are designed to contain specific shapes and colours, with some text and some symbols with high contrast to the background. This paper proposes a hybrid approach for classifying traffic signs by SIFT for image feature extraction and SVM for training and classification. The proposed work is divided into phases: pre-processing, Feature Extraction, Training, and Classification. MATLAB is used for the implementation purpose of the proposed framework, and classification is carried out by utilizing real traffic sign images

A Hierarchical Accident Recognition Method for Highway Traffic Systems

With the rapid development of highway traffic systems, the accident rate of highway remains high. Once an accident occurs, it will cause serious casualties. In this paper, we proposed a highway accident identification method for real-time accident detection through high-speed monitoring. The proposed method focuses on high detecting speed and high recall for car accident. The designed system consists of four parts: first, yolov3 is used to detect the car. Second, identified car frames are put the identified car frame into the Mosse Tracker for tracking. Third, speed and future center point are estimated according to its size and pixel movement. If the speed exceeds a certain threshold or the future center points among other trackers are too close, the object frame will be put into a trained SVM model. Finally, the flow vector estimation calculation will be used to determine whether there is an accident. The system has fast detection speed, high recall and strong real-time performance. Through this system, we can send warning information to upstream car drivers and traffic management personnel to improve the efficiency of accident handling and reduce the rate of highway traffic accidents and casualties.

Mutual clustered redundancy assisted feature selection for an intrusion detection system

Intrusion Detection is very important in computer networks because the widespread of internet makes the computers more prone to several cyber-attacks. With this inspiration, a new paradigm called Intrusion Detection System (IDS) has emerged and attained a huge research interest. However, the major challenge in IDS is the presence of redundant and duplicate information that causes a serious computational problem in network traffic classifications. To solve this problem, in this paper, we propose a novel IDS model based on statistical processing techniques and machine learning algorithms. The machine learning algorithms incudes Fuzzy C-means and Support Vector Machine while the statistical processing techniques includes correlation and Joint Entropy. The main purpose of FCM is to cluster the train data and SVM is to classify the traffic connections. Next, the main purpose of correlation is to discover and remove the duplicate connections from every cluster while the Joint entropy is applied for the discovery and removal of duplicate features from every connection. For experimental validation, totally three standard datasets namely KDD Cup 99, NSL-KDD and Kyoto2006+ are considered and the performance is measured through Detection Rate, Precision, F-Score, and accuracy. A five-fold cross validation is done on every dataset by changing the traffic and the obtained average performance is compared with existing methods.

PLASTIC: Pressure and location-based acoustic sensing touch interface using classification

A method is described for developing a pressure sensor panel that uses a single small electromechanical exciter, delivering a sawtooth wave, and a single sensor attached to a panel or structure. A trained classifier is able to recognize the pressure and location of a finger-sized object pressing on the panel based on the characteristics of the waveform detected at the sensor. The classification results of three experiments are reported: 55 virtual buttons on a panel, the pressure classification sensitivity of a single virtual button, and the location classification resolution of a tightly packed grid of points. All three experiments show excellent accuracy (99.7%, 85.2%, 96.0%, respectively) using a simple trained linear SVM classifier with a 70/30 training/validation data ratio, demonstrating that the PLASTIC method is an effective method for classifying the location and pressure of a touch on a panel or structure.

Incipient Interturn Fault Detection and Severity Evaluation in Electric Drive System Using Hybrid HCNN-SVM Based Model

The importance of induction motor (IM) drives in industrial applications, especially in the renewable energy sector is unarguable due to their fast-dynamic response. Their monitoring is key to avoid downtime and economic losses. Even though a lot of research has been focused on incipient fault diagnosis in induction machines, early diagnosis in power electronic drive-fed machines is still a challenge. In this article, a novel two-level hybrid hierarchical convolution neural network with support vector machine (HCNN-SVM) is proposed for incipient interturn fault diagnosis of drive-fed machines. The first level of the proposed structure is intended to recognize the pattern of interturn fault in drive-fed IM, while the second level is developed to identify the fault severity. First, the effective features are extracted automatically using the shared layers of HCNN for fault diagnosis and fault severity evaluation at the same time. The features obtained using the HCNN are suitably used to train SVM for classification. Comparison of the results with the existing architectures, namely, HCNN and SVM, show the effectiveness of proposed hybrid method. The experimental results show that the hybrid HCNN-SVM is fast and highly accurate in identifying the interturn fault pattern and evaluating its severity.

Detecção automática de áreas urbanas desfavorecidas usando imagens do Google Earth™ de cidades do semiárido brasileiro

Abstract Automatic classification of deprived urban areas provides vital information for implementing pro-poor policies. In this paper, an approach for the classification of these areas in Brazilian cities is presented. Satellite images were obtained free of charge from six cities in the Brazilian semi-arid region using Google Earth Engine software. To assess the discriminative power of census data, data made publicly available by Brazilian Institute of Geography and Statistics (IBGE) were used to train SVM classifiers together with features extracted from images. The image features were extracted using the following approaches: color histograms, LBP histograms, and lacunarity. Four evaluation tests were investigated based on two criteria: use of census data and cross-validation method. Two types of cross-validation were used: 10-fold and leave-one-city-out. The use of census data caused a negative impact on the results. This impact is justified by the criteria on which census tracts are mapped in the country, not only morphological and visually perceptible through satellite images, as opposed to adopted extraction approaches. The best results obtained were average accuracy of 91.81% and average F1-score of 92.27%. This research contributes to the recognition of deprived urban areas and urban socio-spatial dynamics, supporting urban-territorial planning.

Nondestructive Detection of Ceramic Products Based on Tapping Sound Signal Feature Recogition

Recent studies on the test of ceramic non-destructive testing are mainly based on high cost technologies, image processing and so on, these method possesses some drawback of low efficiency, high cost and so on. What’s more, detecting whether the ceramic products by human through listening to sound of tapping is also effectless. This paper proposed a non-destructive method for ceramic products to solve this problem. This non-destructive method consists of a tapping device and a signal processing module. The tapping device will be applied to generate the tapping sound signal and the signal processing system will be applied to analysis signal. After the process of signal analysis, sample length and peak of spectrum 2 parameters is extracted, then use these parameters to train SVM, the results will be compared with BP neural network (BPNN). The result of experiment shows that SVM with different kernels of linear, poly, rbf, sigmoid respectively reach the accuracy of 96.29%, 96.29%, 46.29%, 93.82%, while BPNN reaches the accuracy of 93.21%. This result proves that SVM can effectively complete the task of identifying defective ceramics, and its performance is better than BPNN.

Application of Frequency Features of Optical Flow for Event Detection in Video-EEG Monitoring Data

The work is devoted to the study of the frequency features of the optical flow obtained from the video record of long-term video-electroencephalographic (video-EEG) monitoring data of patients with epilepsy. It is necessary to obtain features to recognize epileptic seizures and differentiate them from non-epileptic events. We propose to analyze the periodograms of the smoothed optical flow computed from the fragments of the patient’s video recordings. We use Welch's method to obtain periodograms. The values of the power spectral density of the optical flow at the selected frequencies are used as features. Using the clustering algorithm, seven groups of events are identified in video recordings and combined into three generalized classes. We train SVM classifier and conduct recognition of events in a test sample of 103 video fragments in four patients. The experiment indicates the accuracy of event classification equal to 90.3%.

Facial Expression Recognition Algorithm Based on Fusion of Transformed Multilevel Features and Improved Weighted Voting SVM

In allusion to the shortcomings of traditional facial expression recognition (FER) that only uses a single feature and the recognition rate is not high, a FER method based on fusion of transformed multilevel features and improved weighted voting SVM (FTMS) is proposed. The algorithm combines the transformed traditional shallow features and convolutional neural network (CNN) deep semantic features and uses an improved weighted voting method to make a comprehensive decision on the results of the four trained SVM classifiers to obtain the final recognition result. The shallow features include local Gabor features, LBP features, and joint geometric features designed in this study, which are composed of distance and deformation characteristics. The deep feature of CNN is the multilayer feature fusion of CNN proposed in this study. This study also proposes to use a better performance SVM classifier with CNN to replace Softmax since the poor distinction between facial expressions. Experiments on the FERPlus database show that the recognition rate of this method is 17.2% higher than that of the traditional CNN, which proves the effectiveness of the fusion of the multilayer convolutional layer features and SVM. FTMS-based facial expression recognition experiments are carried out on the JAFFE and CK+ datasets. Experimental results show that, compared with the single feature, the proposed algorithm has higher recognition rate and robustness and makes full use of the advantages and characteristics of different features.

On-air English Capital Alphabet (ECA) recognition using depth information

On-air writing can be considered as a time-dependent event where hand gesture is produced in a natural environment through index finger movement. A sequence of such movements containing several time steps in 3D space can be utilized to construct an English Capital Alphabet (ECA). While Previous researches investigated 2D features, we believe that depth information may play a significant role along with other features in recognition of these dynamic gestures. We have captured hand finger motion information using a depth camera and represented them as depth images for each ECA. The hand finger trajectory data were extracted from the depth image, and a combination of depth-based features and non-depth features were generated; depth variation was performed in the depth-based features, and then, all the feature values were converted into time-series data. Dynamic Time Warping distances were determined between a template ECA and a test ECA for each ECA collected from 15 participants. These distance-based features were then fed into a multi-class SVM for training and testing and got the recognition accuracy of $$80.77\%$$ without depth and $$88.21\%$$ with depth-based features. To cope with the over-fitting problem, we applied the resampling technique and got the highest recognition accuracy of 96.85%, and at last, we applied some feature selection techniques to analyze the recognition results.

Emotion recognition by deeply learned multi-channel textual and EEG features

Human emotion recognition is a key technique in human–computer interaction. Traditional emotion recognition algorithms rely on external actions such as facial expression, which may fail to capture real human emotion since facial expression signals may be camouflaged. EEG signal is closely close to human emotion, which can directly reflect human emotion. In this paper, we propose to learn multi-channel features from the EEG signal for human emotion recognition, where the EEG signal is generated by sound signal stimulation. Specifically, we apply multi-channel EEG and textual feature fusion in time-domain to recognize different human emotions, where six statistical features in time-domain are fused to a feature vector for emotion classification. The textual feature extraction is based. And we conduct EEG&textual-based feature extraction from both time and frequency domain. Finally, we train SVM for human emotion recognition. Experimental on DEAP dataset show that compared with frequency domain feature-based emotion recognition algorithms, our proposed method improves recognition accuracy rate.

Combined Deep Learning With Directed Acyclic Graph SVM for Local Adjustment of Age Estimation

In order to further improve the accuracy of age estimation, a locally adjusted age estimation algorithm based on deep learning and directed acyclic graph SVM is proposed. In the training phase, SE-ResNet-50 network pre-trained by the VGGFace2 dataset is first fine-tuned. Once the network converges, and the vector consisting of the parameters of the last fully connected layer is used as a representation and train multiple One-Versus-One SVMs. In the test phase, we first sent the face image to be estimated into SE-ResNet-50 to obtain a rough age estimation value, then set the specific neighborhood, and finally combined the trained SVM into a directed acyclic graph SVM and set specific neighborhood with the global estimate as the center for accurate age estimate. In order to show the universality of the proposed coarse-to-fine or/and global-to-local method, experiments were carried out on MORPH and AFAD images of different races, and the results verified the effectiveness of the algorithm.

Image Normalization and Weighted Classification Using an Efficient Approach for SVM Classifiers

The content of massive image changing the brightest brightness is an impasse between most tests of sorted image realizations with low-resolution representation. I have done this research through image security, which will help curb crime in the coming days, and we propose a novel receipt for their strong and effective counterpart. Image classification using low levels of the image is a difficult method, so for this, I have adopted the method of automating the semantic image classification of this research and used it with different SVM classifiers, based on the normalized weighted feature support vector machine for semantic image classification. This is a novel approach given that weighted feature or normalized biased feature is applied and it is found that the normalized method is the best. It also uses normalized weighted features to compute kernel functions and train SVM. The trained SVM is then used to classify new images. During training and generalization, we displayed a decrease of identification error rate and there have been many benefits of using SVM with better performance in normalized image-cataloging systems. The importance of this technique and its role will be highlighted in the years to come.

Automated shopping system using computer vision

The shopping experiment made by amazon go in USA is one of the most interesting applications of computer vision recently. They allow you to shop and automatically charge your virtual card for whatever goods you purchased using cameras and wireless systems, so no checkouts or waiting lines are required. However, amazon didn’t reveal yet the details of how their system components are implemented. In this paper, we introduce a complete system for computer vision based automated shopping. The proposed system contains barcode scanning of objects, data registration, image capturing for offline training stage, motion (change) detection, CNN and SVM for object classification and charging/discharging customers. Our system can be integrated with the wireless data transmission to do the whole shopping process. First, the proposed method extracts the objects’ barcodes to register their details, and take sample images of objects for classifier training. We employ a pre-trained CNN (i.e. ResNet50) for feature extraction and a multi-class SVM for training. After training our classifier, we have a real-time operation stage (i.e. test stage). We assume that a camera is embedded above products on each shelf to capture videos of the products. We employ a change detector to understand any added or removed items. If the item is removed from or added to the shelve, the moving object is input to CNN feature extractor, and then SVM classifier for identification and pricing. Results show that the proposed system is fast and effective.

A Novel Bearing Fault Diagnosis Method Based on GL-mRMR-SVM

A convolutional neural network (CNN) has been used to successfully realize end-to-end bearing fault diagnosis due to its powerful feature extraction ability. However, the CNN is prone to focus on local information, ignoring the relationship between the whole and the part of the signal due to its unique structure. In addition, it extracts some fault features with poor robustness under noisy environment. A novel diagnosis model based on feature fusion and feature selection, GL-mRMR-SVM, is proposed to address this problem in this paper. First, the model combines the global features in the time-domain and frequency-domain of the raw data with the local features extracted by CNN to make full use of the signal information and overcome the weakness of traditional CNNs neglecting the overall signal. Then, the max-relevance min-redundancy (mRMR) algorithm is used to automatically extract the discriminative features from the fused features without any prior knowledge. Finally, the extracted discriminative features are input into the SVM for training and output the fault recognition results. The proposed GL-mRMR-SVM model was evaluated through experiments on bearing data of Case Western Reserve University (CWRU) and CUT-2 platform. The experimental results show that the proposed method is more effective than other intelligent diagnosis methods.