Accurate Classification Results Research Articles

Atom-substituted tensor dictionary learning enhanced convolutional neural network for hyperspectral image classification

A novel sparse tensor dictionary learning algorithm and a convolutional neural network (CNN) classification method based on this algorithm are proposed for hyperspectral image (HSI) in this paper. The HSI is intuitively represented as a three-dimensional (3-D) cube, and utilizing the joint spatial-spectral information can significantly improve the accuracy of HSI classification. Therefore, the proposed atom-substituted tensor dictionary learning (ASTDL) algorithm utilizes tensor techniques to extract 3-D joint spatial-spectral features from HSI cubes directly. Sparsity constraint is enforced on the coefficient tensors, which obeys the sparsity attribute of HSI. The proposed ASTDL enhanced CNN (ASTDL-CNN) classification method utilizes a two-dimensional (2-D) CNN to extract deep features from the feature tensors obtained by the ASTDL, and to perform the pixel-wise classification. The use of ASTDL, which extracts intrinsic tensor features before CNN, alleviates within-class spatial-spectral variation and reduces the requirement of CNN for the labeled data. Whereas, the CNN works as a 3-D classifier, and provides the final classification results. Besides, we perform the majority vote on the classification map obtained by CNN to refine the classification. The performance of the proposed method is evaluated on three real HSI data sets. The competitive results to the compared state-of-the-art methods demonstrate that the ASTDL-CNN can provide accurate and robust classification results.

Open-Set Recognition with Gaussian Mixture Variational Autoencoders

In inference, open-set classification is to either classify a sample into a known class from training or reject it as an unknown class. Existing deep open-set classifiers train explicit closed-set classifiers, in some cases disjointly utilizing reconstruction, which we find dilutes the latent representation's ability to distinguish unknown classes. In contrast, we train our model to cooperatively learn reconstruction and perform class-based clustering in the latent space. With this, our Gaussian mixture variational autoencoder (GMVAE) achieves more accurate and robust open-set classification results, with an average F1 increase of 0.26, through extensive experiments aided by analytical results.

Study of Image Classification Accuracy with Fourier Ptychography

In this research, the accuracy of image classification with Fourier Ptychography Microscopy (FPM) has been systematically investigated. Multiple linear regression shows a strong linear relationship between the results of image classification accuracy and image visual appearance quality based on PSNR and SSIM with multiple training datasets including MINST, Fashion MNIST, Cifar, Caltech 101, and customized training datasets. It is, therefore, feasible to predict the image classification accuracy only based on PSNR and SSIM. It is also found that the image classification accuracy of FPM reconstructed with higher resolution images is significantly different from using the lower resolution images under the lower numerical aperture (NA) condition. The difference is yet less pronounced under the higher NA condition.

Retinal disease identification using upgraded CLAHE filter and transfer convolution neural network

Retinal tissue plays a crucial part in human vision. Infections of retinal tissue and delayed treatment or untreated infection could lead to loss of vision. Additionally, the diagnosis is prone to errors when huge dataset is involved. Therefore, a fully automated model of identification of retinal disease is proposed to reduce human interaction while retaining its high accuracy classification results. This paper introduces an enhanced design of a fully automatic multi-class retina diseases prediction system to assist ophthalmologists in making speedy and accurate investigation. Retinal fundus images, which have been used in this study, were downloaded from the stare website (157 images from five classes: BDR, CRVO, CNV, PDR, and Normal). The five files were categorized according to their annotations conducted by the experienced specialists. The categorized images were first processed with the proposed upgraded contrast-limited adaptive histogram filter for image brightness enhancement, noise reduction, and intensity spectrum normalization. The proposed model was designed with transfer learning method and the fine-tuned pre-trained RESNET50. Eventually, the proposed framework was examined with performance evaluation parameters, recorded a classification rate with 100% sensitivity, 100% specificity, and 100% accuracy. The performance of the proposed model showed a magnificent superiority as compared to the state-of-the-art studies.

Power System Event Identification Based on Deep Neural Network With Information Loading

Online power system event identification and classification are crucial to enhancing the reliability of transmission systems. In this paper, we develop a deep neural network (DNN) based approach to identify and classify power system events by leveraging real-world measurements from hundreds of phasor measurement units (PMUs) and labels from thousands of events. Two innovative designs are embedded into the baseline model built on convolutional neural networks (CNNs) to improve the event classification accuracy. First, we propose a graph signal processing based PMU sorting algorithm to improve the learning efficiency of CNNs. Second, we deploy information loading based regularization to strike the right balance between memorization and generalization for the DNN. Numerical results based on real-world dataset from the Eastern Interconnection of the U.S power transmission grid show that the combination of PMU based sorting and the information loading based regularization techniques help the proposed DNN approach achieve highly accurate event identification and classification results.

Automatic classification of nerve discharge rhythms based on sparse auto-encoder and time series feature

BackgroundNerve discharge is the carrier of information transmission, which can reveal the basic rules of various nerve activities. Recognition of the nerve discharge rhythm is the key to correctly understand the dynamic behavior of the nervous system. The previous methods for the nerve discharge recognition almost depended on the traditional statistical features, and the nonlinear dynamical features of the discharge activity. The artificial extraction and the empirical judgment of the features were required for the recognition. Thus, these methods suffered from subjective factors and were not conducive to the identification of a large number of discharge rhythms.ResultsThe ability of automatic feature extraction along with the development of the neural network has been greatly improved. In this paper, an effective discharge rhythm classification model based on sparse auto-encoder was proposed. The sparse auto-encoder was used to construct the feature learning network. The simulated discharge data from the Chay model and its variants were taken as the input of the network, and the fused features, including the network learning features, covariance and approximate entropy of nerve discharge, were classified by Softmax. The results showed that the accuracy of the classification on the testing data was 87.5%, which could provide more accurate classification results. Compared with other methods for the identification of nerve discharge types, this method could extract the characteristics of nerve discharge rhythm automatically without artificial design, and show a higher accuracy.ConclusionsThe sparse auto-encoder, even neural network has not been used to classify the basic nerve discharge from neither biological experiment data nor model simulation data. The automatic classification method of nerve discharge rhythm based on the sparse auto-encoder in this paper reduced the subjectivity and misjudgment of the artificial feature extraction, saved the time for the comparison with the traditional method, and improved the intelligence of the classification of discharge types. It could further help us to recognize and identify the nerve discharge activities in a new way.

Parameter estimation for high dimensional classification model on colon cancer microarray dataset

In classification problems, logistic regression is among powerful techniques for discrimination. It provides directive probabilities of sample classification and interpretable coefficients. When it comes to model high dimensional dataset, however, logistic regression with its Newton-Raphson method of parameter estimation is no longer applicable, especially on low sample size and extremely high dimension. By applying cross-entropy algorithm on regularized logistic regression, it was able to well performing parameter estimation and highly accurate classification result.

Multi-Level Image Segmentation for Urban Land-Cover Classifications

Variation land-cover features, which include natural and man-made objects, lead to the advent of features that are spectrally similar. Object in urban area tend to have spectral similar response that can easily misclassified from one to another for example in the case of tree and grass as well as asphalt building roof and asphalt road. Object based classification approached instead of pixel based will improved the misclassification yet will increase the accuracy of land-cover classification. Using Worldview-2 multispectral satellite image as a primary data, while normalized Digital Surface Model (nDSM) derived from Light Detection and Ranging (LIDAR) data and indices image, the image segmentation process utilizing multiresolution segmentation algorithm and spectral difference was conducted. Before going through classification process, twelve segmentation levels were constructed to create image objects. Three classification algorithm including Support Vector Machine (SVM), BAYES and K-Nearest Neighbour (KNN) were choose to be tested to identify which algorithm gives the best classification result of the urban area target. The results from the study indicate statistically significant difference in classification accuracy between each algorithm: Based on Kappa statistics, user’s and producer’s accuracy, as well as visual examination and overall accuracy performance, BAYES with overall accuracy of 85.51% has depicted to have the best land-cover classification accuracy result.

Sea Ice Classification of SAR Imagery Based on Convolution Neural Networks

We explore new and existing convolutional neural network (CNN) architectures for sea ice classification using Sentinel-1 (S1) synthetic aperture radar (SAR) data by investigating two key challenges: binary sea ice versus open-water classification, and a multi-class sea ice type classification. The analysis of sea ice in SAR images is challenging because of the thermal noise effects and ambiguities in the radar backscatter for certain conditions that include the reflection of complex information from sea ice surfaces. We use manually annotated SAR images containing various sea ice types to construct a dataset for our Deep Learning (DL) analysis. To avoid contamination between classes we use a combination of near-simultaneous SAR images from S1 and fine resolution cloud-free optical data from Sentinel-2 (S2). For the classification, we use data augmentation to adjust for the imbalance of sea ice type classes in the training data. The SAR images are divided into small patches which are processed one at a time. We demonstrate that the combination of data augmentation and training of a proposed modified Visual Geometric Group 16-layer (VGG-16) network, trained from scratch, significantly improves the classification performance, compared to the original VGG-16 model and an ad hoc CNN model. The experimental results show both qualitatively and quantitatively that our models produce accurate classification results.

Deep Learning Strategy to Recognize Kannada Named Entities

Entity representatives are useful in understanding the natural language tasks including the semantics of the Kannada sentences into various entities. In this paper, we have come up with new pertained tag based representative learning of words and entities based on the bidirectional parsing. The proposed research works on segmenting the sentences of Kannada words into various taken, where every token makes various contributions in understanding the semantics of Kannada Sentences which treats words and entities in a given text as independent tokens, and outputs tagged entities based on representative learning mechanism. The research also has focused its attention towards achieving the results of good classification accuracy while recognizing the entities are through the tagging mechanism that is an extension of the general self-tagging mechanism of the Supervised Machine Learning Technique, and considers the types of tokens (words or entities) when computing attention scores. The erected research work has given its significant contribution in terms of good results over a standard benchmark datasets. In particular, it obtains state-of-the-art results on five well-known datasets: Open Entity (entity typing), TACRED (relation classification), CoNLL-2003 (named entity recognition), ReCoRD (cloze-style question answering), and SQuAD 1.1 (extractive question answering) as well as Kannada Named Entity Recognition of Central Institute of Indian Languages.

Prediction of Diabetes in Females of Pima Indian Heritage: A Complete Supervised Learning Approach

Nowadays, diabetes is a common disease that affects millions of people over the world, and women are mostly affected by this disease. Recent healthcare studies have applied various innovative and advanced technologies to diagnose people and predict their disease based on clinical data. One of such technologies is machine learning (ML) in which diagnosis and prediction can be made more accurately. In this paper, the designed model predicts the diabetes of females of Pima Indians heritage by taking the clinical dataset. Here, this problem is considered as a binary classification problem. Therefore, supervised learning algorithms have been used, such as classification tree (CT), support vector machine (SVM), k-Nearest Neighbour (k-NN), Naive Bayes (NB), Random Forest (RF), Neural Network (NN), AdaBoost (AB) and Logistic Regression (LR). We use the female Pima Indians diabetic dataset from Kaggle and UCI data repository and k-fold cross-validation to carry out the process of training and testing. We determine the area under the curve (AUC), classification accuracy (CA), F1, precision and recall results of all the supervised learning algorithms and compare them to determine the best algorithm that is suitable for prediction. For this, we use the Orange 3.24.1 open-source platform to generate the results, which uses Python open-source libraries. From the results, it is concluded that the LR performs better in comparison to other algorithms

Connected Components-based Colour Image Representations of Vibrations for a Two-stage Fault Diagnosis of Roller Bearings Using Convolutional Neural Networks

Roller bearing failure is one of the most common faults in rotating machines. Various techniques for bearing fault diagnosis based on faults feature extraction have been proposed. But feature extraction from fault signals requires expert prior information and human labour. Recently, deep learning algorithms have been applied extensively in the condition monitoring of rotating machines to learn features automatically from the input data. Given its robust performance in image recognition, the convolutional neural network (CNN) architecture has been widely used to learn automatically discriminative features from vibration images and classify health conditions. This paper proposes and evaluates a two-stage method RGBVI-CNN for roller bearings fault diagnosis. The first stage in the proposed method is to generate the RGB vibration images (RGBVIs) from the input vibration signals. To begin this process, first, the 1-D vibration signals were converted to 2-D grayscale vibration Images. Once the conversion was completed, the regions of interest (ROI) were found in the converted 2-D grayscale vibration images. Finally, to produce vibration images with more discriminative characteristics, an algorithm was applied to the 2-D grayscale vibration images to produce connected components-based RGB vibration images (RGBVIs) with sets of colours and texture features. In the second stage, with these RGBVIs a CNN-based architecture was employed to learn automatically features from the RGBVIs and to classify bearing health conditions. Two cases of fault classification of rolling element bearings are used to validate the proposed method. Experimental results of this investigation demonstrate that RGBVI-CNN can generate advantageous health condition features from bearing vibration signals and classify the health conditions under different working loads with high accuracy. Moreover, several classification models trained using RGBVI-CNN offered high performance in the testing results of the overall classification accuracy, precision, recall, and F-score.

Cervical Cancer Diagnosis System Using Ant-Miner for Managing the Knowledge in Medical Database

The fourth most frequent cause of cancer death in women is cervical cancer. No sign can be observed in the early stages of the disease. In addition, cervical cancer diagnosis methods used in health centers are time-consuming and costly. Data classification has been widely applied in the diagnosis of cervical cancer for knowledge acquisition. However, none of the existing intelligent methods are comprehensible, and they look like a black box to clinicians. In this paper, an ant colony optimization-based classification algorithm, Ant-Miner is applied to analyze the cervical cancer data set. The cervical cancer data set used was obtained from the repository of the University of California, Irvine. The proposed algorithm outperforms the previous approach, support vector machine, in the same domain, in terms of the better result of classification accuracy. The proposed method is implemented as an engine in a prototype system named as the cervical cancer detection system. Evaluation of the prototype system demonstrates a good result on its usability and functionality.

Fast identification of the geographical origin of Gastrodia elata using excitation-emission matrix fluorescence and chemometric methods

Geographical origin is an important factor affecting the quality of traditional Chinese medicine. In this paper, the identification of geographical origin of Gastrodia elata was performed by using excitation-emission matrix fluorescence and chemometric methods. Firstly, excitation-emission matrix (EEM) fluorescence spectra of Gastrodia elata samples from different geographical origins were obtained. And then three chemometric methods, including multilinear partial least squares discriminant analysis (N-PLS-DA), unfold partial least squares discriminant analysis (U-PLS-DA), and k-nearest neighbor (kNN) method, were applied to build discriminant models. Finally, 45 Gastrodia elata samples could be differentiated from each other by these classification models according to their geographical origins. The results showed that all models obtained good classification results. Compared with the N-PLS-DA and U-PLS-DA, kNN got more accurate and reliable classification results and could identify Gastrodia elata samples from different geographical origins with 100% accuracy on the training and test set. Therefore, the proposed method was available for easily and quickly distinguishing the geographical origin of Gastrodia elata, which can be considered as a promising alternative method for determining the geographic origin of other traditional Chinese medicines.

Classification of soil quality using K-Nearest Neighbors methods

The damage or deterioration of the material due to reaction with the environment is called corrosion. The corrosive of soil can cause natural disasters such as landslides. Soil corrosivity is a serious problem in industrial and infrastructure activities. Soil corrosivity is also a problem at PT. IPMOMI, because of the large number of metal pipes that are planted underground which functions to drain fluid in the form of seawater as the main material for steam production. PT. IPMOMI wants to install metal pipes underground, Therefore PT. IPMOMI requires information about the status of soil corrosivity at that location. To overcome soil corrosivity, PT. IMOMI will install cathodic protection on the soil with high corrosivity. One of the efforts to deal with this problem is by mapping the subsurface corrosion zone. Before mapping the corrosion zone, it is necessary to know the level of soil corrosivity whether it is very high, high, medium, low, or very low. So that to determine the level of soil corrosivity whether very high, high, medium, low, or very low, a classification method is needed. In this study, the classification method used is the nonparamteric classification method, namely K Nearest Neigbors. The results of classification accuracy were 83.3% and variable chargebility had a higher contribution than the depth level in the classification model.

Intelligent Fault Diagnosis of Rotary Machinery by Convolutional Neural Network with Automatic Hyper-Parameters Tuning Using Bayesian Optimization.

Intelligent fault diagnosis can be related to applications of machine learning theories to machine fault diagnosis. Although there is a large number of successful examples, there is a gap in the optimization of the hyper-parameters of the machine learning model, which ultimately has a major impact on the performance of the model. Machine learning experts are required to configure a set of hyper-parameter values manually. This work presents a convolutional neural network based data-driven intelligent fault diagnosis technique for rotary machinery which uses model with optimized hyper-parameters and network structure. The proposed technique input raw three axes accelerometer signal as high definition 1-D data into deep learning layers with optimized hyper-parameters. Input is consisted of wide 12,800 × 1 × 3 vibration signal matrix. Model learning phase includes Bayesian optimization that optimizes hyper-parameters of the convolutional neural network. Finally, by using a Convolutional Neural Network (CNN) model with optimized hyper-parameters, classification in one of the 8 different machine states and 2 rotational speeds can be performed. This study accomplished the effective classification of different rotary machinery states in different rotational speeds using optimized convolutional artificial neural network for classification of raw three axis accelerometer signal input. Overall classification accuracy of 99.94% on evaluation set is obtained with the CNN model based on 19 layers. Additionally, more data are collected on the same machine with altered bearings to test the model for overfitting. Result of classification accuracy of 100% on second evaluation set has been achieved, proving the potential of using the proposed technique.

Deep learning through LSTM classification and regression for transmission line fault detection, diagnosis and location in large-scale multi-machine power systems

Fault detection, diagnosis, identification and location are crucial to improve the sensitivity and reliability of system protection. This maintains power systems continuous proper operation; however, it is challenging in large-scale multi-machine power systems. This paper introduces three novel Deep Learning (DL) classification and regression models based on Deep Recurrent Neural Networks (DRNN) for Fault Region Identification (FRI), Fault Type Classification (FTC), and Fault Location Prediction (FLP). These novel models explore full transient data from pre- and post-fault cycles to make reliable decisions; whereas current and voltage signals are measured through Phasor Measurement Units (PMUs) at different terminals and used as input features to the DRNN models. Sequential Deep Learning (SDL) is employed herein through Long Short-Term Memory (LSTM) to model spatiotemporal sequences of high-dimensional multivariate features to achieve accurate classification and prediction results. The proposed algorithms were tested in a Two-Area Four-Machine Power System. Training and testing data are collected during transmission lines faults of different types introduced at various locations in different regions. The presented algorithms achieved superior detection, classification and location performance with high accuracy and robustness compared to contemporary techniques.

PENGARUH SELEKSI FITUR CITRA TERHADAP KLASIFIKASI TINGKAT KESEGARAN DAGING SAPI LOKAL

Identifying beef manually has some drawbacks because human visual has limitations and there are differences of human perception in assessing object quality. Several researches developed beef quality assessment methods based on image feature extraction. However, not all features support for obtaining the classification results that have high accuracy. The efficiency will be achieved if the classification analyzes only the relevant features. Therefore, a feature selection process is required to select relevant features and to eliminate irrelevant features to obtain more accurate and faster classification results. One of the feature selection algorithms is the F-Score which is a simple technique that measures the discrimination of two sets of real numbers. The features with the lowest ranking from the F-Score will be eliminated one by one until the most relevant features are obtained. The test is carried out by analyzing the classification results in the form of sensitivity, specificity, and accuracy values. The results of this research showed that by using the F-Score feature, the most relevant features for the classification of freshness level of local beef are obtained using the K-Nearest Neighbor (KNN) method. These features include the average color intensity R and standard deviation with a sensitivity of 0.8, a specificity of 0.93, and an accuracy of 86%. Keywords: Classification, Fiture Selection, F-Score, K-Nearest Neighbor, Local beef

Multi-Language Spam/Phishing Classification by Email Body Text: Toward Automated Security Incident Investigation

Spamming and phishing are two types of emailing that are annoying and unwanted, differing by the potential threat and impact to the user. Automated classification of these categories can increase the users’ awareness as well as to be used for incident investigation prioritization or automated fact gathering. However, currently there are no scientific papers focusing on email classification concerning these two categories of spam and phishing emails. Therefore this paper presents a solution, based on email message body text automated classification into spam and phishing emails. We apply the proposed solution for email classification, written in three languages: English, Russian, and Lithuanian. As most public email datasets almost exclusively collect English emails, we investigate the suitability of automated dataset translation to adapt it to email classification, written in other languages. Experiments on public dataset usage limitations for a specific organization are executed in this paper to evaluate the need of dataset updates for more accurate classification results.

Multimodal Gait Recognition for Neurodegenerative Diseases.

In recent years, single modality-based gait recognition has been extensively explored in the analysis of medical images or other sensory data, and it is recognized that each of the established approaches has different strengths and weaknesses. As an important motor symptom, gait disturbance is usually used for diagnosis and evaluation of diseases; moreover, the use of multimodality analysis of the patient's walking pattern compensates for the one-sidedness of single modality gait recognition methods that only learn gait changes in a single measurement dimension. The fusion of multiple measurement resources has demonstrated promising performance in the identification of gait patterns associated with individual diseases. In this article, as a useful tool, we propose a novel hybrid model to learn the gait differences between three neurodegenerative diseases, between patients with different severity levels of Parkinson's disease, and between healthy individuals and patients, by fusing and aggregating data from multiple sensors. A spatial feature extractor (SFE) is applied to generating representative features of images or signals. In order to capture temporal information from the two modality data, a new correlative memory neural network (CorrMNN) architecture is designed for extracting temporal features. Afterward, we embed a multiswitch discriminator to associate the observations with individual state estimations. Compared with several state-of-the-art techniques, our proposed framework shows more accurate classification results.