Search

The handwritten Urdu text recognition is a challenging area in pattern recognition and has gained much importance after the rapid emergence of several camera-based applications on portable devices, which facilitate the daily processing of plenty of images. The various challenges encountered in handwritten Urdu recognition are writer-dependent variations amongst different Urdu writers, irregular positioning of diacritics associated with a character, context sensitivity of characters, and cursive nature of Urdu script. These challenges also make it difficult to formulate a large generalized handwritten Urdu dataset. The state-of-the-art approaches proposed for the recognition of handwritten Urdu text mostly focus on implicit approaches. These approaches are error prone and do not yield significant recognition rates. The holistic approach of handwritten Urdu recognition has been least explored to date and the existing holistic approaches are complex and time consuming as they mostly rely on convolutional/recurrent neural networks or statistical methods. Hence, in this research, a novel and efficient vision transformer-based methodology using BERT architecture has been proposed to the recognition of handwritten Urdu text. The proposed approach uses convolution feature maps as word embedding in the transformer that makes full use of the powerful attention mechanism of the vision transformer to focus on a particular connected component (ligature) in handwritten Urdu text. To cover the entire Urdu corpus, we have pre-trained several benchmark handwritten Urdu datasets such as UNHD and NUST-UHWR and tested unconstrained handwritten Urdu text. In comparison with the state-of-the-art techniques, the experimental evaluation of the proposed approach reports the better results of the various performance parameters such as Ligature Error Rate (LER), precision, sensitivity, specificity, f1-score, and accuracy. The great success of the proposed approach lies in (i) the significant reduction of training time needed to train a large handwritten Urdu dataset, (ii) minimum computational complexity as there is no overhead of diacritic separation and re-association as used in most of the state-of-the-art techniques, and (iii) the proposed approach registers a new state-of-the-art LER of up to 3% only on unconstrained handwritten Urdu text. [ABSTRACT FROM AUTHOR]

The seizure is an unusual event of the brain, which leads to the second most common disease of the brain called epilepsy. Electroencephalography (EEG) has the potential to provide insight into the diagnosis of seizure. Our objective is to explore the practical efficiency of the convolutional neural network (CNN) in detecting seizures using EEG signal. A novel CNN-FCM architecture is proposed to classify the seizure signals. The conventional CNN is modified with Fuzzy C-means (FCM) clustering algorithm. The competency of the clustering method is confirmed with the cluster validity index (CVI) parameters such as partition entropy (PE), partition coefficient (PC) and the Xie-Beni index (XB). The efficiency of proposed CNN-FCM architecture is validated and confirmed by considering the standard classification parameters such as accuracy, sensitivity, specificity and F-measure. The two different seizure EEG datasets are utilized to examine the proposed system. The proposed CNN-FCM architecture achieved the classification accuracy of 98.33% and outperformed with other existing deep learning methods, achieving a less computational time of 0.3286 s in classification. The performance outcomes exhibit the efficiency of the proposed one-dimensional CNN-FCM in the diagnosis of epilepsy. In contrast to other existing automatic seizure detection techniques, the CNN-FCM architecture can perform real-time seizure detection and classification. [ABSTRACT FROM AUTHOR]

In recent years, how to discover causality rather than correlation among climate variables and how to use causality to help in time-series tasks have received great concern. However, the high dimensionality and nonlinearity of climate variables are the main issues for causal inference based on historical large-scale climate time series. Therefore, a method based on neural Granger causality inference is proposed to study the interactions of climate variables, with focus on the variables commonly used in the energy field, especially in photovoltaics. Firstly, for each climate variable, the time-varying causality and global causality are, respectively, obtained on each time window and on the whole series by neural Granger causality inference. Secondly, the global causality is used as a feature selection map of the input variables in the prediction task. Finally, compared with some existing feature selection methods, the experiments determine that the proposed method not only reveals the appropriate causality rather than the correlation among climate variables, but also efficiently reduces the input dimensionality and improves the performance and interpretability of the predicting model. [ABSTRACT FROM AUTHOR]

Most time series forecasting tasks using Artificial Neural Networks (ANNs) relegate trend-cycle modeling to a simple preprocessing step. In this work, we propose an automatic feature engineering framework for modeling the trend-cycle (tofee-tree) in time series forecasting. The first stage of the framework automatically creates over 286 deterministic linear and nonlinear engineered features to model the trend-cycle. These features are based only on the time of observation and length of the time series, making them domain-agnostic. In the second stage of the framework, a SHapley Additive exPlanations (SHAP)—based feature selection procedure using Light Gradient Boosted Machine (LightGBM) selects the most relevant features. These relevant features can be used for forecasting with ANNs in addition to the auto-regressive lags. Two popular ANNs—Multi-Layer Perceptron (MLP) and Long Short Term Memory network (LSTM) are used to evaluate our proposed tofee-tree framework. Comparisons against two empirical studies using the M3 competition dataset show that the proposed framework improved the overall Symmetric Mean Absolute Percentage Error (SMAPE) in the one-step, medium- and long-term. The relative improvement in one-step SMAPE is 3% for MLP and 23% for LSTM. We also show that the residual seasonality left after deseasonalization can be modeled using the tofee-tree framework. [ABSTRACT FROM AUTHOR]

A new equalization model for digital communication systems is proposed, based on a multi-layer perceptron (MLP) artificial neural network with a backpropagation algorithm. Unlike earlier techniques, the proposed model, called the bidimensional neural equalizer, is composed of two independent MLP networks that operate in parallel for each dimension of the digital modulation scheme. A heuristic method to combine the errors of the two MLP networks is also proposed, with the aim of reducing the convergence time. Simulations performed for linear and nonlinear channels demonstrated that the new model could improve performance in terms of the bit error rate and the convergence time, compared to existing models. [ABSTRACT FROM AUTHOR]

Nowadays, genetic disorders, like cancer and birth defects, are a great threat to human life. Since the first noticing of these types of diseases, many efforts have been made and researches performed in order to recognize them and find a cure for them. These disorders affect genes and they appear as abnormal traits in a genetic organism. In order to recognize abnormal genes, we need to predict splice sites in a DNA signal; then, we can process the genetic codes between two continuous splice sites and analyze the trait that it represents. In addition to abnormal genes and their consequent disorders, we can also identify other normal human traits like physical and mental features. So the primary issue here is to estimate splice sites precisely. In this paper, we have introduced two new methods in using neuro-fuzzy network and clustering for DNA splice site prediction. In this method, instead of using raw data and nucleotide sequence as an input to neural network, a survey on the first bunch of the nucleotide sequence of true and false categories of the input is carried out and training of the neuro-fuzzy network is achieved based on the similarities and dissimilarities of the selected sequences. In addition, sequences of the large input data are clustered into smaller categories to improve the prediction as they are really spliced based on different mechanisms. Experimental results show that these improvements have increased the recognition rate of the splice sites. [ABSTRACT FROM AUTHOR]

In the industrial environment, specifically in the automotive industry, an accurate prediction of execution times for each production task is very useful in order to plan the work and to optimize the human, technical and material resources. In this paper, we applied several regression neural networks to predict the execution times of the tasks in the production of parts for plastic injection molds. These molds are used to make a variety of car components in automotive industry. The prediction is based on the geometric features of the mold parts to be made. The accuracy of the predicted times is high enough to be used as a tool for the design stage of the mold parts, e.g. guiding the design process in order to get the lowest production time. [ABSTRACT FROM AUTHOR]

The main aim of this paper is to predict NO and NO2 concentrations 4 days in advance by comparing two artificial intelligence learning methods, namely, multi-layer perceptron and support vector machines, on two kinds of spatial embedding of the temporal time series. Hourly values of NO and NO2 concentrations, as well as meteorological variables were recorded in a cross-road monitoring station with heavy traffic in Szeged, in order to build a model for predicting NO and NO2 concentrations several hours in advance. The prediction of NO and NO2 concentrations was performed partly on the basis of their past values, and partly on the basis of temperature, humidity and wind speed data. Since NO can be predicted more accurately, its values were considered primarily when forecasting NO2. Time series prediction can be interpreted in a way that is suitable for artificial intelligence learning. Two effective learning methods, namely, multi-layer perceptron and support vector regression are used to provide efficient non-linear models for NO and NO2 time series predictions. Multi-layer perceptron is widely used to predict these time series, but support vector regression has not yet been applied for predicting NO and NO2 concentrations. Three commonly used linear algorithms were considered as references: 1-day persistence, average of several day persistence and linear regression. Based on the good results of the average of several day persistence, a prediction scheme was introduced, which forms weighted averages instead of simple ones. The optimization of these weights was performed with linear regression in linear case and with the learning methods mentioned in non-linear case. Concerning the NO predictions, the non-linear learning methods give significantly better predictions than the reference linear methods. In the case of NO2, the improvement of the prediction is considerable, however, it is less notable than for NO. [ABSTRACT FROM AUTHOR]

An automatic road sign recognition system first locates road signs within images captured by an imaging sensor on-board of a vehicle, and then identifies the detected road signs. This paper presents an automatic neural-network-based road sign recognition system. First, a study of the existing road sign recognition research is presented. In this study, the issues associated with automatic road sign recognition are described, the existing methods developed to tackle the road sign recognition problem are reviewed, and a comparison of the features of these methods is given. Second, the developed road sign recognition system is described. The system is capable of analysing live colour road scene images, detecting multiple road signs within each image, and classifying the type of road signs detected. The system consists of two modules: detection and classification. The detection module segments the input image in the hue-saturation-intensity colour space, and then detects road signs using a Multi-layer Perceptron neural-network. The classification module determines the type of detected road signs using a series of one to one architectural Multi-layer Perceptron neural networks. Two sets of classifiers are trained using the Resillient-Backpropagation and Scaled-Conjugate-Gradient algorithms. The two modules of the system are evaluated individually first. Then the system is tested as a whole. The experimental results demonstrate that the system is capable of achieving an average recognition hit-rate of 95.96% using the scaled-conjugate-gradient trained classifiers. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach to the control of the cutting force on the basis of the internal model control (IMC) principle. The main goal is to control a single output variable, the cutting force, by changing a single input variable, the feedrate. A neural model is used as an internal model to determine the control inputs (feedrate) necessary to keep the cutting force constant. Three approaches, the fuzzy logic controller (FLC), the direct inverse controller (DIC) and the IMC, based on artificial neural networks (IMC-NN), are simulated and their performances are assessed in terms of several performance measurements. The results demonstrate that IMC-NN strategy provides a better disturbance rejection than FLC for the cases analysed. [ABSTRACT FROM AUTHOR]

This paper discusses the application of a class of feed-forward Artificial Neural Networks (ANNs) known as Multi-Layer Perceptrons (MLPs) to two vision problems: recognition and pose estimation of 3D objects from a single 2D perspective view; and handwritten digit recognition. In both cases, a multi-MLP classification scheme is developed that combines the decisions of several classifiers. These classifiers operate on the same feature set for the 3D recognition problem whereas different feature types are used for the handwritten digit recognition. The backpropagation learning rule is used to train the MLPs. Application of the MLP architecture to other vision problems is also briefly discussed. [ABSTRACT FROM AUTHOR]

Data fusion (integration) techniques are combined with MultiLayer Feedforward (backpropagation) neural networks in order to improve the inversion —extraction of the key describing parameters — of oblique-incidence ionograms (plots of apparent height of reflection versus transmission frequency). Two separate investigations were undertaken: first, the incorporation of vertical ionogram data to improve inversion; and secondly, the fusion of ionogram data gathered from a 2D array of ionosondes (the ground-based radio frequency transmitters). With the former, the average percentage errors obtained by incorporating data fusion dropped by a factor of five when compared with single ionogram inversion. Moreover, gradients of ionospheric parameters (critical frequency, layer height and thickness) were also obtained. In the case of the latter, the error rate dropped by a similar factor, and by even more when vertical ionograms were incorporated. Better results were forthcoming when a hierarchical network was used to invert the ionograms prior to fusion, compared with directly fusing the ionogram array data. [ABSTRACT FROM AUTHOR]

A major obstacle in point-of-regard monitoring for human-computer interaction has been the contaminating effect of head movement. A novel solution to this problem has been simulated. A multilayer perceptron converts the distorted pattern of four infrared sources, reflected in the cornea of the user, into an estimate for the point-of-regard. Using an idealised model of the eye the simulation combined vertical, horizontal, pitch and yaw head movements with eye excursions typical of individuals seated in front of a computer display. Results show that the method is capable of dramatically reducing the effects of small head movements. Under these viewing conditions the error in the point-of-regard estimate exhibited a standard deviation of under 0.5 mm. It is concluded that such a scheme could form an attractive solution to the long-standing problem of head movement artefact. [ABSTRACT FROM AUTHOR]