Cascade Neural Network Research Articles

Self-organizing Cascade Neural Network Based on Differential Evolution with Better and Nearest Option for System Modeling

Self-organizing Cascade Neural Network Based on Differential Evolution with Better and Nearest Option for System Modeling

Position Control of Quadrotor UAV Based on Cascade Fuzzy Neural Network

In this article, a cascade fuzzy neural network (FNN) control approach is proposed for position control of quadrotor unmanned aerial vehicle (UAV) system with high coupling and underactuated. For the attitude loop with limited range, the FNN controller parameters were trained offline using flight data, whereas for the position loop, the method based on FNN compensation proportional-integral-derivative (PID) was adopted to tune the system online adaptively. This method not only combined the advantages of fuzzy systems and neural networks but also reduced the amount of calculation for cascade neural network control. Simulations of fixed set point flight and spiral and square trajectory tracking flight were then conducted. The comparison of the results showed that our method had advantages in terms of minimizing overshoot and settling time. Finally, flight experiments were carried out on a DJI Tello quadrotor UAV. The experimental results showed that the proposed controller had good performance in position control.

Comparison of the prediction performance of separating coal in separation equipment using machine learning based cubic regression modelling and cascade neural network modelling

ABSTRACT The availability of low-grade coal with a high amount of ash has urged the improvisation of separation equipment with minimal or no water utilization. The present work addresses the study on the separation equipment performance with different moisture coal. The experimental results were obtained in terms of separation efficiency. After obtaining the experimental results, the mathematical modeling results were obtained using different techniques. The cubic regression and cascade neural network models were considered to study the mathematical correlation with experimental results. The R-squared value of each mathematical modeling technique was correlated with the model fitting to check the model’s validity. The results clearly showed that the cubic model fitting for the experimental condition had provided an excellent R-squared value varying from 92% to 99%. The cascade model fitting for the experimental condition has provided a higher R-squared value, i.e., more than 99%. Results show that for all experimental conditions, the cascade model fitting of the neural network technique provides the significant mathematical modeling technique suitable for predicting the separation equipment’s performance compared to the cubic model of the regression technique.

Predicting the surfactant-polymer flooding performance in chemical enhanced oil recovery: Cascade neural network and gradient boosting decision tree

Surfactant-polymer flooding is one of the most important enhanced oil recovery (EOR) techniques, which refers to the injection of surfactant slugs and polymer drives. Two crucial decision-making parameters in EOR operations are net present value (NPV) and oil recovery factor (RF). Herein, various intelligent models, based on multilayer perceptron (MLP), cascade neural network (CNN), radial basis function (RBF), neural networks as well as support vector regression (SVR), and decision tree (DT) algorithms are proposed toward estimating these two parameters with respect to polymer drive size, surfactant slug size, the salinity of polymer drive, Kv/Kh ratio, surfactant concentration, and polymer concentration in polymer drive and surfactant slug. The results exhibited the outperformance of the CNN model trained with the Levenberg Marquardt algorithm in forecasting the RF and NPV with average absolute errors of 0.66% and 1.95%, respectively. Moreover, the results of the sensitivity analysis reflected that the most effective inputs on the predicted value of RF were surfactant concentration and surfactant slug size, while surfactant concentration and polymer concentration in surfactant slug could considerably affect the NPV model’s output. Lastly, the outlier detection analysis revealed that the employed data is valid and only two points were detected as outliers.

Optimal Deep-Learning-Enabled Intelligent Decision Support System for SARS-CoV-2 Classification.

Intelligent decision support systems (IDSS) for complex healthcare applications aim to examine a large quantity of complex healthcare data to assist doctors, researchers, pathologists, and other healthcare professionals. A decision support system (DSS) is an intelligent system that provides improved assistance in various stages of health-related disease diagnosis. At the same time, the SARS-CoV-2 infection that causes COVID-19 disease has spread globally from the beginning of 2020. Several research works reported that the imaging pattern based on computed tomography (CT) can be utilized to detect SARS-CoV-2. Earlier identification and detection of the diseases is essential to offer adequate treatment and avoid the severity of the disease. With this motivation, this study develops an efficient deep-learning-based fusion model with swarm intelligence (EDLFM-SI) for SARS-CoV-2 identification. The proposed EDLFM-SI technique aims to detect and classify the SARS-CoV-2 infection or not. Also, the EDLFM-SI technique comprises various processes, namely, data augmentation, preprocessing, feature extraction, and classification. Moreover, a fusion of capsule network (CapsNet) and MobileNet based feature extractors are employed. Besides, a water strider algorithm (WSA) is applied to fine-tune the hyperparameters involved in the DL models. Finally, a cascaded neural network (CNN) classifier is applied for detecting the existence of SARS-CoV-2. In order to showcase the improved performance of the EDLFM-SI technique, a wide range of simulations take place on the COVID-19 CT data set and the SARS-CoV-2 CT scan data set. The simulation outcomes highlighted the supremacy of the EDLFM-SI technique over the recent approaches.

Finger Joint Angle Estimation With Visual Attention for Rehabilitation Support: A Case Study of the Chopsticks Manipulation Test

Most East Asian rehabilitation centers offer chopsticks manipulation tests (CMT). In addition to impaired hand function, approximately two-thirds of stroke survivors have visual impairment related to eye movement. This article investigates the significance of combining finger joint angle estimation and a visual attention measurement in CMT. We present a multiscopic framework that consists of microscopic, mesoscopic, and macroscopic levels. We develop a feature extraction technique to extract the kinematic finger model at the microscopic level. At the mesoscopic level, we propose an active perception ability to detect the position and geometry of the finger on the chopsticks. The proposed framework estimates the proximal interphalangeal (PIP) joint angle on the index finger during CMT using fully connected cascade neural networks (FCC-NN). At the macroscopic level, we implement a cognitive ability by measuring visual attention during CMT. We further evaluate the proposed framework with a conventional test that counts the number of peanuts (NP) which are moved from one bowl to another using chopsticks within a particular time frame. For the evaluation, we introduce three parameters, namely joint angle estimation movement (JAEM), chopstick attention movement (CAM), and chopstick tip movement (CTM), by detecting the local minima and maxima of the signal. According to the experiment results, the velocity of these three parameters could indicate improvement in hand and eye function during CMT. We expect this study to benefit therapists and researchers by providing valuable information that is not accessible in the clinic. Code and datasets are available online at https://github.com/anom-tmu/cmt-attention.

Interactive Visualization of Hyperspectral Images Based on Neural Networks.

It is challenging to interpret hyperspectral images in an intuitive and meaningful way, as they usually contain hundreds of dimensions. We develop a visualization tool for hyperspectral images based on neural networks, which allows a user to specify the regions of interest, select bands of interest, and obtain hyperspectral classification results in a scatterplot generated from hyperspectral features. A cascade neural network is trained to generate a scatterplot that matches the cluster centers labeled by the user. The inferred scatterplot not only shows the clusters of points, but also reveals relationships of substances. The trained neural network can be reused for time-varying hyperspectral data analysis without retraining. Our visualization solution can keep domain experts in the analytical loop and provide an intuitive analysis of hyperspectral images while identifying different substances, which are difficult to be realized using existing hyperspectral image analysis techniques.

Machine Learning Classifiers for Surface Crack Detection in Fracture Experiments

Correctly determining the onset of fracture is crucial when performing mechanical experiments. Commonly carried out by visual inspection, here an image-based machine learning approach is proposed to classify cracked and un-cracked specimens. It yields the potential to objectify and automate crack detection, thereby removing sources of uncertainty and error from the post-processing of experiments. More than 30’000 speckle-pattern images obtained from 77 experiments on three specimen geometries are evaluated. They comprise uniaxial tension, notched tension as well as axisymmetric V-bending experiments. Statistical texture features are extracted from all images. They include both first-order (variance, skewness, kurtosis) and higher-order statistical texture features, i.e. Haralick features. The discriminatory power of the texture information is evaluated based on the Fisher's Discriminant Ratio and feature correlations are identified and quantified. Image texture feature subsets of high discriminatory power are used to parse neural network architectures of different complexities from simple perceptron to feed-forward and cascade neural networks. It is found that a small subset of the investigated texture features is highly significant for all experiments. Using this feature subset in conjunction with multi-layer, non-linear and low complexity feed-forward network architectures classification accuracies in the order of 99% are obtained. At the same time, it is shown that linear classifiers are not sufficient to robustly distinguish the state of the specimens, even when high discriminatory power features are used.Graphical Abstract: [Display omitted]

Criminal face identification system using deep learning algorithm multi-task cascade neural network (MTCNN)

Nowadays criminal activities are growing at an exponential rate. Crime prevention by effective identification of criminals is the main issue before the police and on the other hand, the availability of police officers is inadequate. There are various technological solutions for detecting criminals but they are not up to the mark. In this project, a face detection and recognition system for criminal identification is developed using the multi-task cascade neural network. This system will be able to detect faces and recognize faces of criminals automatically in real-time. This system would also just require a single image of the criminal to recognize him, also known as one-shot learning. The purpose is to identify the criminal face, retrieve the information stored in the database for the identified criminal and a notification is sent to the police personnel with all the details and the location at which he was under the surveillance of the camera.

Evolving Hybrid Cascade Neural Network Genetic Algorithm Space–Time Forecasting

Design: At the heart of time series forecasting, if nonlinear and nonstationary data are analyzed using traditional time series, the results will be biased. At the same time, if just using machine learning without any consideration given to input from traditional time series, not much information can be obtained from the results because the machine learning model is a black box. Purpose: In order to better study time series forecasting, we extend the combination of traditional time series and machine learning and propose a hybrid cascade neural network considering a metaheuristic optimization genetic algorithm in space–time forecasting. Finding: To further show the utility of the cascade neural network genetic algorithm, we use various scenarios for training and testing while also extending simulations by considering the activation functions SoftMax, radbas, logsig, and tribas on space–time forecasting of pollution data. During the simulation, we perform numerical metric evaluations using the root-mean-square error (RMSE), mean absolute error (MAE), and symmetric mean absolute percentage error (sMAPE) to demonstrate that our models provide high accuracy and speed up time-lapse computing.

Performance Evaluation of Classifiers Created using Elman Back-Propagation and Cascade Feed-forward Neural Networks

Huge data being captured in our day-to-day activities are mostly imbalance. Such data therefore, calls for fast, accurate and robust techniques, through which they could be analyzed in order to fast-track early decision making. A Cascade Feed-forward Neural Networks and Elman Backpropagation are known techniques in neural network domain and their efficacies is therefore tested on separable data in this study. The objective of this study is to evaluate the performance of these techniques in solving a linear classification problem. The linear classification of data involves, splitting of separable data into two distinct clusters. In order to achieve the goal of this study, linear classifiers were created using the two aforementioned techniques. Both network structures were exposed to the same dataset and similar parameter configurations were set for each technique. The model that emanates through each technique was simulated using the set of untrained data. In order to determine the accuracy of each model created through each technique, their Mean Absolute Errors (MAE) was computed. The performance of each model was determined based on the value of MAE. The error computation for the simulated output reveals that, cascade neural network gives an error of 0.0928; while the model created using Elman Backpropagation network gives a relatively lower error of 0.0661. It can be inferred from this study, that, both techniques are capable of fitting accurate classifiers from dataset and specifically, both techniques are very suitable for binary classification of separable data.

Grey-Fuzzy Hybrid Optimization and Cascade Neural Network Modelling in Hard Turning of AISI D2 Steel

Nowadays hard turning is noticed to be the most dominating machining activity especially for difficult to cut metallic alloys. Attributes of dry hard turning are highly influenced by the amount of heat generation during cutting. Some major challenges are rapid tool wear, lower tool-life span, and poor surface finish but simultaneously generated heat is enough to provide thermal softening of hard work material and facilitates easier shear deformation thus easy cutting. Also, plenty of works reported the utilization of various cooling methods as well as coolants which successfully retard the intensity of cutting heat but this leads to additional cost as well as environmental and health issues. However, still, there is scope to select proper cutting tool materials, its geometry, and appropriate values of cutting parameters to get favorable machining outcomes under dry hard turning and avoid the cooling cost, environmental, and health issue. Considering these challenges, current work utilizes PVD-coated (TiAlN) carbide insert in dry hard turning of AISI D2 steel. The multi-responses like tool-flank wear, chip morphology, and chip reduction coefficient are considered. The amalgamation of input processing variables must be optimum for the effectual performance of hard to process materials turning. Generally, the Fuzzy logic hypothesis represents the uncertainties co-related with fuzziness, and deficiency in the data concerned with the problem. Further, to achieve the best combination of input cutting terms, grey-fuzzy hybrid optimization (Type I and Type II) is utilized considering the Gaussian membership function. Type II grey-fuzzy system attributed to 15 % less error (between GRG and GFG) compared to Type I. Hence, Type II grey-fuzzy system is utilized to get the optimal set of input terms. The optimal combination of input terms is found as t-1 (0.15 mm), s-4 (0.25 mm/rev) and is Vc-2 (100 m/min) which is comparable to the results obtained under spray impingement cooling using CVD tool in the literature. However, hard turning can be assessed under the dry condition with a PVD tool at the obtained optimal input condition for industrial uses. Further, six different types of cascade-forward-back propagation neural network modelling is accomplished. Among all models, CFBNN-4 model exhibited the best prediction results with a mean absolute error of 2.278% for flank wear (VBc) and 0.112% for the chip reduction coefficient (CRC). However, this model can be recommended for other engineering modelling problems. The outcomes of this research may be of immense importance to the tool manufacturers and machining industry.

Predict the required cost to develop Software Engineering projects by Using Machine Learning

Software project cost prediction is a very important task during building and developing software projects. This process helps software project engineers to accurately manage and plan their resources in terms of cost estimation. However, Need for accurate cost development prediction model for a software project is not a simple procedure. Predicting the cost required while developing software engineering projects is the most difficult challenge that attracts the attention of researchers and practitioners. This paper adopts a new model in estimating the cost of building or developing software engineering projects using a machine learning approach. The results proves that machine learning methods can be used to predict program cost with high accuracy rate compared with traditional software estimation techniques. The proposed model in this research was trained on the NASA (National Aeronautics and Space Administration) data set, which contains the characteristics of 60 projects in addition to the real cost of the projects. An analysis of the results of the implementation for the proposed methods showed that the cost Predicting process using K-Nearest Neighbours algorithm (KNN), Cascade Neural Networks (CNN) and Elman Neural Networks (ENN) It has the ability to predict the costs required to build or develop software engineering projects, K-Nearest Neighbours algorithm has shown high accuracy for Predict the required cost to develop Software Engineering projects Compared to Cascade Neural Networks and Elman Neural Networks ENN.

Intelligent LED Certification System in Mass Production.

It is impossible to effectively use light-emitting diodes (LEDs) in medicine and telecommunication systems without knowing their main characteristics, the most important of them being efficiency. Reliable measurement of LED efficiency holds particular significance for mass production automation. The method for measuring LED efficiency consists in comparing two cooling curves of the LED crystal obtained after exposure to short current pulses of positive and negative polarities. The measurement results are adversely affected by noise in the electrical measuring circuit. The widely used instrumental noise suppression filters, as well as classical digital infinite impulse response (IIR), finite impulse response (FIR) filters, and adaptive filters fail to yield satisfactory results. Unlike adaptive filters, blind methods do not require a special reference signal, which makes them more promising for removing noise and reconstructing the waveform when measuring the efficiency of LEDs. The article suggests a method for sequential blind signal extraction based on a cascading neural network. Statistical analysis of signal and noise values has revealed that the signal and the noise have different forms of the probability density function (PDF). Therefore, it is preferable to use high-order statistical moments characterizing the shape of the PDF for signal extraction. Generalized statistical moments were used as an objective function for optimization of neural network parameters, namely, generalized skewness and generalized kurtosis. The order of the generalized moments was chosen according to the criterion of the maximum Mahalanobis distance. The proposed method has made it possible to implement a multi-temporal comparison of the crystal cooling curves for measuring LED efficiency.

Comparative study of neural network architectures for modelling nonlinear optical pulse propagation

Ultrashort pulses have a crucial role in the evolution of different areas of science such as ultra fast imaging, femtochemistry and high harmonic spectroscopy and therefore, diagnosing and reconstructing ultrashort pulses is important. The interplay of dispersion and nonlinear effects gives rise to a wide variety of pulse dynamics. On the other hand, dispersion, linear and nonlinear effects have proven to be a fundamental bottleneck for high speed communications. Conventionally, time consuming and computationally inefficient algorithms are used to solve these problems. Since machine learning (ML) has proved to be more powerful than other analytical methods we show a comprehensive comparison of different neural network (NN) architectures to learn the mapping of the nonlinear schrödinger equation (NLSE). We have used a ML based approach to construct the distorted output pulse resulting from nonlinear propagation through the fiber. Additionally, the trained network can also predict the dispersion and nonlinear parameters. We have also reconstructed the temporal and spectral profile of transmitted pulse from the pulse distorted due to propagation through a highly nonlinear fiber (HNLF). These techniques can work without the knowledge of fiber parameters. A detailed comparison of six different NN based techniques namely fully connected NN (FCNN), cascade NN (CaNN), convolutional NN (CNN), long short term memory networks (LSTM), bidirectional LSTM (BiLSTM) and gated recurrent unit (GRU) is presented. Results show that a FCNN regressor outperforms the all other architectures.

Application of Elman and Cascade neural network (ENN and CNN) in comparison with adaptive neuro fuzzy inference system (ANFIS) to predict key fuel properties of ABE-diesel blends

ABSTRACT Today, a growing interest to use Acetone-Butanol-Ethanol (ABE) as a biofuel has emerged. Fuel properties play important roles to determine engine’s performance, combustion, and emission behaviors. Yet, the determination of fuel properties is expensive and time-consuming. Previous studies on ABE did not provide information on how to predict its fuel properties. This study developed an Elman and Cascade neural networks (ENN and CNN) and compared their results with adaptive neuro inference system (ANFIS) to predict ABE’s key fuel properties. Three properties, i.e., calorific value, density, and kinematic viscosity were used as the target outputs, while ABE, acetone, butanol, and ethanol ratio were selected as the input parameters. The ENN and CNN models were trained using 10 different training algorithms, while the ANFIS model was examined using eight unique membership functions. To evaluate the prediction accuracy of each model, six different parameters were employed. Results showed that, compared to ENN and CNN, the ANFIS model gave the best performance accuracy with the least errors to predict the key fuel properties of ABE-diesel blends. For calorific value, density, and kinematic viscosity prediction, the best results of the ANFIS model were given by triangular, Pi curve, and trapezoidal membership functions, respectively. Therefore, ANFIS gave the best model of all the investigated models in this study.

An Improved Method for Extracting Photovoltaic Module I–V Characteristic Curve Using Hybrid Learning Machine System

Abstract This paper provides an improved method for predicting the I–V curve of the photovoltaic module using a hybrid machine learning system. The proposed method is based on a random forest algorithm and a cascade forward neural network. A random forest algorithm is used to predict a specific factor that is subsequently used as an input for the cascade neural network to remove the correlation between voltage and current. Then, the actual current is predicted using the cascade neural network. This procedure assures the ability of the proposed model to extract the I–V curve of any photovoltaic module regardless of its rating or type. A dataset that contains values for air temperature, solar radiation, voltage, and current of two polycrystalline photovoltaic modules is used in the training process of the proposed algorithm. The hybrid model has general inputs such as ambient temperature, solar radiation, and data from the photovoltaic module datasheet (Voc and Isc). The proposed model is trained, tested, and validated by 86% of the data. Meanwhile, 14% of the data are used for testing. Thus, the proposed model is tested using unknown data so as to avoid overfitting. Results show that the proposed model is very accurate in predicting I–V curves based on three types of errors which are mean absolute percentage error (0.68%), mean bias error (0.0191 A), and root-mean-squared error (0.04458 A). This hybrid model can be used to obtain the I–V curves for several types of photovoltaic modules.

DRPnet: automated particle picking in cryo-electron micrographs using deep regression

BackgroundIdentification and selection of protein particles in cryo-electron micrographs is an important step in single particle analysis. In this study, we developed a deep learning-based particle picking network to automatically detect particle centers from cryoEM micrographs. This is a challenging task due to the nature of cryoEM data, having low signal-to-noise ratios with variable particle sizes, shapes, distributions, grayscale variations as well as other undesirable artifacts.ResultsWe propose a double convolutional neural network (CNN) cascade for automated detection of particles in cryo-electron micrographs. This approach, entitled Deep Regression Picker Network or “DRPnet”, is simple but very effective in recognizing different particle sizes, shapes, distributions and grayscale patterns corresponding to 2D views of 3D particles. Particles are detected by the first network, a fully convolutional regression network (FCRN), which maps the particle image to a continuous distance map that acts like a probability density function of particle centers. Particles identified by FCRN are further refined to reduce false particle detections by the second classification CNN. DRPnet’s first CNN pretrained with only a single cryoEM dataset can be used to detect particles from different datasets without retraining. Compared to RELION template-based autopicking, DRPnet results in better particle picking performance with drastically reduced user interactions and processing time. DRPnet also outperforms the state-of-the-art particle picking networks in terms of the supervised detection evaluation metrics recall, precision, and F-measure. To further highlight quality of the picked particle sets, we compute and present additional performance metrics assessing the resulting 3D reconstructions such as number of 2D class averages, efficiency/angular coverage, Rosenthal-Henderson plots and local/global 3D reconstruction resolution.ConclusionDRPnet shows greatly improved time-savings to generate an initial particle dataset compared to manual picking, followed by template-based autopicking. Compared to other networks, DRPnet has equivalent or better performance. DRPnet excels on cryoEM datasets that have low contrast or clumped particles. Evaluating other performance metrics, DRPnet is useful for higher resolution 3D reconstructions with decreased particle numbers or unknown symmetry, detecting particles with better angular orientation coverage.

Industrial Temperature Measurement of Flame and Flue gases using IOT and Image processing

This study effort contracts with measuring the flame temperature in the combustion chamber along with a scheme to monitor the temperature of flue gases from the power plants. Presently, the gases from the exhaust are monitore dusing gas analyzers which are positioned at the exit point of the chimney. These gas analyzers are prone to cold end corrosion which will lead to malfunctioning of the equipment causing erroneous results during measurements. Thiskind of indigenous technique facilitates to develop an Industrial IoT (IIoT) temperature monitoring system at power plants. A combination of Principal Component Analysis (PCA) with Cascaded Neural Network (CNN) is used. The features extracted and reduced serve as inputs to CNN. An efficient technology to scrutinize and measure the temperature inside the combustion chamber thereby reducing the amount of flue gas emissions by the partnership of two methods.

Fast and Robust Cascade Model for Multiple Degradation Single Image Super-Resolution.

Single Image Super-Resolution (SISR) is one of the low-level computer vision problems that has received increased attention in the last few years. Current approaches are primarily based on harnessing the power of deep learning models and optimization techniques to reverse the degradation model. Owing to its hardness, isotropic blurring or Gaussians with small anisotropic deformations have been mainly considered. Here, we widen this scenario by including large non-Gaussian blurs that arise in real camera movements. Our approach leverages the degradation model and proposes a new formulation of the Convolutional Neural Network (CNN) cascade model, where each network sub-module is constrained to solve a specific degradation: deblurring or upsampling. A new densely connected CNN-architecture is proposed where the output of each sub-module is restricted using some external knowledge to focus it on its specific task. As far we know, this use of domain-knowledge to module-level is a novelty in SISR. To fit the finest model, a final sub-module takes care of the residual errors propagated by the previous sub-modules. We check our model with three state-of-the-art (SOTA) datasets in SISR and compare the results with the SOTA models. The results show that our model is the only one able to manage our wider set of deformations. Furthermore, our model overcomes all current SOTA methods for a standard set of deformations. In terms of computational load, our model also improves on the two closest competitors in terms of efficiency. Although the approach is non-blind and requires an estimation of the blur kernel, it shows robustness to blur kernel estimation errors, making it a good alternative to blind models.