Deep Neural Network Approach Research Articles

Accelerating regional-scale groundwater flow simulations with a hybrid deep neural network model incorporating mixed input types: A case study of the northeast Qatar aquifer

ABSTRACT This study presents the ‘Dual Path CNN-MLP’, a novel hybrid deep neural network (DNN) architecture that merges the strengths of convolutional neural networks (CNNs) and multilayer perceptrons (MLPs) for regional groundwater flow simulations. This model stands out from previous DNN approaches by managing mixed input types, including both imagery and numerical vectors. Such flexibility allows the diverse nature of groundwater data to be efficiently utilized without the need to convert it into a uniform format, which often leads to oversimplification or unnecessary expansion of the dataset. When applied to the northeast Qatar aquifer, the model demonstrates high accuracy in simulating transient groundwater flow fields, benchmarked against the well-established MODFLOW model. The model's efficacy is confirmed through k-fold cross-validation, showing an error margin of less than 12% across all examined locations. The study also examines the model's ability to perform uncertainty analysis using Monte Carlo simulations, finding that it achieves around 1% average absolute percentage error in estimating the mean hydraulic head. Errors are mostly found in areas with significant variations in the hydraulic head. Switching to this ML model from the conventional MODFLOW simulator boosts computational efficiency by about 99%, showcasing its advantage for tasks like uncertainty analysis in repetitive groundwater simulations.

From order to reorder: Assessment of the living environment of hydropower resettlement for just energy transition in China

Hydropower projects have contributed significantly to the world’s just energy transition, while at the cost of massive-scale resettlement and displacement. The principles of a just transition ensure that the low-carbon transition process is fair, inclusive and socially responsible. However, since hydropower resettlers are often involuntary, it incurs many socio-environmental disruptions for the affected. Whether the post-resttlement living environment of the resettlers can also be just and sustainable needs to be effectively evaluated. This paper establishes and validates an optimized deep neural network (DNN) approach for predictability and reliability. A comprehensive set of living environment development indicators is then used to assess the just transition in the living environment of resettlers, based on three typical hydropower station resettlement cases in the Wujiang River Basin in Guizhou Province, China. The results indicate that individual living condition, community development, and fairness in market and social participation have important impacts on the just transition of resettlers’ living environments. By summarizing the composite affiliation and score values of each sub-indicator, it is found that the Hongjiadu hydropower station has the lowest composite score of the resettlers’ living environment, ranging from 0.476 to 0.680, followed by the Dongfeng hydropower station, from 0.483 to 0.709. The resettlers of Qianzhong Hydropower Hub have the highest living environment status after resettlement, from 0.497 to 0.794. Although there are differences in the living environment scores of different hydropower stations, it is argued that the transition of the living environment in the stepped power stations along the Wujiang River is just and sustainable in the long term. This paper provides valuable insights into the just transition of hydropower and the living environment of the resettlers, so as to foster sustainable development policymaking.

Machine Learning Analysis of Enhanced Biodegradable Phoenix dactylifera L./HDPE Composite Thermograms.

Worldwide, environmental groups and policymakers are focusing on waste recycling to create economic value and on the decomposition of waste by leveraging on scarce resources. This work, therefore, explores the thermal decomposition of enhanced biodegradable polymer matrices made from a mixture of discarded Phoenix dactylifera L./high-density polyethylene (PD/HDPE) using the machine learning analysis of experimental data. The experimental results of these samples were obtained via thermogravimetric (TGA) analysis under an oxidation-free environment, with heating rates of 10, 20, and 40 °C·min-1 and a degradation temperature range from 25 to 600 °C. The TGA analyses revealed the continued dependence of the actual percentage weight loss by these materials as a test function of the degradation temperature, shifting thermograms to temperature maxima consistent with increasing heating rates. Although high-density polyethylene (HDPE) materials were found to be thermally more stable than Phoenix dactylifera L. (PD) materials, PD/HDPE composite materials contained a significant amount of residual ash. Using a machine learning deep neural network approach for this process, significantly improved learning algorithms have been developed, which reduces the overall cost function (residual error) to almost zero (0.025) after just over a million iterations (epochs) and provides predictions that overlap with the experimental results (R2~1). Learning algorithms, along with optimized synaptic weights and biases, were employed to predict the behaviour of PD materials based on experimental thermograms conducted at higher degradation temperatures, typically ranging between 600 and 1000 °C. Predicted data using the enhanced learning algorithms completely overlapped the experiments (R2~1) for these higher degradation temperatures with near unity correlation if the decomposition of the materials continued until the residue was attained. With this approach, it is possible to predict and optimize the thermal characteristics of PD and HDPE with greater efficiency, which reduces the need for multiple design iterations and experimentation.

An Effective Deep Learning Approach For Handwritten Ol-Chiki Character Recognition

In this paper, we propose a deep convolution neural network approach to recognize Ol-Chiki handwritten characters. Here we describe recognition of handwritten basic characters of Ol-Chiki script, used by more than 10 million tribal people in India mostly from Assam, Bengal, Bihar, Odisha and Jharkhand. There are 30 basic characters and 10 numeral digits in Ol-Chiki. We have used a dataset of 10000 handwritten isolated character samples written by 100 persons. Convolution Neural Network (CNN) architecture has been used for the recognition of handwritten isolated Ol-Chiki characters. Our system has been tested on Ol-Chiki isolated dataset and we have achieved 92% accuracy on almost on all Ol-Chiki characters.

A Deep Learning Approach for Chromium Detection and Characterization from Soil Hyperspectral Data.

High levels of chromium (Cr) in soil pose a significant threat to both humans and the environment. Laboratory-based chemical analysis methods for Cr are time consuming and expensive; thus, there is an urgent need for a more efficient method for detecting Cr in soil. In this study, a deep neural network (DNN) approach was applied to the Land Use and Cover Area frame Survey (LUCAS) dataset to develop a hyperspectral soil Cr content prediction model with good generalizability and accuracy. The optimal DNN model was constructed by optimizing the spectral preprocessing methods and DNN hyperparameters, which achieved good predictive performance for Cr detection, with a correlation coefficient value of 0.79 on the testing set. Four important hyperspectral bands with strong Cr sensitivity (400-439, 1364-1422, 1862-1934, and 2158-2499 nm) were identified by permutation importance and local interpretable model-agnostic explanations. Soil iron oxide and clay mineral content were found to be important factors influencing soil Cr content. The findings of this study provide a feasible method for rapidly determining soil Cr content from hyperspectral data, which can be further refined and applied to large-scale Cr detection in the future.

Adaptive deep homogenization theory for periodic heterogeneous materials

We present an adaptive physics-informed deep homogenization neural network (DHN) approach to formulate a full-field micromechanics model for elastic and thermoelastic periodic arrays with different microstructures. The unit cell solution is approximated by fully connected multilayers via minimizing a loss function formulated in terms of the sum of residuals from the stress equilibrium and heat conduction partial differential equations (PDEs), together with interfacial traction-free or adiabatic boundary conditions. In comparison, periodicity boundary conditions are directly satisfied by introducing a network layer with sinusoidal functions. Fully trainable weights are applied on all collocation points, which are simultaneously trained alongside the network weights. Hence, the network automatically assigns higher weights to the collocation points in the vicinity of the interface (particularly challenging regions of the unit cell solution) in the loss function. This compels the neural networks to enhance their performance at these specific points. The accuracy of adaptive DHN is verified against the finite element and the elasticity solution respectively for elliptical and circular cylindrical pores/fibers. The advantage of the adaptive DHN over the original DHN technique is justified by considering locally irregular porous architecture where pore–pore interaction makes training the network particularly slow and hard to optimize.

Deep Neural Networks with Spacetime RBF for Solving Forward and Inverse Problems in the Diffusion Process

Deep Neural Networks with Spacetime RBF for Solving Forward and Inverse Problems in the Diffusion Process

Deep learning neural network approach to thermal-wave imaging of damage in solids with application to diffusivity measurements of a green (unsintered) metal powder compact slab

A method is proposed to measure the thermal diffusivity of a green (unsintered) metal powder compact slab by combining numerical solution and deep learning. The thermal diffusion and thermal-wave equation is first discretized in space and time domains. Square wave excitation signals at different frequencies are considered, and the corresponding thermal wave responses are obtained. The space and frequency dependencies of the thermal-wave amplitude and phase are obtained by using lock-in thermography. Then, deep learning candidate neural network sets composed of spatial coordinates, excitation frequency, amplitude, and phase values are quickly and massively obtained for training the deep learning network. The validity of the deep learning network is verified by predicting the known thermal diffusivity, and the parameters and robustness of the deep learning network are discussed. Experimentally, a square wave modulated laser beam is used to illuminate one side of a green metal powder compact slab and the spatial distribution of amplitude and phase at three excitation frequencies is obtained. The predicted slab thermal diffusivity is extracted through the formed deep learning network, and the validity of the predicted value is verified by comparing with independent measurements.

Enhancing breast cancer segmentation and classification: An Ensemble Deep Convolutional Neural Network and U-net approach on ultrasound images

Breast cancer is a condition where the irregular growth of breast cells occurs uncontrollably, leading to the formation of tumors. It poses a significant threat to women’s lives globally, emphasizing the need for enhanced methods of detecting and categorizing the disease. In this work, we propose an Ensemble Deep Convolutional Neural Network (EDCNN) model that exhibits superior accuracy compared to several transfer learning models and the Vision Transformer model. Our EDCNN model integrates the strengths of the MobileNet and Xception models to improve its performance in breast cancer detection and classification. We employ various preprocessing techniques, including image resizing, data normalization, and data augmentation, to prepare the data for analysis. By following these measures, the formatting is optimized, and the model’s capacity to make generalizations is improved. We trained and evaluated our proposed EDCNN model using ultrasound images, a widely available modality for breast cancer imaging. The outcomes of our experiments illustrate that the EDCNN model attains an exceptional accuracy of 87.82% on Dataset 1 and 85.69% on Dataset 2, surpassing the performance of several well-known transfer learning models and the Vision Transformer model. Furthermore, an AUC value of 0.91 on Dataset 1 highlights the robustness and effectiveness of our proposed model. Moreover, we highlight the incorporation of the Grad-CAM Explainable Artificial Intelligence (XAI) technique to improve the interpretability and transparency of our proposed model. Additionally, we performed image segmentation using the U-Net segmentation technique on the input ultrasound images. This segmentation process allowed for the identification and isolation of specific regions of interest, facilitating a more comprehensive analysis of breast cancer characteristics. In conclusion, the study presents a creative approach to detecting and categorizing breast cancer, demonstrating the superior performance of the EDCNN model compared to well-established transfer learning models. Through advanced deep learning techniques and image segmentation, this study contributes to improving diagnosis and treatment outcomes in breast cancer.

Hybrid active shape model and deep neural network approach for lung cancer detection

Hybrid active shape model and deep neural network approach for lung cancer detection

An analysis of fractional piecewise derivative models of dengue transmission using deep neural network

This manuscript investigates a fractional piecewise dengue transmission model using singular and non-singular kernels. The existence results and uniqueness of the solution are established by using the approach of fixed point and in the framework of piecewise derivative and integral. To obtain the approximate solution of the considered models we apply a piecewise numerical iteration scheme which is based on Newton interpolation polynomials. Furthermore, the numerical scheme for piecewise derivatives encompasses singular and non-singular kernels. This study aims to enhance our understanding of dengue internal transmission dynamics by using a novel piecewise derivative approach that considers both singular and non-singular kernels. This work contributes to clarifying the concept of piecewise derivatives and their significance in understanding crossover dynamics. Moreover, a deep neural network approach is employed with high accuracy in training, testing, and validation of data to investigate the specified disease problem. This methodology is employed to thoroughly investigate the intricacies of the specified disease problem.

Novel hybrid optimization based adaptive deep convolution neural network approach for human activity recognition system

Novel hybrid optimization based adaptive deep convolution neural network approach for human activity recognition system

DeepAHR: a deep neural network approach for recognizing Arabic handwritten recognition

DeepAHR: a deep neural network approach for recognizing Arabic handwritten recognition

Revealing neural dynamical structure of C. elegans with deep learning

Caenorhabditis elegans serves as a common model for investigating neural dynamics and functions of biological neural networks. Data-driven approaches have been employed in reconstructing neural dynamics. However, challenges remain regarding the curse of high-dimensionality and stochasticity in realistic systems. In this study, we develop a deep neural network (DNN) approach to reconstruct the neural dynamics of C.elegans and study neural mechanisms for locomotion. Our model identifies two limit cycles in the neural activity space: one underpins basic pirouette behavior, essential for navigation, and the other introduces extra turns. The combination of two limit cycles elucidates predominant locomotion patterns in neural imaging data. The corresponding energy landscape explains the switching strategies between two limit cycles, quantitatively, and provides testable predictions on neural functions and circuit roles. Our work provides a general approach to study neural dynamics by combining imaging data and stochastic modeling.

High Dynamic Range Image Reconstruction from Saturated Images of Metallic Objects.

This study considers a method for reconstructing a high dynamic range (HDR) original image from a single saturated low dynamic range (LDR) image of metallic objects. A deep neural network approach was adopted for the direct mapping of an 8-bit LDR image to HDR. An HDR image database was first constructed using a large number of various metallic objects with different shapes. Each captured HDR image was clipped to create a set of 8-bit LDR images. All pairs of HDR and LDR images were used to train and test the network. Subsequently, a convolutional neural network (CNN) was designed in the form of a deep U-Net-like architecture. The network consisted of an encoder, a decoder, and a skip connection to maintain high image resolution. The CNN algorithm was constructed using the learning functions in MATLAB. The entire network consisted of 32 layers and 85,900 learnable parameters. The performance of the proposed method was examined in experiments using a test image set. The proposed method was also compared with other methods and confirmed to be significantly superior in terms of reconstruction accuracy, histogram fitting, and psychological evaluation.

Massive Open Online Course (MOOCs) English Courses Based On Hamiltonian Deep Neural Network

Numerous researchers are now concentrating on the English course on intercultural communication. Notwithstanding, these courses encounter several obstacles, including student variations in traditional English for cross-cultural instruction, developing students' cross-cultural communication skills, and improving the quality of instruction. To overcome these problems, in this manuscript, a Hamiltonian deep neural network (NN) approach is proposed for improving the teaching quality of massive open online course (MOOCs) based English course. Initially, data is given from English-language massive open online courses (MOOCs) teaching adult basic life support (ABLS) dataset. Afterward the data is fed to Switching Hierarchical Gaussian Filter. The pre-processing output is provided to the Hamiltonian Deep Neural Network (HDNN) used to improve the teaching quality of MOOC English course. The learnable of the optimized is using Coati Optimization Algorithm (COA). The proposed technique is executed in Python and the effectiveness of the proposed MOOCEC-HDNN method is improved by using various performances evaluating metrics like, teaching Efficiency, Robustness Gain, teaching quality and system Efficiency are analysed. The proposed MOOCEC-HDNN method is shows the higher teaching efficiency of 78%, Robustness Gain of 28dB and teaching quality 68% while comparing other existing methods such as massive open online course (MOOCs) based English course based on Gannet algorithm and Neural Network (MOOCEC-GA-NN), MOOCs based English course based on K-modes algorithm and Neural Network (MOOCEC-KMA-NN) and MOOCs based English course based on Conventional Neural Network (MOOCEC-CNN), respectively.

A Review on Sign Language Recognition and Learning System

This paper presents innovative approaches to various aspects of sign language recognition and learning, catering to the diverse needs within this field. It introduces a system utilizing 2D image sampling and concatenation, trained with convolutional neural networks, achieving accurate and robust sign recognition even with low-cost cameras. Another system, SignQuiz, offers a cost-effective web-based solution for learning finger-spelled signs in Indian Sign Language (ISL), outperforming traditional printed mediums. Additionally, a dynamic hand gesture recognition system employing deep learning architectures demonstrates superior performance over existing methods, addressing challenges such as hand segmentation and sequence modeling. Furthermore, an efficient deep convolutional neural network approach for hand gesture recognition, including transfer learning, achieves high recognition rates across various datasets. Finally, a robust model for static sign recognition in ISL, utilizing CNNs and evaluating on a diverse dataset, attains exceptional accuracy and outperforms previous works. These contributions collectively advance the field of sign language recognition and learning, offering solutions that are accurate, cost-effective, and efficient, thereby facilitating better communication and interaction for the hearing-impaired community. Key Words: CNN, Sign language recognition, Hand gesture recognition, Machine learning, Transfer Learning, SVM.

Predicting the Number of Software Faults using Deep Learning

The software testing phase requires considerable time, effort, and cost, particularly when there are many faults. Thus, developers focus on the evolution of Software Fault Prediction (SFP) to predict faulty units in advance, therefore, improving software quality significantly. Forecasting the number of faults in software units can efficiently direct software testing efforts. Previous studies have employed several machine learning models to determine whether a software unit is faulty. In this study, a new, simple deep neural network approach that can adapt to the type of input data was designed, utilizing Convolutional Neural Networks (CNNs) and Multi-Layer Perceptron (MLP), to predict the number of software faults. Twelve open-source software project datasets from the PROMISE repository were used for testing and validation. As data imbalance can negatively impact prediction accuracy, the new version of synthetic minority over-sampling technique (SMOTEND) was used to resolve data imbalance. In experimental results, a lower error rate was obtained for MLP, compared to CNN, reaching 0.195, indicating the accuracy of this prediction model. The proposed approach proved to be effective when compared with two of the best machine learning models in the field of prediction. The code will be available on GitHub.

A Review on Low-Dose Emission Tomography Post-Reconstruction Denoising with Neural Network Approaches.

Low-dose emission tomography (ET) plays a crucial role in medical imaging, enabling the acquisition of functional information for various biological processes while minimizing the patient dose. However, the inherent randomness in the photon counting process is a source of noise which is amplified low-dose ET. This review article provides an overview of existing post-processing techniques, with an emphasis on deep neural network (NN) approaches. Furthermore, we explore future directions in the field of NN-based low-dose ET. This comprehensive examination sheds light on the potential of deep learning in enhancing the quality and resolution of low-dose ET images, ultimately advancing the field of medical imaging.

A deep neural network approach with attention mechanism to improve the quality of target observation for UAVs

Owing to the limitations of onboard equipment, the direct acquisition of high-resolution images in unmanned aerial vehicle (UAV) target observation tasks is a challenge. However, with the development of deep learning, super-resolution reconstruction has become a popular technique for enhancing image resolution. Though, current super-resolution reconstruction methods can enhance UAV target observation images to some extent, most of them are unable to strike a balance between the number of parameters, computational cost, and reconstruction effect. Furthermore, limited research has been conducted for overcoming the cascade coupling problem in UAV target observation images. To solve these problems and provide a lightweight, fast, and high-quality super-resolution method for UAV target observation images, an enhanced attention dense cross network (EADCN) is proposed. This method includes excellent attention mechanisms, such as the double large kernel attention module (D-LKA) and multi-aware channel attention distillation block (MAADB). In addition to benchmark datasets, a UAV target observation dataset (UAV-TOD) is created to evaluate the performance of EADCN. The experimental results demonstrate that the EADCN has fewer parameters, low computational complexity, high accuracy, and fast processing speed, making it suitable for deployment in UAVs.