Network Method Research Articles

Neural Network-Based Fusion of InSAR and Optical Digital Elevation Models with Consideration of Local Terrain Features

InSAR and optical techniques represent two principal approaches for the generation of large-scale Digital Elevation Models (DEMs). Due to the inherent limitations of each technology, a single data source is insufficient to produce high-quality DEM products. The increasing deployment of satellites has generated vast amounts of InSAR and optical DEM data, thereby providing opportunities to enhance the quality of final DEM products through the more effective utilization of the existing data. Previous research has established that complete DEMs generated by InSAR technology can be combined with optical DEMs to produce a fused DEM with enhanced accuracy and reduced noise. Traditional DEM fusion methods typically employ weighted averaging to compute the fusion results. Theoretically, if the weights are appropriately selected, the fusion outcome can be optimized. However, in practical scenarios, DEMs frequently lack prior information on weights, particularly precise weight data. To address this issue, this study adopts a fully connected artificial neural network for elevation fusion prediction. This approach represents an advancement over existing neural network models by integrating local elevation and terrain as input features and incorporating curvature as an additional terrain characteristic to enhance the representation of terrain features. We also investigate the impact of terrain factors and local terrain feature as training features on the fused elevation outputs. Finally, three representative study areas located in Oregon, USA, and Macao, China, were selected for empirical validation. The terrain data comprise InSAR DEM, AW3D30 DEM, and Lidar DEM. The results indicate that compared to traditional neural network methods, the proposed approach improves the Root-Mean-Squared Error (RMSE) ranges, from 5.0% to 12.3%, and the Normalized Median Absolute Deviation (NMAD) ranges, from 10.3% to 26.6%, in the test areas, thereby validating the effectiveness of the proposed method.

Three-stage binarization of color document images based on discrete wavelet transform and generative adversarial networks

The efficient extraction of text information from the background in degraded color document images is an important challenge in the preservation of ancient manuscripts. The imperfect preservation of ancient manuscripts has led to different types of degradation over time, such as page yellowing, staining, and ink bleeding, seriously affecting the results of document image binarization. This work proposes an effective three-stage network method to image enhancement and binarization of degraded documents using generative adversarial networks (GANs). Specifically, in Stage-1, we first split the input images into multiple patches, and then split these patches into four single-channel patch images (gray, red, green, and blue). Then, three single-channel patch images (red, green, and blue) are processed by the discrete wavelet transform (DWT) with normalization. In Stage-2, we use four independent generators to separately train GAN models based on the four channels on the processed patch images to extract color foreground information. Finally, in Stage-3, we train two independent GAN models on the outputs of Stage-2 and the resized original input images (512 × 512) as the local and global predictions to obtain the final outputs. The experimental results show that the Avg-Score metrics of the proposed method are 77.64, 77.95, 79.05, 76.38, 75.34, and 77.00 on the (H)-DIBCO 2011, 2013, 2014, 2016, 2017, and 2018 datasets, which are at the state-of-the-art level. The implementation code for this work is available at https://github.com/abcpp12383/ThreeStageBinarization.

A likelihood-based monitoring method for social networks with auto-correlated labeled Poisson model

Social network monitoring aims to detect any significant change in the structure of communications from the normal condition. The normal condition can be a set of networks considered as a reference set. This set is represented in dynamic cases by a probability model. Statistical process control methods can be applied to signal changes in the reference set. These signals can be useful for decision-makers in various applications, from marketing analysis to criminal investigations. In most of the developed models, the networks are assumed independent over time. However, this assumption may be questioned in real-world applications. Accordingly, this article formulates the social network as an auto-correlated labeled Poisson model and monitors it using a control method based on likelihood statistics. A simulation study is performed to evaluate the performance of the proposed methods in terms of the average run length criterion. The obtained results revealed that the proposed methods can reliably detect significant changes.

Counting the number of solutions in satisfiability problems with tensor-network message passing.

We investigate how to count the number of solutions in the Boolean satisfiability (SAT) problem, a fundamental problem in theoretical computer science that has applications in various domains. We convert the problem to a spin-glass problem in statistical physics and approximately compute the entropy of the problem, which corresponds to the logarithm of the number of solutions. We propose a new method for the entropy computing problem based on a combination of the tensor network method and the message-passing algorithm. The significance of the proposed method is its ability to consider the effects of both long loops and short loops present in the factor graph of the SAT problem. We validate the efficacy of our approach using 3-SAT problems defined on the random graphs and structured graphs, and show that the proposed method gives more accurate results on the number of solutions than the standard belief propagation algorithms. We also discuss the applications of our method across a wide range of combinatorial optimization problems.

Micro segment analysis and machine learning prediction for thermal-hydraulic parameters of propane condensation flow in a PCHE straight channel

In the LNG regasification process, the printed circuit heat exchangers(PCHE) is an ideal choice for vaporizer, with propane as the intermediate working medium in the hot side of PCHE. In this work, numerical simulations of propane condensation flow and heat transfer at different operating parameters was carried out in a semi-circular straight channel of PCHE. The micro-segment analysis method was used to present the local heat transfer coefficient and pressure drop in the channel. The results indicate that the propane condensation process can be divided into three stages: the “constant temperature condensation stage” in the droplet condensation state, the “varying temperature condensation stage” in the film condensation state, and the “subcooled stage” in the liquid phase state. Different operating parameters can affect the start and end position of each stage, as well as the parameter values in the same state. An artificial neural network method was used to predict the local heat transfer coefficient and pressure drop, with the predicting error of less than 5 % and 10 % respectively. This study provides a machine learning method to accurately predict the flow and heat transfer parameters for propane condensation flow in PCHE channel and valuable for future design of LNG vaporizer.

Effects of different treatment measures on the efficacy of diabetic foot ulcers: a network meta-analysis.

Through a network meta-analysis, we compared different treatment measures for patients with diabetic foot ulcers (DFU), assessing their impact on the healing of DFU and ranking them accordingly. We searched the PubMed, the China National Knowledge Infrastructure (CNKI), Embase, the WanFang and the WeiPu database. The retrieval time was from database establishment to January 2024, and retrieval entailed subject and free words. Randomized controlled trials (RCTs) with different treatment measures for DFU were included. Data extraction and evaluation were based on the PRISMA guidelines. Meta-analyses using pairwise and network methods were employed to compare and rank the effectiveness of different treatments for DFU. Ultimately, we included 57 RCTs involving a total of 4,826 patients with DFU. When it comes to ulcer healing rates, compared to standard of care(SOC),platelet-rich plasma(PRP), hyperbaric oxygen therapy(HBOT), topical oxygen therapy(TOT), acellular dermal matrix(ADM), and stem cells(SCs) in both direct meta-analysis(DMA) and network meta-analysis(NMA) can effectively increase the complete healing rate. For Scs+PRP, a statistically significant improvement was only observed in the NMA. Moreover, when compared to the negative pressure wound therapy(NPWT) group, the PRP+NPWT group was more effective in promoting the complete healing of ulcers. In terms of promoting the reduction of ulcer area, no statistical differences were observed among various treatment measures. When it comes to ulcer healing time, both PRP and NPWT can effectively shorten the healing time compared to SOC. Furthermore, when compared to the NPWT group, the combined treatment of PRP and ultrasonic debridement(UD) with NPWT is more effective in reducing healing time. In terms of amputation rates and adverse reactions, the PRP group effectively reduced the amputation rate and adverse reactions for patients with DFU. Additionally, compared to the NPWT group, the combined treatment of PRP and UD with NPWT reduced the incidence of adverse reactions. However, no significant differences were observed among other treatment measures in terms of amputation rates and adverse reactions. The ranking results showed that the efficacy of PRP+NPWT and UD+NPWT in promoting ulcer healing, reducing ulcer area, shortening healing time, decreasing amputation rates and adverse reactions is superior to that of the alone PRP group, NPWT group, and UD group. Conversely, the SOC group demonstrates the least effective performance in all aspects. Due to the particularity of the wound of DFU, the standard of care is not effective, but the new treatment scheme has a remarkable effect in many aspects. And the treatment of DFU is not a single choice, combined with a variety of methods often achieve better efficacy, and will not bring more adverse reactions.

Deep learning-based post-disaster energy management and faster network reconfiguration method for improvement of restoration time

Natural disasters in the world often cause power outages. To improve post-disaster response and restoration time, a method for energy management and network reconfiguration in emergencies has been proposed, where energy storages systems (ESSs) in distribution networks are used to support the system balance immediately and then the network reconfiguration is accelerated based on deep learning techniques. First, when power failures occur, the outputs of ESSs are calculated optimally in terms of the proposed optimization model under the constraints of minimizing line losses. After the injections of ESSs in emergencies, the network reconfiguration is still needed due to the limitation of ESS capacities. To avoid calculations of power flow repeatedly, a deep-learning based reconfiguration method for distribution networks has been proposed, where the deep-learning model is trained offline and can output the parameters of power flow immediately when a state combination of distribution generators (DGs), loads and power lines, which is an effective way to accelerate the network reconfiguration. If many reconfiguration schemes are found, a decision-making method with preferences are used to select one according to different situations. Finally, cases are designed, and simulation results show that the calculation time of a reconfiguration schemes are reduced significantly by almost one hundred multiples.

The underlying mechanisms of alignment in error backpropagation through arbitrary weights

Understanding the mechanisms by which plasticity in millions of synapses in the brain is orchestrated to achieve behavioral and cognitive goals is a fundamental question in neuroscience. In this regard, insights from learning methods in artificial neural networks (ANNs) and in particular supervised learning using backpropagation (BP) seem inspiring. However, the implementation of BP requires exact matching between forward and backward weights, which is unrealistic given the known connectivity pattern in the brain (known as the “weight transport problem”). Notably, it has been shown that under certain conditions, error BackPropagation Through Arbitrary Weights (BP-TAW) can lead to a partial alignment between forward and backward weights (weight alignment or WA). This learning algorithm, which is also known as feedback alignment (FA), can result in surprisingly good degrees of accuracy in simple classification tasks. However, the underlying mechanisms and mathematical basis of WA are not thoroughly understood. In this work, we demonstrate the mathematical basis of WA and answer the question of why and in what conditions WA occurs. We show that the occurrence of WA in ANNs is induced by statistical properties of the output and error signals of neurons, such as autocorrelation and cross-correlation, and can happen even in the absence of learning or reduction of the loss function. Moreover, we show that WA can be improved significantly by limiting the norm of input weights to neurons and that such a weight normalization (WN) method can improve the classification accuracy of BP-TAW. The findings presented can be used to further improve the performance of BP-TAW and open new ways for exploring possible learning mechanisms in biological neural networks without exact matching between forward and backward weights.

Efficient Quantum Circuit Simulation by Tensor Network Methods on Modern GPUs

Efficient simulation of quantum circuits has become indispensable with the rapid development of quantum hardware. The primary simulation methods are based on state vectors and tensor networks. As the number of qubits and quantum gates grows larger in current quantum devices, traditional state-vector based quantum circuit simulation methods prove inadequate due to the overwhelming size of the Hilbert space and extensive entanglement. Consequently, brutal force tensor network simulation algorithms become the only viable solution in such scenarios. The two main challenges faced in tensor network simulation algorithms are optimal contraction path finding and efficient execution on modern computing devices, with the latter determines the actual efficiency. In this study, we investigate the optimization of such tensor network simulations on modern GPUs and propose general optimization strategies from two aspects: computational efficiency and accuracy. Firstly, we propose to transform critical Einstein summation operations into GEMM operations, leveraging the specific features of tensor network simulations to amplify the efficiency of GPUs. Secondly, by analyzing the data characteristics of quantum circuits, we employ extended precision to ensure the accuracy of simulation results and mixed precision to fully exploit the potential of GPUs, resulting in faster and more precise simulations. Our numerical experiments demonstrate that our approach can achieve a 3.96x reduction in verification time for random quantum circuit samples in the 18-cycle case of Sycamore, with sustained performance exceeding 21 TFLOPS on one A100. This method can be easily extended to the 20-cycle case, maintaining the same performance, accelerating by 12.5x compared to the state-of-the-art CPU-based results and 4.48-6.78x compared to the state-of-the-art GPU-based results reported in the literature. Furthermore, the strategies presented in this article hold general applicability for accelerating problems that can be tackled by contracting tensor networks such as graphical models and combinatorial optimizations.

Research on Temperature-Rise Characteristics of Motor Based on Simplified Lumped-Parameter Thermal Network Model

The thermal management of a driving motor is related to the performance and economy of the vehicle. The lumped-parameter thermal network (LPTN) method can provide a model basis for motor temperature-rise control and effectively shorten the development cycle of motor thermal performance design. In this study, an 80 kw water-cooled permanent magnet synchronous motor was used and the accurate thermal model of a motor was built using the LPTN. On the basis of the accurate thermal model, the simplified thermal model of a motor was obtained by reducing the complexity of the model. The temperature-rise test of the end winding and magnetic steel of the motor was carried out under some working conditions. The test conditions were selected according to continuous external characteristics and operating characteristics of the motor. Compared with the experimental results, the temperature-rise error of the two thermal models was less than 5%. The temperature-rise error of the simplified thermal model was less than 3% compared with the accurate thermal model. Therefore, the simplified thermal model can be used to quickly predict the temperature rise of the motor.

Identification of Distorted Gamma-Ray Signature Patterns Using Digital Filtering and Auto-Associative Memory Implemented with a Hopfield Neural Network

The detection and identification of radioactive sources in search applications involve analyzing passive gamma-ray emissions from high-level radioactive materials. This process uses a mobile detector-spectrometer in a complex field test environment. Recently, the use of artificial intelligence for gamma-ray spectrum analysis has shown promising results. However, challenges persist in identifying isotopic signatures from spectral measurements that may be distorted due to source shielding, random variations in natural radioactive background, or insufficient measurement time to obtain clear spectral lines. This paper presents a novel intelligent signature recognition method that combines digital filtering techniques with an artificial Hopfield Neural Network (HNN). The HNN leverages auto-associative memory to store training sample patterns and match them with incoming gamma spectra from distorted sources. It restores the testing sources’ measurements by finding the closest matching signature patterns in the spectral library. Before HNN recognition, the measured spectrum undergoes preprocessing with a digital image filter to reduce fluctuations. Performance of the proposed method is evaluated using a set of gamma-ray spectra measured with a sodium iodide detector. The data collected include measurements from six pure samples: 241Am, 60Co, 137Cs, 192Ir, 239Pu, and 235U, which are used for training and validation (i.e. six cases). Additionally, the data set contains 24 distorted synthesized sources with various fluctuating backgrounds. Test results demonstrate the potential of the proposed method to accurately recognize the correct isotope with high precision, achieving an accuracy rate exceeding 85%. Furthermore, the proposed method exhibits superior performance compared to the conventional multiple regression fitting and simple feedforward neural network methods.

Fault diagnosis method of Bayesian network based on association rules

Fault diagnosis method of Bayesian network based on association rules

An Interval Neural Network Method for Identifying Static Concentrated Loads in a Population of Structures

During the design and validation of structural engineering, the focus is on a population of similar structures, not just one. These structures face uncertainties from external environments and internal configurations, causing variability in responses under the same load. Identifying the real load from these dispersed responses is a significant challenge. This paper proposes an interval neural network (INN) method for identifying static concentrated loads, where the network parameters are internalized to create a new INN architecture. Additionally, the paper introduces an improved interval prediction quality loss function indicator named coverage and mean square criterion (CMSC), which balances the interval coverage rate and interval width of the identified load, ensuring that the median of the recognition interval is closer to the real load. The efficiency of the proposed method is assessed through three examples and validated through comparative research against other loss functions. Our research findings indicate that this approach enhances the interval accuracy, robustness, and generalization of load identification. This improvement is evident even when faced with challenges such as limited training data and significant noise interference.

Regulating Modality Utilization within Multimodal Fusion Networks.

Multimodal fusion networks play a pivotal role in leveraging diverse sources of information for enhanced machine learning applications in aerial imagery. However, current approaches often suffer from a bias towards certain modalities, diminishing the potential benefits of multimodal data. This paper addresses this issue by proposing a novel modality utilization-based training method for multimodal fusion networks. The method aims to guide the network's utilization on its input modalities, ensuring a balanced integration of complementary information streams, effectively mitigating the overutilization of dominant modalities. The method is validated on multimodal aerial imagery classification and image segmentation tasks, effectively maintaining modality utilization within ±10% of the user-defined target utilization and demonstrating the versatility and efficacy of the proposed method across various applications. Furthermore, the study explores the robustness of the fusion networks against noise in input modalities, a crucial aspect in real-world scenarios. The method showcases better noise robustness by maintaining performance amidst environmental changes affecting different aerial imagery sensing modalities. The network trained with 75.0% EO utilization achieves significantly better accuracy (81.4%) in noisy conditions (noise variance = 0.12) compared to traditional training methods with 99.59% EO utilization (73.7%). Additionally, it maintains an average accuracy of 85.0% across different noise levels, outperforming the traditional method's average accuracy of 81.9%. Overall, the proposed approach presents a significant step towards harnessing the full potential of multimodal data fusion in diverse machine learning applications such as robotics, healthcare, satellite imagery, and defense applications.

CPINet: Towards A Novel Cross-Polarimetric Interaction Network for Dual-Polarized SAR Ship Classification

With the rapid development of the modern world, it is imperative to achieve effective and efficient monitoring for territories of interest, especially for the broad ocean area. For surveillance of ship targets at sea, a common and powerful approach is to take advantage of satellite synthetic aperture radar (SAR) systems. Currently, using satellite SAR images for ship classification is a challenging issue due to complex sea situations and the imaging variances of ships. Fortunately, the emergence of advanced satellite SAR sensors has shed much light on the SAR ship automatic target recognition (ATR) task, e.g., utilizing dual-polarization (dual-pol) information to boost the performance of SAR ship classification. Therefore, in this paper we have developed a novel cross-polarimetric interaction network (CPINet) to explore the abundant polarization information of dual-pol SAR images with the help of deep learning strategies, leading to an effective solution for high-performance ship classification. First, we establish a novel multiscale deep feature extraction framework to fully mine the characteristics of dual-pol SAR images in a coarse-to-fine manner. Second, to further leverage the complementary information of dual-pol SAR images, we propose a mixed-order squeeze–excitation (MO-SE) attention mechanism, in which the first- and second-order statistics of the deep features from one single-polarized SAR image are extracted to guide the learning of another polarized one. Then, the intermediate multiscale fused and MO-SE augmented dual-polarized deep feature maps are respectively aggregated by the factorized bilinear coding (FBC) pooling method. Meanwhile, the last multiscale fused deep feature maps for each single-polarized SAR image are also individually aggregated by the FBC. Finally, four kinds of highly discriminative deep representations are obtained for loss computation and category prediction. For better network training, the gradient normalization (GradNorm) method for multitask networks is extended to adaptively balance the contribution of each loss component. Extensive experiments on the three- and five-category dual-pol SAR ship classification dataset collected from the open and free OpenSARShip database demonstrate the superiority and robustness of CPINet compared with state-of-the-art methods for the dual-polarized SAR ship classification task.

Impact Load Localization Based on Multi-Scale Feature Fusion Convolutional Neural Network.

In order to achieve impact load localization of complex structures such as ships, this paper proposes a multi-scale feature fusion convolutional neural network (MSFF-CNN) method for impact load localization. An end-to-end machine learning model is used, where the raw vibration signals of impact loads are directly fed into the network model to avoid the process of feature extraction. Automatic feature learning and feature concatenation of the signal are achieved through four independent convolutional layers, each using a different size of convolutional kernel. Data normalization and L2 regularization techniques are introduced to enhance the data and prevent overfitting. Classification and localization of impact loads are accomplished using a softmax classification layer. Validation experiments are carried out using a ship's stern compartment model. Our results show that the classification and localization accuracy of the impact load sample group of MSFF-CNN reaches 94.29% compared with a traditional CNN. The method further improves the ability of the network to extract state features, takes local perception and global vision into account, effectively improves the classification ability of the model, and has good prospects for engineering applications.

EAV: EEG-Audio-Video Dataset for Emotion Recognition in Conversational Contexts

Understanding emotional states is pivotal for the development of next-generation human-machine interfaces. Human behaviors in social interactions have resulted in psycho-physiological processes influenced by perceptual inputs. Therefore, efforts to comprehend brain functions and human behavior could potentially catalyze the development of AI models with human-like attributes. In this study, we introduce a multimodal emotion dataset comprising data from 30-channel electroencephalography (EEG), audio, and video recordings from 42 participants. Each participant engaged in a cue-based conversation scenario, eliciting five distinct emotions: neutral, anger, happiness, sadness, and calmness. Throughout the experiment, each participant contributed 200 interactions, which encompassed both listening and speaking. This resulted in a cumulative total of 8,400 interactions across all participants. We evaluated the baseline performance of emotion recognition for each modality using established deep neural network (DNN) methods. The Emotion in EEG-Audio-Visual (EAV) dataset represents the first public dataset to incorporate three primary modalities for emotion recognition within a conversational context. We anticipate that this dataset will make significant contributions to the modeling of the human emotional process, encompassing both fundamental neuroscience and machine learning viewpoints.

A study of neural network-based evaluation methods for pipelines with multiple corrosive regions

In recent years, significant developments have been made in methods for assessing the remaining strength of corroded pipelines. However, existing methods have limitations as they mainly focus on the local impact of corrosion defects. This study explores evaluation methods using neural networks to predict the ultimate resistance of pipelines containing multiple corrosive regions. Firstly, based on the validated method, the study generates a dataset comprising 3,000 corroded pipeline models and pixelates the corrosion information of these models via digital images. Then, three neural network evaluation frameworks are constructed: a Multilayer Perceptron (MLP) using the overall corrosion matrix, an MLP based on corrosion feature parameters, and Convolutional Neural Networks (CNN) based on corrosion images. Following this, the study analyzes the relationship between various corrosion parameters and failure pressure, compares the training effectiveness of the three neural network methods, and validates the accuracy and applicability of the proposed approach. The results indicated that various corrosion features should be considered when evaluating corroded pipelines, particularly depth. In addition, all three neural network-based methods show improved applicability and reliability compared to traditional evaluation methods, with CNN-image having the highest evaluation accuracy (correlation coefficient = 0.9564, average error = 3.46%).

Regression prediction of critical exhaust volumetric flow rate in tunnel fire with two-point extraction ventilation based on neural network method

In this study, a Genetic Algorithm-Backpropagation Neural Network (GA-BPNN) model was developed to predict critical exhaust volumetric flow rate in tunnel fires with two-point extraction ventilation system. Seven influencing factors served as inputs for the model, while the critical exhaust volumetric flow rate was the output. Experimental data from two reduced-scale tunnels were utilized to both train and validate the GA-BPNN model. The results demonstrate a satisfactory prediction performance, with key metrics such as Mean Absolute Percentage Error (MAPE), Relation Coefficient (R2), and Root Mean Square Error (RMSE) achieving values of 0.0384, 0.9989, and 3.5258, respectively. Importantly, the model maintains a maximum Absolute Percentage Error (APE) below 10 %. In contrast, the traditional BPNN model exhibited an APE of approximately 18.9 %. Moreover, the predictive results of the GA-BPNN model were compared with those of a previously semi-empirical formula. It was observed that the semi-empirical predictions exhibited APE exceeding 20 % for certain data points. This further underscores the superiority of the GA-BPNN model in predicting critical exhaust volumetric flow rates in tunnel fires with two-point ventilation system. These findings affirm the feasibility and effectiveness of GA-BPNN regression model as a reliable method for predicting critical exhaust volumetric flow rates under multiple factors.

Evaluating and Reducing Subgroup Disparity in AI Models: An Analysis of Pediatric COVID-19 Test Outcomes.

Artificial Intelligence (AI) fairness in healthcare settings has attracted significant attention due to the concerns to propagate existing health disparities. Despite ongoing research, the frequency and extent of subgroup fairness have not been sufficiently studied. In this study, we extracted a nationally representative pediatric dataset (ages 0-17, n=9,935) from the US National Health Interview Survey (NHIS) concerning COVID-19 test outcomes. For subgroup disparity assessment, we trained 50 models using five machine learning algorithms. We assessed the models' area under the curve (AUC) on 12 small (<15% of the total n) subgroups defined using social economic factors versus the on the overall population. Our results show that subgroup disparities were prevalent (50.7%) in the models. Subgroup AUCs were generally lower, with a mean difference of 0.01, ranging from -0.29 to +0.41. Notably, the disparities were not always statistically significant, with four out of 12 subgroups having statistically significant disparities across models. Additionally, we explored the efficacy of synthetic data in mitigating identified disparities. The introduction of synthetic data enhanced subgroup disparity in 57.7% of the models. The mean AUC disparities for models with synthetic data decreased on average by 0.03 via resampling and 0.04 via generative adverbial network methods.