Deep Convolutional Generative Adversarial Networks Research Articles

Application of Deep Convolutional Generative Adversarial Network to Identification of Bridge Structural Damage

In recent years, the increase in the number of vehicles on the road has impacted the service life of bridges, and thus, their performance. In China, bridge inspections must be performed and documented at least once per year using labor-intensive and subjective approaches. In this study, a computationally efficient damage identification methodology for bridges was developed based on a deep convolutional generative adversarial network (DCGAN). To predict the damage to bridges, the vibration data of the structural damage are extracted from sensor measurements on the bridges and applied to the DCGAN model. The proposed damage identification method was validated on a simple supported beam bridge and a continuous girder bridge. The effects of sample size and noise on the identification accuracy were also evaluated. The performance of the DCGAN in the damage identification process is robust and reliable. This method for structural health monitoring offers a more objective evaluation of the overall performance and condition of a bridge, thus enabling bridge owners to efficiently allocate finite maintenance resources.

Simulation data-driven fault diagnosis method for metro traction motor bearings under small samples and missing fault samples

Abstract Traction motor bearings are crucial for guaranteeing the safe operation of metro vehicles. However, in the metro traction motor bearing fault diagnosis, there are usually problems of small samples and missing fault samples, leading to inaccurate results. Therefore, a novel bearing fault diagnosis method utilizing a track-vehicle-bearing coupled dynamic model and the improved deep convolutional generative adversarial network-multiscale convolutional neural network with mixed attention (IDCG-MAMCNN) model is proposed in this paper. The IDCG-MAMCNN model combines an improved deep convolutional generative adversarial network (IDCGAN) with a multi-scale convolutional neural network with mixed attention (MA-MCNN). Specifically, simulation data is first provided by the coupled dynamic model to supplement missing fault samples. Secondly, the IDCGAN, along with a training method that involves pre-training models with simulation samples and fine-tuning models with experimental samples, is introduced to generate high-quality samples and augment experimental samples under small samples. Lastly, the MA-MCNN serves as the classification model, trained with the augmented dataset comprising experimental, simulation, and generated samples. The fault diagnosis performance of the proposed method is evaluated on the experimental samples of two bearing datasets under small samples and various conditions of missing fault samples. It has been demonstrated by the experimental results that the proposed method exhibits robust fault diagnosis performance and generates high-quality samples under small samples and missing fault samples. Furthermore, the proposed method showcases its adaptability to different operation speeds.

HQ-DCGAN: Hybrid quantum deep convolutional generative adversarial network approach for ECG generation

The class imbalance of electrocardiogram (ECG) data is a serious impediment to the development of diagnostic systems for heart disease. To address this issue, this paper proposes HQ-DCGAN, a hybrid quantum deep convolutional generative adversarial network, specifically designed for the generation of ECGs. The proposed algorithm employs different quantum convolutional layers for the generator and discriminator as feature extractors and utilizes parameterized quantum circuits (PQCs) to enhance computational capabilities, along with the model-feature mapping process. Moreover, this algorithm preserves the nonlinearity and scalability inherent to classical convolutional neural networks (CNNs), thereby optimizing the utilization of quantum resources, and ensuring compatibility with contemporary quantum devices. In addition, this paper proposes a novel evaluation metric, 1D Fréchet Inception Distance (1DFID), to assess the quality of the generated ECG signals. Simulation experiments show that HQ-DCGAN exhibits strong performance in ECG signal generation. Furthermore, the generated signals achieve an average classification accuracy of 82.2%, outperforming the baseline algorithms. It has been experimentally proven that HQ-DCGAN is friendly to currently noisy intermediate-scale quantum (NISQ) computers, in terms of both number of qubits and circuit depths, while improving the stability.

High energy absorption design of porous metals using deep learning

Due to its remarkable energy absorption properties, porous metals have widespread applications in engineering. However, the high randomness of pore morphology greatly hinders the effective design and analysis of high energy absorption structures. To address this challenge, this paper first introduces a deep learning-based framework for high energy absorption-oriented design of random porous metals structures. The framework comprises two steps: (i) a generator powered by Wasserstein deep convolutional generative adversarial network is developed to swiftly generate a vast design space (∼one million samples) of porous metals with real random pore morphology. (ii) an inverse search strategy based on convolutional neural network is applied to quickly pick out the optimal structure with the best energy absorption from the design space. Results show that the optimal energy absorption is about 17.71 % higher than the maximum value of initial structures from CT scan. Additionally, a 575-fold increase in computational efficiency is achieved compared to the traversal search using finite element method. Subsequently, the deformation process of the optimal structure is analyzed focusing on the pore morphology and compression performance, showing that random porous metals with uniformly sized pores are capable of withstanding higher stress under the same strain and exhibit no yield band during compression. Inspired by this, a structural homogenization method is introduced and validated to create porous metal structure with stable microstructure evolution, extended plateau stress and high energy absorption.

Predicting multiphase flow behavior of methane in shallow unconfined aquifers using conditional deep convolutional generative adversarial network

Numerical modeling is an essential tool for geoscience applications involving multiphase flow behavior. Performing numerical simulation is, however, computationally intensive due to multi-scale, non-linear, and multi-physics nature of mechanisms controlling multiphase flow dynamics. Additionally, the computational challenges associated with traditional numerical simulations limit their applicability for repetitive simulations required in inverse modeling, uncertainty analysis, and optimization tasks. To overcome these limitations, surrogate modeling has recently emerged as a viable solution. A surrogate model, which functions as a regression model, is trained using numerical simulation data. It approximates the input–output relationships within a dynamic fluid environment. We design surrogate models, based on conditional deep convolutional generative adversarial network (cDC-GAN), to map the cross-domain between input and output pairs in a multiphase multicomponent system, focusing on methane (CH4) migration in shallow unconfined aquifers. The cDC-GAN technique is selected due to its ability to generate high-fidelity surrogate models that effectively capture complex spatial distributions and heterogeneities. This technique excels in handling high-dimensional data, learning the intricate relationships between inputs and outputs, and producing realistic representations of subsurface phenomena. The trained cDC-GANs consider capillary entry pressure heterogeneity as input because this geologic attribute places first-order control on CH4 migration in subsurface. The outputs include CH4 saturation, water saturation, residual saturation of CH4, CH4 mole fraction, and CH4 molality. For each output, a separate cDC-GAN model is trained. Once trained, cDC-GANs can at any given time determine CH4 distribution in a shallow unconfined aquifer, both qualitatively and quantitatively. The cDC-GAN approach can be considered as a valuable complement to numerical simulations for predicting multiphase flow behavior in other geoscience-, energy-, and environmental applications such as contaminant transport, hydrocarbon production, and geological carbon dioxide and hydrogen storage.

Rolling bearing fault diagnosis in electric motors based on IDIG-GAN under small sample conditions

Abstract The bearing fault diagnosis based on deep learning algorithms requires a substantial amount of data. However, in practical industrial production, the diagnostic algorithms tend to work ineffectively due to the limitations of samples. Therefore, in this article, we propose an improved method of deep convolutional generative adversarial networks (DCGAN) with discriminator gradient gap regularization (IDIG-GAN), which can effectively solve the problems of unstable training and poor training performance under a small sample dataset. Firstly, the self-attention mechanism is integrated into the DCGAN to capture global information to enhance the generalization capability of the network. Moreover, gradient normalization is applied to the discriminator to address the problem of vanishing gradients in the network. Furthermore, gradient gap regularization is incorporated into the loss function to narrow the gap between the discriminator gradient norms, thereby improving network stability when dealing with small fault datasets. Through training with the improved IDIG-GAN, then the generated samples are used to expand the dataset and construct a fault diagnosis model. By verifying under two bearing datasets, the results demonstrate that the proposed method can generate high-quality samples and effectively enhance the diagnostic capability of the network when working with small datasets.

Specific wavelength peak emulation with amorphous metastructures.

The conventional design process for metasurfaces is time-consuming and computationally expensive. To address this challenge, we utilize a deep convolutional generative adversarial network (DCGAN) to generate new nanohole metastructure designs that match a desired transmittance spectrum in the visible range. The trained DCGAN model demonstrates an exceptional performance in generating diverse and manufacturable metastructure designs that closely resemble the target optical properties. The proposed method provides several advantages over existing approaches. These include its capability to generate new designs without prior knowledge or assumptions regarding the relationship between metastructure geometries and optical properties, its high efficiency, and its generalizability to other types of metamaterials. The successful fabrication and experimental characterization of the predicted metastructures further validate the accuracy and effectiveness of our proposed method.

Efficient high‐fidelity deep convolutional generative adversarial network model for received signal strength reconstruction in indoor environments

AbstractWith the rapid development of wireless communication systems, particularly in the era of 5G and the Internet of Things, deploying wireless communication networks in indoor environments has become crucial. Indoor infrastructure deployment necessitates innovative approaches for efficiently and accurately obtaining received signal strength (RSS) maps. However, traditional methods for acquiring RSS maps, such as empirical and deterministic models, are limited by significant inaccuracies and high computational demands. Empirical models often fail to capture the complex dynamics of indoor environments, resulting in deviations from actual signal behaviours. On the other hand, deterministic models, while more accurate, are computationally intensive due to their reliance on detailed physical modelling of wave propagation. This study introduces a machine learning approach based on deep convolutional generative adversarial networks (DCGAN) aimed at reconstructing indoor RSS maps with minimal RSS measurements. By leveraging DCGAN's generative and adversarial training capabilities, the method not only surpasses traditional interpolation methods in efficiency and precision but also offers new possibilities for the rapid deployment and optimization of wireless communication systems in indoor environments.

In Silico Design of Heterogeneous Microvascular Trees Using Generative Adversarial Networks and Constrained Constructive Optimization.

Designing physiologically adequate microvascular trees is of crucial relevance for bioengineering functional tissues and organs. Yet, currently available methods are poorly suited to replicate the morphological and topological heterogeneity of real microvascular trees because the parameters used to control tree generation are too simplistic to mimic results of the complex angiogenetic and structural adaptation processes invivo. We propose a method to overcome this limitation by integrating a conditional deep convolutional generative adversarial network (cDCGAN) with a local fractal dimension-oriented constrained constructive optimization (LFDO-CCO) strategy. The cDCGAN learns the patterns of real microvascular bifurcations allowing for their artificial replication. The LFDO-CCO strategy connects the generated bifurcations hierarchically to form microvascular trees with a vessel density corresponding to that observed in healthy tissues. The generated artificial microvascular trees are consistent with real microvascular trees regarding characteristics such as fractal dimension, vascular density, and coefficient of variation of diameter, length, and tortuosity. These results support the adoption of the proposed strategy for the generation of artificial microvascular trees in tissue engineering as well as for computational modeling and simulations of microcirculatory physiology.

YOLOX-CS: An Automatic Search Algorithm for Low Surface Brightness Galaxies

The characteristics of Low Surface Brightness Galaxies (LSBGs) are very important for understanding the overall characteristics of galaxies. It is of great significance to search and expand the samples of Low Surface Brightness Galaxies by modern machine learning, especially deep learning algorithm. LSBGs are difficult to discern automatically and accurately with traditional methods because of their obscure features. However, deep learning does have the advantage of automatically identifying complex and effective features. To solve this problem, an algorithm named You Only Look Once version X-CS (YOLOX-CS) is proposed to search LSBG in large sample sky survey. Firstly, five classical target detection algorithms are compared through experiments and the optimal YOLOX algorithm is selected as the basic algorithm. Then, the YOLOX-CS framework is constructed by combining different attention mechanisms and different optimizers. The data set uses images from the Sloan Digital Sky Survey (SDSS), labelled from LSBG in the α.40-SDSS DR7 (the cross coverage area of 40% HI Arecibo Legacy Fast ALFA Survey and SDSS Data Release7) survey. Due to the small number of samples in this data set, Deep Convolutional Generative Adversarial Networks (DCGAN) model is used to expand the experimental test data. After comparing with a series of target detection algorithms, YOLOX-CS has a good test result in searching LSBG recall rate and Average Precision (AP) value in two data sets before and after expansion. The recall rate and AP value in the test set without expansion data set reach 97.75% and 97.83%, respectively. In the expanded data set of DCGAN model, under the same test set, the recall rate reaches 99.10% and the AP value reaches 98.94%, which proves that the algorithm has excellent performance in LSBG search. Finally, the algorithm is applied to SDSS photometric data, and 765 LSBG candidates are obtained.

Enhancing Image Captioning Using Deep Convolutional Generative Adversarial Networks

Introduction:: Introduction: Image caption generation has long been a fundamental challenge in the area of computer vision (CV) and natural language processing (NLP). In this research, we present an innovative approach that harnesses the power of Deep Convolutional Generative Adversarial Networks (DCGAN) and adversarial training to revolutionize the generation of natural and contextually relevant image captions. Method:: Our method significantly improves the fluency, coherence, and contextual relevance of generated captions and showcases the effectiveness of RL reward-based fine-tuning. Through a comprehensive evaluation of COCO datasets, our model demonstrates superior performance over baseline and state-of-the-art methods. On the COCO dataset, our model outperforms current state-of-the-art (SOTA) models across all metrics, achieving BLEU-4 (0.327), METEOR (0.249), Rough (0.525) and CIDEr (1.155) scores. Result:: The integration of DCGAN and adversarial training opens new possibilities in image captioning, with applications spanning from automated content generation to enhanced accessibility solutions. Conclusion:: This research paves the way for more intelligent and context-aware image understanding systems, promising exciting future exploration and innovation prospects.

Generation of virtual mandibular first molar teeth and accuracy analysis using deep convolutional generative adversarial network

Generation of virtual mandibular first molar teeth and accuracy analysis using deep convolutional generative adversarial network

Optimized Wasserstein Deep Convolutional Generative Adversarial Network fostered Groundnut Leaf Disease Identification System

ABSTRACT Groundnut is a noteworthy oilseed crop. Attacks by leaf diseases are one of the most important reasons causing low yield and loss of groundnut plant growth, which will directly diminish the yield and quality. Therefore, an Optimized Wasserstein Deep Convolutional Generative Adversarial Network fostered Groundnut Leaf Disease Identification System (GLDI-WDCGAN-AOA) is proposed in this paper. The pre-processed output is fed to Hesitant Fuzzy Linguistic Bi-objective Clustering (HFL-BOC) for segmentation. By using Wasserstein Deep Convolutional Generative Adversarial Network (WDCGAN), the input leaf images are classified into Healthy leaf, early leaf spot, late leaf spot, nutrition deficiency, and rust. Finally, the weight parameters of WDCGAN are optimized by Aquila Optimization Algorithm (AOA) to achieve high accuracy. The proposed GLDI-WDCGAN-AOA approach provides 23.51%, 22.01%, and 18.65% higher accuracy and 24.78%, 23.24%, and 28.98% lower error rate analysed with existing methods, such as Real-time automated identification and categorization of groundnut leaf disease utilizing hybrid machine learning methods (GLDI-DNN), Online identification of peanut leaf diseases utilizing the data balancing method along deep transfer learning (GLDI-LWCNN), and deep learning-driven method depending on progressive scaling method for the precise categorization of groundnut leaf infections (GLDI-CNN), respectively.

Deep Convolutional Generative Adversarial Networks in Image-Based Android Malware Detection

The recent advancements in generative adversarial networks have showcased their remarkable ability to create images that are indistinguishable from real ones. This has prompted both the academic and industrial communities to tackle the challenge of distinguishing fake images from genuine ones. We introduce a method to assess whether images generated by generative adversarial networks, using a dataset of real-world Android malware applications, can be distinguished from actual images. Our experiments involved two types of deep convolutional generative adversarial networks, and utilize images derived from both static analysis (which does not require running the application) and dynamic analysis (which does require running the application). After generating the images, we trained several supervised machine learning models to determine if these classifiers can differentiate between real and generated malicious applications. Our results indicate that, despite being visually indistinguishable to the human eye, the generated images were correctly identified by a classifier with an F-measure of approximately 0.8. While most generated images were accurately recognized as fake, some were not, leading them to be considered as images produced by real applications.

Policy-making optimization based on generative adversarial networks: A case study of mapping energy transition pathways to China's carbon neutrality

Mapping energy transition pathways is pivotal for achieving carbon neutrality. However, potential pathways delineated by rule-based models differ significantly due to model characteristics, posing a grand challenge in subsequent policy-making. Inspired by the ability of deep convolutional generative adversarial networks (DCGAN) to extract features and generate images, we integrate model outputs concerning 16 energies’ transition pathways to carbon neutrality through DCGAN, assimilating the uncertainties among these outputs. Since DCGAN absorbs the patterns of published data, it offers new insights into policy-making of energy transitions. DCGAN indicates that natural gas and its application with carbon capture and storage play more crucial roles than these currently suggested levels. Additionally, hydrogen and nuclear energies require further development over 2020−2060, serving as a cushion during the substantial energy restructuring. Our study not only provides novel insights into methods mapping the trajectories of critical variables to carbon neutrality but extends DCGAN's application into policy-making optimization.

Generative Model for Dual-Band Filters Based on Modified Complementary Split-Ring Resonators

This article presents a generative model for the inverse design of dual-band filters based on a type of modified complementary split-ring resonator (CSRR). It consists of a series of convolutional neural networks that incorporate the conditional deep convolutional generative adversarial network (GAN) technique. The filters are designed by etching the modified CSRRs on the surface of substrate-integrated waveguides. This design allows us to achieve two passbands with a compact size. In this GAN-based generative model, the CSRRs are represented as two-dimensional matrices. Each matrix corresponds to a training sample of the designed filter, and its S-parameters are extracted through an HFSS simulation. Both the matrices and the S-parameters are fed into the model as the training datasets. Different CSRRs with various sizes are employed for a wider applicable frequency band. Normalized matrices and normalized S-parameters are utilized to simplify the complex generative model resulting from the variations in CSRR sizes. The effectiveness of the generative model is validated through four design examples of dual-band filters, with their center frequencies located within 5 to 18 GHz. The inference time for each design is approximately 18.5 min. The measurement results of the fabricated filters are in good agreement with the simulation ones.

Employing generative adversarial neural networks as surrogate model for reactive transport modeling in the hyporheic zone

This study explores the application of advanced deep learning techniques, specifically conditional deep convolutional generative adversarial networks (cDC-GANs), to model and predict the complex dynamics of the hyporheic zone (HZ) driven by river stage fluctuations. The HZ is a critical transitional mixing zone between surface water and groundwater, where significant biogeochemical reactions occur. Traditional process-based reactive transport modeling methods are computationally intensive and require fine-scale parameterization. To overcome these challenges, we employ the cDC-GAN as a deep-learning-based surrogate model to predict zones of enhanced biogeochemical activity (i.e., hotspots) and reaction product concentrations in the HZ under varying bimodal sedimentary heterogeneity scenarios. The cDC-GAN model can efficiently capture nonlinear relationships between input data and output parameters without extensive parameterization, making it computationally less demanding. The model’s generative capabilities allow for the creation of new data instances, enabling the exploration of diverse scenarios and interpolation between observed data points. The model uses maps of liquid saturation and solute concentrations at different time steps to generate corresponding maps of reaction rates and reaction product concentrations. Results demonstrate the model’s qualitative and quantitative ability to capture complex relationships without relying on rigid assumptions about linearity or specific probability distribution functions. Notably, even amid variations in aquifer heterogeneity, the model consistently exhibits robust performance, validating its ability to adapt to dynamic geological settings. Our use of a simplified reaction scheme serves as a proof of concept for the cDC-GAN model’s ability to simulate solute transport in the HZ, opening avenues for future applications including more complex reaction networks within HZ system such as nutrient cycling, organic matter degradation, redox reactions, and beyond.

Visual Ship Image Synthesis and Classification Framework Based on Attention-DCGAN

To improving ship image generation and classification tasks, a deep convolution generative adversarial network based on attention mechanism (ADCGAN) model was constructed. The rectified linear unit (ReLU) activation function was adopted, and three Deconv layers and Conv layers were added to both the generator and discriminator. Subsequently, an attention mechanism was added to the generator, while spectral normalization (SN) was added to the discriminator. Mean squared error (MSE) was used as loss function to stabilize the training process. Furthermore, ship classification tasks were performed using the generated ship images by end-to-end training of the classification network, enabling ship data augmentation and co-learning with other tasks. Experimental results on the Ship700 and Seaship7000 datasets demonstrate that the ADCGAN model can generate clear and robust ship images, with PSNR, LIPIPS, MS-SSIM values of 20.279 and 27.523, 0.596 and 0.096, 0.781 and 0.947, respectively. The effectiveness of the proposed method in ship image classification tasks was also verified, providing a data foundation for other collaborative tasks.

Assessing the Efficacy of Synthetic Optic Disc Images for Detecting Glaucomatous Optic Neuropathy Using Deep Learning.

Deep learning architectures can automatically learn complex features and patterns associated with glaucomatous optic neuropathy (GON). However, developing robust algorithms requires a large number of data sets. We sought to train an adversarial model for generating high-quality optic disc images from a large, diverse data set and then assessed the performance of models on generated synthetic images for detecting GON. A total of 17,060 (6874 glaucomatous and 10,186 healthy) fundus images were used to train deep convolutional generative adversarial networks (DCGANs) for synthesizing disc images for both classes. We then trained two models to detect GON, one solely on these synthetic images and another on a mixed data set (synthetic and real clinical images). Both the models were externally validated on a data set not used for training. The multiple classification metrics were evaluated with 95% confidence intervals. Models' decision-making processes were assessed using gradient-weighted class activation mapping (Grad-CAM) techniques. Following receiver operating characteristic curve analysis, an optimal cup-to-disc ratio threshold for detecting GON from the training data was found to be 0.619. DCGANs generated high-quality synthetic disc images for healthy and glaucomatous eyes. When trained on a mixed data set, the model's area under the receiver operating characteristic curve attained 99.85% on internal validation and 86.45% on external validation. Grad-CAM saliency maps were primarily centered on the optic nerve head, indicating a more precise and clinically relevant attention area of the fundus image. Although our model performed well on synthetic data, training on a mixed data set demonstrated better performance and generalization. Integrating synthetic and real clinical images can optimize the performance of a deep learning model in glaucoma detection. Optimizing deep learning models for glaucoma detection through integrating DCGAN-generated synthetic and real-world clinical data can be improved and generalized in clinical practice.

A Comparative Analysis of the Novel Conditional Deep Convolutional Neural Network Model, Using Conditional Deep Convolutional Generative Adversarial Network-Generated Synthetic and Augmented Brain Tumor Datasets for Image Classification.

Disease prediction is greatly challenged by the scarcity of datasets and privacy concerns associated with real medical data. An approach that stands out to circumvent this hurdle is the use of synthetic data generated using Generative Adversarial Networks (GANs). GANs can increase data volume while generating synthetic datasets that have no direct link to personal information. This study pioneers the use of GANs to create synthetic datasets and datasets augmented using traditional augmentation techniques for our binary classification task. The primary aim of this research was to evaluate the performance of our novel Conditional Deep Convolutional Neural Network (C-DCNN) model in classifying brain tumors by leveraging these augmented and synthetic datasets. We utilized advanced GAN models, including Conditional Deep Convolutional Generative Adversarial Network (DCGAN), to produce synthetic data that retained essential characteristics of the original datasets while ensuring privacy protection. Our C-DCNN model was trained on both augmented and synthetic datasets, and its performance was benchmarked against state-of-the-art models such as ResNet50, VGG16, VGG19, and InceptionV3. The evaluation metrics demonstrated that our C-DCNN model achieved accuracy, precision, recall, and F1 scores of 99% on both synthetic and augmented images, outperforming the comparative models. The findings of this study highlight the potential of using GAN-generated synthetic data in enhancing the training of machine learning models for medical image classification, particularly in scenarios with limited data available. This approach not only improves model accuracy but also addresses privacy concerns, making it a viable solution for real-world clinical applications in disease prediction and diagnosis.