Deep Learning Network Research Articles

A Survey on Detecting Deep Fakes Using Advanced AI-Based Approaches

Today, artificial intelligence is used to clone human faces, which leads to a new technology known as “deepfakes.” Recently, machine learning (ML) approaches and the use of deep learning (DL) networks have captured researchers' competition to achieve the highest classification accuracy in building efficient models for digital content deepfake detection. Therefore, this review analyzes and compares existing deepfake detection methods based on advanced artificial intelligence algorithms. Thus, deepfake detection techniques were classified into three major categories based on the classifier model used (machine learning, deep learning, or hybrid) and then compared to show the aspects that influence the efficiency and accuracy of the algorithms. This research helps researchers develop efficient classification models for deepfake detection applications. Based on the survey information reviewed in this study, a discussion of open issues and future directions is presented. The most important challenges and research directions related to deepfake detection methods are discussed.

Power data analysis and mining technology in smart grid

This study proposes a smart grid model named “GridOptiPredict”, which aims to achieve efficient analysis and processing of power system data through deep fusion of deep learning and graph neural network, so as to improve the intelligent level and overall efficiency of power grid operation. The model integrates three core functions of load forecasting, power grid state sensing and resource optimization into one, forming a closely connected and complementary framework. Through carefully designed experimental scheme, the practical value and effectiveness of “Grid OptiPredict” model are fully verified from three aspects: accuracy of load forecasting, sensitivity of power grid state sensing and efficiency of resource allocation strategy. Experimental results show that the model has significant advantages in prediction accuracy, model stability and robustness, resource optimization, security, information security, social and economic benefits and user experience.

LIDAR3DNETV2: Smart Approach to Classify 3D Objects Based on Real Point Clouds

Point clouds generated in simulators avoid collection time costs and provide a high and organized amount of point clouds, an ideal scenario for deep learning networks. However, these networks have limitations when applied to real point clouds. This work proposes a multilayer perceptron-based method to classify 3D objects based on real point clouds obtained using LiDAR sensors. The method includes a pre-processing step that normalizes and adjusts the point clouds in the 3D Cartesian plane to overcome discrepancies in the point distribution. Furthermore, we created a dataset and used the ModelNet dataset for comparison purposes. The proposed neural network, Lidar3DNetV2, achieved 98.47% and 125 μs in accuracy and test time with real data, respectively. The pre-processing step provided a significant increase in the classifier’s performance. Finally, the proposed method performs better than other state-of-the-art networks considering real point clouds.

A Survey on Classifying Ocular Diseases Using Deep Learning and Machine Learning Techniques

In ophthalmology, using fundus images to identify ocular diseases early may pose challenges for clinicians. Manually diagnosing ocular conditions is time-consuming, difficult, and requires experimentation. As a result, technology was created to help computers differentiate between ocular diseases. It is possible to create a system of this type due to different learning algorithms based on visual capabilities. Recent breakthroughs in deep learning and machine learning have led to the development of intelligent systems that improve accuracy and efficiency in classifying eye diseases. The purpose of this study is to conduct a comprehensive survey of modern systems that classify ocular problems using different methods, including pre-trained deep learning networks, using the Ocular Disease Intelligent Recognition (ODIR) dataset. The goal is to build and train a model that can recognize and classify ocular disorders. Previous research indicates the increasing use of deep learning techniques. CNN-based methods have spread widely in this field, compared to traditional manual procedures, due to their outstanding results. The most prominent deep learning techniques are convolutional neural networks (CNNs), recurrent neural networks (RNNs), and various learning methods for increasing and transferring data. The survey highlights the potential of these systems to enhance classification accuracy and sensitivity while addressing challenges such as data availability, interpretability, and integration with clinical practice.

Efficient Training Data Acquisition Technique for Deep Learning Networks in Radar Applications

In the field of radar, deep learning techniques have shown considerably superior performance over traditional classifiers in detecting and classifying targets. However, acquiring sufficient training data for deep learning applications is often challenging and time consuming. In this study, we propose a technique for acquiring training data efficiently using a combination of synthesized data and measured background data. We utilized graphics processing unit (GPU)-based physical optics methods to obtain the backscattered field of moving targets. We then generated a virtual dataset by mixing the synthesized target signal with the background signal real data. Subsequently, we trained a convolutional neural network using the virtual dataset to identify three different classes—Bird, Drone, and Background—from a range-Doppler map. When tested using the measurement data, the trained model achieved an accuracy of over 90%, demonstrating the effectiveness of the proposed method in acquiring training data for radar-based deep learning applications.

MULTI-LEVEL CLASSIFICATION BASED ON DEEP LEARNING FOR ACCURATE RISK STRATIFICATION OF ARRHYTHMIAS

Arrhythmias, also known as irregular heartbeats, are important health problems that must be accurately identified to diagnose and treat cardiovascular disease. Within the scope of this study, a network for classifying arrhythmias, which are important in the diagnosis and treatment of cardiovascular diseases, was proposed by using one-dimensional convolutional neural network (1D CNN), one of the deep learning techniques. With the proposed 1D-CNN architecture, arrhythmia types and normal rhythm ECGs were subjected to a more detailed examination from general to specific according to urgency situations. In the classifications made, first of all, a binary classification was made and an evaluation was made as whether there was a life risk or not. In triple, quadruple and six-fold classification, the detection of arrhythmia status is detailed. More complex classifications have helped to define different types of arrhythmias in more detail. This study proposes a deep learning network for automatic identification and classification of arrhythmias and shows that different arrhythmia conditions can be diagnosed with a single network model by applying the proposed network structure to multi-class arrhythmia disorders.

Plant Species Classification and Biodiversity Estimation from UAV Images with Deep Learning

Biodiversity is a characteristic of ecosystems that plays a crucial role in the study of their evolution, and to estimate it, the species of all plants need to be determined. In this study, we used Unmanned Aerial Vehicles to gather RGB images of mid-to-high-altitude ecosystems in the Zao mountains (Japan). All the data-collection missions took place in autumn so the plants present distinctive seasonal coloration. Patches from single trees and bushes were manually extracted from the collected orthomosaics. Subsequently, Deep Learning image-classification networks were used to automatically determine the species of each tree or bush and estimate biodiversity. Both Convolutional Neural Networks (CNNs) and Transformer-based models were considered (ResNet, RegNet, ConvNeXt, and SwinTransformer). To measure and estimate biodiversity, we relied on the Gini–Simpson Index, the Shannon–Wiener Index, and Species Richness. We present two separate scenarios for evaluating the readiness of the technology for practical use: the first scenario uses a subset of the data with five species and a testing set that has a very similar percentage of each species to those present in the training set. The models studied reach very high performances with over 99 Accuracy and 98 F1 Score (the harmonic mean of Precision and Recall) for image classification and biodiversity estimates under 1% error. The second scenario uses the full dataset with nine species and large variations in class balance between the training and testing datasets, which is often the case in practical use situations. The results in this case remained fairly high for Accuracy at 90.64% but dropped to 51.77% for F1 Score. The relatively low F1 Score value is partly due to a small number of misclassifications having a disproportionate impact in the final measure, but still, the large difference between the Accuracy and F1 Score highlights the complexity of finely evaluating the classification results of Deep Learning Networks. Even in this very challenging scenario, the biodiversity estimation remained with relatively small (6–14%) errors for the most detailed indices, showcasing the readiness of the technology for practical use.

Direction of Arrival Estimation Based on DNN and CNN

The accuracy of Direction of Arrival (DOA) estimation primarily depends on the precision of the data. When the receiver uses a low-precision analog-to-digital converter (ADC), traditional DOA estimation algorithms exhibit poor accuracy. To face the challenge of multi-target DOA estimation in scenarios with low-precision ADC quantized sampling, this paper proposes a novel DOA estimation algorithm for quantized signals based on classification problems. A deep learning network was constructed using Deep Neural Networks (DNNs) and Convolutional Neural Networks (CNNs), divided into the quantized signal recovery framework and the DOA estimation framework. The DNN network is utilized to recover signals that have undergone low-precision quantization, while the CNN network addresses the classification problem to estimate the DOA from received data with an unknown number of signal sources. A comprehensive analysis of the impact of signal-to-noise ratio (SNR), the number of array elements, and the number of quantization bits on the proposed algorithm was conducted. Simulation results indicate that the proposed algorithm exhibits superior DOA estimation performance in low-precision scenarios, characterized by reduced computational complexity, thereby facilitating real-time DOA estimation.

Automated High-Precision Recognition of Solar Filaments Based on an Improved U2-Net

Solar filaments are a significant solar activity phenomenon, typically observed in full-disk solar observations in the H-alpha band. They are closely associated with the magnetic fields of solar active regions, solar flare eruptions, and coronal mass ejections. With the increasing volume of observational data, the automated high-precision recognition of solar filaments using deep learning is crucial. In this study, we processed full-disk H-alpha solar images captured by the Chinese H-alpha Solar Explorer in 2023 to generate labels for solar filaments. The preprocessing steps included limb-darkening removal, grayscale transformation, K-means clustering, particle erosion, multiple closing operations, and hole filling. The dataset containing solar filament labels is constructed for deep learning. We developed the Attention U2-Net neural network for deep learning on the solar dataset by introducing an attention mechanism into U2-Net. In the results, Attention U2-Net achieved an average Accuracy of 0.9987, an average Precision of 0.8221, an average Recall of 0.8469, an average IoU of 0.7139, and an average F1-score of 0.8323 on the solar filament test set, showing significant improvements compared to other U-net variants.

Adaptive Hybrid Deep Learning Based Effective Channel Estimation in MIMO-Noma for Millimeter-Wave Systems with an Enhanced Optimization Algorithm

The channel estimation is crucial in the “millimeter Wave (mmWave) Massive Multiple-Input MultipleOutput (MIMO) and Non-Orthogonal Multiple Access (NOMA)” devices. Hybrid beamforming techniques are employed nowadays to minimize the complexity and equipment price. However, the absence of digital beam forming in mmWave affects the dynamic range and accuracy of the channel estimation. Previous research has concentrated mainly on predicting narrow-band mmWave channels using deep learning networks, as the wideband channels of mmWave create a considerable range and noise issue. Accurate channel estimation in the MIMO system is challenging because of the increased number of antennas and radio frequency (RF) chains. MIMO system communications using mmWave are frequently chosen because of their massive spectrum resources. Therefore, it is essential to tackle the obstacles obtained in the standard channel estimation framework by developing a MIMO-NOMA network with the help of deep learning methods Hence, this paper proposes an efficient hybrid deep learning model for channel estimation in MIMO-NOMA for mmWave systems. At first, the channel estimation is carried out using the Adaptive Hybrid Deep Learning (AHDL) model, where it combines both Autoencoder and Recurrent Neural Network (RNN). Here, the parameters are optimized using the Improved Red-tailed Hawk Algorithm (IRHA). Later the hardware cost and system complexity are reduced by performing the hybrid beam-forming process. Numerical results show that the proposed channel estimation and pilot estimation process outperforms the state-of-the-art approaches.

Artificial Intelligence and its Applications in the Field of Internet of Things (Iot)

Artificial Intelligence (AI) and the Internet of Things (IoT), are innovative technologies that have revolutionized many fields independently. Altogether, this convergence offers a daunting possibility for innovation of a number of aspects of smartness in a range of domains on the one hand and the optimization of those aspects for sustainability on the other hand. Machine learning and deep learning help IoT devices and networks perform analytics on big data and achieve real-time decision making and automation. In smart cities AI based IoT helps in the self-organizing and self-optimizing of traffic management, and utilization of energy, and increases public safety by applying predictive analytics to the actual management of a city. The telemedicine receiving aid from AI IoT includes the detecting of sickness onset with a device that monitors patients and creates corresponding customized treatment regimens. The same way applied to IIoT where AI is implemented for predictive maintenance, quality assurance, inventory control and in turn, the achievement of alleviated downtime and improved production. Ethical issues related to bias in algorithms, and subsequent decision making via the use of Artificial Intelligence add another layer of concern in terms of deployment and utilization. Prospects for the future include such trends as edge computing, the AIoT is AI + IoT amalgamation, and the disruptive effect of quantum computing on AI applications for the IoT. The mitigation of these difficulties and the exploitation of these opportunities will open the path to proper optimizations of AI-based IoT systems as concerns effectiveness, security and/or ethics. This article stresses how AI is mandatory to the implementation of an ideal IoT system by exploring current uses, future developments, and problems that are vital for the advancement of IoT.

A Deep–Learning Network for Wheat Yield Prediction Combining Weather Forecasts and Remote Sensing Data

Accurately predicting winter wheat yield before harvest could greatly benefit decision-makers when making management decisions. In this study, we utilized weather forecast (WF) data combined with Sentinel-2 data to establish the deep-learning network and achieved an in-season county-scale wheat yield prediction in China’s main wheat-producing areas. We tested a combination of short-term WF data from the China Meteorological Administration to predict in-season yield at different forecast lengths. The results showed that explicitly incorporating WF data can improve the accuracy in crop yield predictions [Root Mean Square Error (RMSE) = 0.517 t/ha] compared to using only remote sensing data (RMSE = 0.624 t/ha). After comparing a series of WF data with different time series lengths, we found that adding 25 days of WF data can achieve the highest yield prediction accuracy. Specifically, the highest accuracy (RMSE = 0.496 t/ha) is achieved when predictions are made on Day of The Year (DOY) 215 (40 days before harvest). Our study established a deep-learning model which can be used for early yield prediction at the county level, and we have proved that weather forecast data can also be applied in data-driven deep-learning yield prediction tasks.

Electric load forecasting under false data injection attacks via denoising deep learning and generative adversarial networks

AbstractAccurate electric load forecasting at various time periods is considered a necessary challenge for electricity consumers and generators to maximize their economic efficiency in energy markets. Hence, the accuracy and effectiveness of existing electric load forecasting approaches depends on the data quality. Nowadays, with the implementation of modern power systems and Internet of Things technology, forecasting models are faced with a large volume of data, which puts the security and health of data at risk due to the use of numerous measuring devices and the threat of cyber‐attackers. In this study, a cyber‐resilient hybrid deep learning‐based model is developed that accurately forecasts electric load in short‐term and long‐term time horizons. The architecture of the proposed model systematically integrates stacked multilayer denoising autoencoder (SMDAE) and generative adversarial network (GAN) and is called SMDAE‐GAN. In the proposed framework, SMDAE layer is used to pre‐process and remove real fs and intentional anomalies in data, and GAN layer is also utilized to forecast electric load values. The effectiveness of the SMDAE‐GAN structure is studied using realistic electrical load data monitored in the distribution network of Tabriz, Iran, and meteorological data measured in weather station there. Compared with other conventional load forecasting approaches, the developed framework has the highest accuracy in both cases of using normal data with real‐world noise and damaged data under false data injection attacks.

ASLAM-FD: a multi-sensor adaptive collaborative fusion SLAM framework based on degradation detection and deep reinforcement learning

Abstract Aiming at the difficulty in ensuring accuracy and performance of existing Simultaneous Localization And Mapping (SLAM) technology based on multi-sensor fusion in complex dynamic environments, a multi-sensor adaptive fusion SLAM framework based on degradation detection and deep reinforcement learning (ASLAM-FD) is proposed. This framework can achieve adaptive collaborative precise adjustment of fusion weights (FWs) based on the real-time self-degradation states and relative degradation states quantified by each sensor, and can adapt to different tightly coupled SLAM algorithms based on FWs. In addition, within this framework, the continuous quantification models for the degradation states of internal/external sensors with certain versatility (we refer to them as EX-DM and IN-DM) is proposed. These quantitative models can achieve continuous quantification of degradation states of various mainstream internal/external sensors. Based on the above sensor degradation state quantification models, this paper further proposes a deep reinforcement learning (DRL) network suitable for adaptive collaborative adjustment of FWs. This network focuses more on the temporal nature of sensor observation data and degradation states, making it more suitable for modeling relationships between data with temporal features. In the experimental section, we adapted the proposed ASLAM-FD to different multi-sensor fusion SLAM algorithms on multiple datasets, and compared it with multiple advanced fusion SLAM algorithms. We found that adapting ASLAM-FD can effectively improve the accuracy and performance of fusion SLAM in complex dynamic environments.

Identifying defects and varieties of Malting Barley Kernels

This study introduces a comprehensive approach for classifying individual malting barley kernels, involving dual-sided kernel imaging, a specifically designed image processing algorithm, an optimized deep neural network architecture, and a mechanical sorting system. The proposed method achieves precise classification into multiple classes, aligning with quality standards for malting material assessment. Throughout the study, various image analysis techniques were assessed, including traditional feature engineering, established transfer learning deep neural network architectures, and our custom-designed convolutional neural network tailored for barley kernel image analysis. Comparative analysis underscores the superior performance of our network model. The study reveals that our proposed deep learning network achieves a 94% accuracy in classifying barley kernel defects and varieties, outperforming well-established transfer learning models to complex architectures that attain 93% accuracy. Additionally, it surpasses the traditional machine learning approach involving feature extraction and support vector machine classifiers, which achieve accuracy below 90% in detecting defective kernels and below 70% in varietal classification. However, we also noted the traditional approach’s advantage in morphological feature recognition. This observation guides new research toward integrating morphological feature extraction techniques with modern convolutional networks. This paper presents a deep neural network designed specifically for the analysis of cereal kernel images in two applications: defect and variety classification. It emphasizes the importance of standardizing kernel orientation and merging images from both sides of the kernel, and introduces a device for image acquisition that fulfills this need.

DB-Net: dual-branch deep learning network for cloud detection utilizing attention mechanism and cascaded multiscale feature fusion

DB-Net: dual-branch deep learning network for cloud detection utilizing attention mechanism and cascaded multiscale feature fusion

Detecting emerald ash borer boring vibrations using an encoder-decoder and improved DenseNet model.

Forest ecosystems are under constant threat from wood-boring pests such as the Emerald ash borer (EAB), which remain elusive owing to their hidden life cycles within tree trunks. Early detection is vital to mitigate economic and ecological damage. The main current monitoring method is manual detection which is ineffective at early stages of infestation. This study introduces VibroEABNet, a deep learning-based joint recognition network designed to enhance the detection of EAB boring vibration signals, with a novel approach integrating denoising and recognition modules. The proposed VibroEABNet model demonstrated exceptional performance, achieving an average accuracy of 98.98% across multiple signal-to-noise ratios (SNRs) in test datasets and a remarkable 97.5% accuracy in real forest datasets, surpassing traditional models and other deep learning networks evaluated in this study. These findings were supported by rigorous noise resistance analysis and real dataset evaluation, indicating the model's robustness and reliability in practical applications. Furthermore, the model's efficiency was highlighted by its inference time of 26 ms and a compact model size of 8.43 MB, underscoring its suitability for deployment in resource-limited environments. The development of VibroEABNet marks a significant advancement in pest detection methodologies, offering a scalable, accurate and efficient solution for early monitoring of wood-boring pests. The integration of a denoising module within the network structure addresses the challenge of environmental noise, one of the primary limitations in acoustic monitoring of pests. Currently, this research is limited to a specific pest. Future work will focus on the applicability of this network to other wood-boring pests. © 2024 Society of Chemical Industry.

Efficient damage prediction and sensitivity analysis in rectangular welded plates subjected to repeated blast loads utilizing deep learning networks

Efficient damage prediction and sensitivity analysis in rectangular welded plates subjected to repeated blast loads utilizing deep learning networks

PSbBO-Net: A Hybrid Particle Swarm and Bayesian Optimization-based DenseNet for Lung Cancer Detection using Histopathological and CT Images

Lung cancer remains a substantial global fatality; early detection is imperative for successful intervention and treatment. Deep learning (DL) models have shown promise in predicting lung cancer from medical images, but optimizing their parameters remains a challenging task. To improve prediction capability, this study introduces an approach by merging Particle Swarm Optimization and Bayesian Optimization (PSbBO) to optimize deep learning parameters. PSO provides an effective way for exploring the hyperparameter space, while Bayesian optimization provides a probabilistic framework for the effective evaluation and refining of a DL network. The simulation study showcases the effectiveness of the proposed model, achieving notable metrics for histopathological images, including an accuracy of 99.5%, precision of 98.3%, recall of 99.2%, F1-score of 99.4%, and an error rate of 1.19%. Furthermore, when applied to lung CT images, the proposed PSbBO demonstrates an accuracy of 98.8%, precision of 97.4%, recall of 98.3%, F1-score of 98.6%, and an error rate of 1.21%.

ResNet-101 based anomaly detection in additive manufacturing: Thermal modeling for quality control in heat exchanger production

In the production of heat exchangers using additive manufacturing, maintaining consistent quality is a significant challenge due to irregular energy transformations during the manufacturing process. These irregularities can result in faulty products, leading to issues such as reduced efficiency and reliability of the heat exchangers. Traditional quality control methods are often inadequate for real-time detection and correction of such faults. This paper focuses on analyzing the occurrence of faulty products caused by irregular energy transformations in heat exchangers. A deep learning network, specifically ResNet-101, is employed to address this issue. The network is trained with characteristic features of the product and instantaneous variations in the heat exchanger to accurately identify faulty patterns resulting from degradation in the heat exchanger’s performance. By leveraging the deep learning network, which comprehensively understands both the product and heat exchanger features, the system can detect and classify faulty products in real time. This enables immediate corrective actions, such as pushing defective products into a recycling process, thereby maintaining high-quality standards in continuous production.