Layers Of Convolutional Neural Network Research Articles

River plume identification through a deep-learning model: an innovative approach

ABSTRACT River plumes are complex physical phenomena that occur at the interface between riverine and marine systems. The lack of direct measurements of sediment concentration complicates the study of these geomorphological features, since the boundaries of river plumes are often gradual and unclear. Therefore, the identification and digitalization of river plumes is not simple and the methods applied in different study areas are not always objective and replicable. The aim of this work is to provide a valid approach based on a deep-learning model that uses Convolution Neural Network (CNN) layers for the digitalization of river plumes. We describe the methodology applied to implement the input dataset used for training the model, the errors obtained, and an application for a study area of about 300 km located in the Mediterranean. The model uses Sentinel-2 Level-1C images. The application of the model to a specific study area allowed us to understand the possibility of investigating these geomorphological features to obtain results in agreement with previous works. As a matter of fact, by using the red band as a proxy of sediment concentration, we were able to investigate the average behaviours of sediment dispersion along the coast and to extract innovative data related to specific events for the study of morphological characteristics such as dimension and direction.

An Algorithmic Study of Transformer-Based Road Scene Segmentation in Autonomous Driving

Applications such as autonomous driving require high-precision semantic image segmentation technology to identify and understand the content of each pixel in the images. Compared with traditional deep convolutional neural networks, the Transformer model is based on pure attention mechanisms, without convolutional layers or recurrent neural network layers. In this paper, we propose a new network structure called SwinLab, which is an improvement upon the Swin Transformer. Experimental results demonstrate that the improved SwinLab model achieves a segmentation accuracy comparable to that of deep convolutional neural network models in applications such as autonomous driving, with an MIoU of 77.61. Additionally, comparative experiments on the CityScapes dataset further validate the effectiveness and generalization of this structure. In conclusion, by refining the Swin Transformer, this paper simplifies the model structure, improves the training and inference speed, and maintains high accuracy, providing a more reliable semantic image segmentation solution for applications such as autonomous driving.

Advanced Mineral Deposit Mapping via Deep Learning and SVM Integration With Remote Sensing Imaging Data

ABSTRACTAutomating mineral delineation and rock type analysis using remote sensing imaging data is a critical application of machine learning. Traditional machine learning methods often struggle with accuracy and precise map generation. This study aims to enhance performance through a refined deep learning model. In this work, we present a deep learning pipeline to map the mineral deposits in the study area. Initially, we apply a deep convolutional neural network (CNN) to a specialized mineral dataset to map mineral deposits within the study area. Subsequently, we build a hybrid model combining deep CNN layers with a support vector machine (SVM). This merger significantly improves classification accuracy from an initial 92.7% to 95.3%. In our approach, CNN layers function as feature extractors while the SVM serves as the classification model. Moreover, we conduct an evaluation of the SVM using polynomial kernels of degrees 3, 6, 9, and 12. The results indicate that the SVM with a degree of 12 achieved the highest classification accuracy, followed by degrees 9, 6, and 3. Experimental results demonstrate the effectiveness of our proposed method for classifying remote sensing imaging data, showcasing its potential for advancing mineral delineation and rock type analysis.

Real-life boxing activity recognition with smartphones using attention assisted deep learning models

Mobile computing technologies play a pivotal role in narrowing the gap between athletes and sports monitoring systems, while also advancing towards intelligent and automatic supervision of individual sports activities. The widespread presence of smartphones equipped with multifunctional sensors has enabled the acquisition and analysis of data, facilitating Human Activity Recognition (HAR). Although recognising sports activities using a smart mobile phone’s accelerometer is widely used, traditional methods struggle with intricate and real time activities due to the high-dimensional sensor data. This study introduces a smartphone-based architecture for HAR utilising inertial accelerometers to record athlete’s game actions data sequences, extract relevant features and derive the player’s activity data through multiple three-axis accelerometers. In this paper an Attention assisted Deep Convolution Neural Network based Bi-LSTM (AT-DCNN-BL) model is proposed. The raw data undergoes sliding window pre-processing technique, the deep learning framework of deep convolution neural network layers helps in automatic feature extraction, the bi-LSTM structure process the boxing activity data and the attention is focussed over the boxing activity classification data. The attention mechanism redistributes the weights of the extracted features which aids in increasing the classification accuracy. Other classification models like Deep Convolutional Neural Network (DCNN), Bi-directional Long Short-Term Memory (bi-LSTM), Multi-Layer Perceptron Neural Network (MLPNN) and Random Forest (RF) are also applied to the dataset collected from a mobile sensor. The study explores training of deep learning methods and illustrates the superiority of the proposed approach over others using the collected mobile sensor dataset. The AT-DCNN-BL model achieved a higher classification accuracy compared to the other models. The AT-DCNN-BL model outperformed all the other models achieving a higher accuracy rate (92.71%).

Retraction notice: COVID-19 prediction through X-ray images using various layers in convolutional neural network

Retraction notice: COVID-19 prediction through X-ray images using various layers in convolutional neural network

Imperial Study on Convolutional Layer in CNN for Data Mining

Convolutional Neural Networks (CNNs) have become a cornerstone in the field of data mining, particularly for tasks involving large-scale image and pattern recognition. This paper presents an imperial study on the efficacy of convolutional layers within CNN architectures in data mining applications. We analyze the impact of varying convolutional layer configurations, such as kernel size, stride, and depth, on performance metrics including accuracy, computational cost, and feature extraction quality. By conducting a series of experiments across different datasets, we explore how these configurations influence the model's ability to generalize and detect complex patterns in structured and unstructured data. Our findings indicate that optimizing the convolutional layers significantly improves the efficiency of CNNs for data mining tasks, offering insights into best practices for architectural design in such applications. This study provides valuable guidelines for leveraging CNNs in the realm of data mining, pushing the boundaries of automated data analysis and knowledge discovery

Multimodal biometric identification: leveraging convolutional neural network (CNN) architectures and fusion techniques with fingerprint and finger vein data

Advancements in multimodal biometrics, which integrate multiple biometric traits, promise to enhance the accuracy and robustness of identification systems. This study focuses on improving multimodal biometric identification by using fingerprint and finger vein images as the primary traits. We utilized the “NUPT-FPV” dataset, which contains a substantial number of finger vein and fingerprint images, which significantly aided our research. Convolutional neural networks (CNNs), renowned for their efficacy in computer vision tasks, are used in our model to extract distinct discriminative features. Specifically, we incorporate three popular CNN architectures: ResNet, VGGNet, and DenseNet. We explore three fusion strategies used in security applications: early fusion, late fusion, and score-level fusion. Early fusion integrates raw images at the input layer of a single CNN, combining information at the initial stages. Late fusion, in contrast, merges features after individual learning from each CNN model. Score-level fusion employs weighted aggregation to combine scores from each modality, leveraging the complementary information they provide. We also use contrast limited adaptive histogram equalization (CLAHE) to enhance fingerprint contrast and vein pattern features, improving feature visibility and extraction. Our evaluation metrics include accuracy, equal error rate (EER), and ROC curves. The fusion of CNN architectures and enhancement methods shows promising performance in identifying multimodal biometrics, aiming to increase identification accuracy. The proposed model offers a reliable authentication system using multiple biometrics to verify identity.

Inverse feedforward control of piezoelectric actuators using optimized composite neural network-based Hammerstein model

This paper utilizes the optimized composite neural network (OCNN)-Hammerstein model to directly identify the dynamic inverse hysteresis effect, employing it as a feedforward controller to compensate for the rate-dependent and amplitude-dependent dynamic hysteresis of the piezoelectric actuator. The OCNN-Hammerstein model consists of an optimized composite neural network and an auto-regressive exogenous model in series. The OCNN comprises convolutional neural network layer and radial basis function neural network layer, with its hyperparameters optimized using a modified grey wolf optimizer algorithm. Compared to the existing dynamic Prandtl-Ishlinskii-Hammerstein and dynamic Bouc-Wen-Hammerstein models in the literature, the feedforward controller based on the proposed model in this paper demonstrates superior performance, with an RSME below 0.45 μm and a 74.5% reduction in relative hysteresis at the condition of 300 Hz and maximum amplitude. The feedforward controller exhibits universality across all amplitude ranges and within the 50–300 Hz frequency range, while also providing predictive compensation for dynamic hysteresis at 300–400 Hz. The feedforward controller based on the OCNN-Hammerstein model in this paper provides a crucial methodology for open-loop control of piezoelectric actuators.

A graph convolutional neural network model based on fused multi-subgraph as input and fused feature information as output

The graph convolution neural network (GCN)-based node classification model tackles the challenge of classifying nodes in graph data through learned feature representations. However, most existing graph neural networks primarily focus on the same type of edges, which might not accurately reflect the intricate real-world graph structure. This paper introduces a novel graph neural network model, MF-GCN, which integrates subgraphs with various edge types as input and combines feature information from each graph convolutional neural network layer to produce the final output. This model learns node feature representations by separately feeding subgraphs with different edge types into the graph convolutional layer. It then computes the weight vectors for fusing various edge type subgraphs based on the learned node features. Additionally, to efficiently extract feature information, the outputs of each graph convolution layer, without an activation function, are weighted and summed to obtain the final node features. This approach resolves the challenges of determining fusion weights and effectively extracting feature information during subgraph fusion. Experimental results show that the proposed model significantly improves performance on all three datasets, highlighting its effectiveness in node representation learning tasks.

Numerical investigation of the effective receptive field and its relationship with convolutional kernels and layers in convolutional neural network

The receptive field (RF) plays a crucial role in convolutional neural networks (CNNs) because it determines the amount of input information that each neuron in a CNN can perceive, which directly affects the feature extraction ability. As the number of convolutional layers in CNNs increases, there is a decay of the RF according to the two-dimensional Gaussian distribution. Thus, an effective receptive field (ERF) can be used to characterize the available part of the RF. The ERF is calculated by the kernel size and layer number within the neural network architecture. Currently, ERF calculation methods are typically applied to single-channel input data that are both independent and identically distributed. However, such methods may result in a loss of effective information if they are applied to more general (i.e., multi-channel) datasets. Therefore, we proposed a multi-channel ERF calculation method. By conducting a series of numerical experiments, we determined the relationship between the ERF and the convolutional kernel size in conjunction with the layer number. To validate the new method, we used the recently published global wave surrogate model for climate simulation (GWSM4C) and its accompanying dataset. According to the newly established relationship, we refined the kernel size and layer number in each neural network of the GWSM4C to produce the same ERF but lower RF attenuation rates than those of the original version. By visualizing the gradient map at several points in West African and East Pacific areas, the high gradient value regions confirmed the known swell sources, which indicated effective feature extraction in these areas. Furthermore, the new version of the GWSM4C yielded better prediction accuracy for significant wave height in global swell pools. The root mean square errors in the West African and East Pacific regions reduced from approximately 0.3 m, in the original model to about 0.15 m, in the new model. Moreover, these improvements were attributed to the higher efficiency of the newly modified neural network structure that allows the inclusion of more historical winds while maintaining acceptable computational consumption.

Accurate and 30-plus days reliable cuffless blood pressure measurements with 9-minutes personal photoplethysmograph data and mixed deduction learning

Cuffless blood pressure (BP) measurements have long been anticipated, and the PPG (Photoplethysmography)-only method is the most promising one since already embedded in many wearable devices. To further meet the clinical accuracy requirements, PPG-only BP predictions with personalized modeling for overcoming personal deviations have been widely studied, but all required tens to hundreds of minutes of personal PPG measurements for training. Moreover, their accurate test periods without calibration practice were not reported. In this work, we collected records of PPG data from our recruited subjects in real-life scenarios instead of relying on the openly available MIMIC dataset obtained from intensive care unit (ICU) patients. Since our objective is commercial application and a substantial reduction in training data, we tailored our model training to closely mimic real-world usage. To achieve this, we developed a training approach that only requires 9-minutes of personal PPG signal recordings and mixed with other PPG data from our recruited 364 subjects. The modeling is conducted with two-channel paired inputs to the convolutional neural network (CNN)-based model, which we called Mixed Deduction Learning (MDL). The test results of 88 samples from 15 subjects, under testing period up to 30-plus days without extra calibration, revealed that MDL meets most of the standards of AAMI, BHS, and IEEE 1708–2014 (for static test only) for BP measurement devices, which indicates MDL’s long-term stability and consistency. Furthermore, we found that the model with two-channel inputs presents a trend of improving performance as the pool of mixed training data increased, while the conventional one-channel input revealed degraded performance. The outperformance of MDL is attributed to many significant features remained in the first CNN layer even when mixing personal 9-minutes data with the other 364 subjects. Consequently, PPG-only with MDL introduces a new avenue for overcoming challenges in training due to personal physiological variations. Given our consideration of real-life usage, this technology can be seamlessly translated to commercial applications.

Nested transformer decoder using dense skip-connections for change detection in high-resolution remote sensing image

ABSTRACT Change detection is a critical task in remote sensing image analysis, with applications in land management and urban development. Convolutional neural networks (CNNs) have shown success in this area, but existing CNN-based methods often struggle with irrelevant changes caused by interfering factors, leading to unsatisfactory results. To address this, some studies have replaced CNN layers with transformers to enhance the receptive field and utilize global information better. However, this can make models overly reliant on global information, potentially overlooking shallow, fine-grained features, especially at the edges of changed targets. To overcome these limitations, we propose a Nested Transformer Decoder (NTD) that combines the strengths of transformers and CNNs by introducing dense skip connections. While having a large receptive field, it can also better retain shallow and fine-grained features by adding skip-connection. These innovations enable our model to more accurately identify changes and preserve the edges of the changed areas, including small changes that are difficult to detect using existing methods. Additionally, we introduce a Residual Fusion Module (RFM) to reduce false detections caused by irrelevant factors like light and shadow. The RFM employs residual information and spatiotemporal attention to suppress differences in multi-scale features, leading to better detection outcomes. The experimental results demonstrate that our proposed model outperforms the current state-of-the-art change detection methods in various metrics, and represents a novel and effective solution to the challenge of change detection in remote sensing images. Experimental results show that our model surpasses state-of-the-art change detection methods across various metrics. Tested on three public change detection datasets, our method improves the Intersection over Union (IoU) for the change class by 2.83% to 16.39% and enhances the F1-score by 1.60% to 9.63%. These results demonstrate that the proposed approach provides an effective solution to the challenges of change detection in remote sensing images.

Design of model fusion learning method based on deep bidirectional GRU neural network in fault diagnosis of industrial processes

This paper proposes an end-to-end model fusion feature learning method based on deep bidirectional gated recurrent unit (MCNN-DBiGRU) for fault diagnosis in industrial processes. First, a feature-aligned multi-scale feature extraction model (MCNN) is designed by analyzing the working principles of convolutional and pooling layers of convolutional neural networks. Secondly, a deep bidirectional mechanism is proposed to better extract the time series features in the process data. This mechanism makes the recurrent neural network not only present the forward processing input features from the past to the future, but also the reverse processing from the future to the past. By integrating these features, the diagnostic performance of the network model is improved. To verify that the proposed model has effective diagnostic accuracy for fault diagnosis, we conduct simulation experiments on the Tennessee-Eastman (TE) process and a chemical coking furnace, and compare with several conventional network models. In the end, not only the effectiveness of the model is proved, but it is also confirmed that the model is superior to other conventional neural networks in both diagnostic accuracy and feature robustness.

Industrial robot energy consumption model identification: A coupling model-driven and data-driven paradigm

Due to wide distribution and low energy efficiency, the energy-saving in industrial robots (IRs) is attracting extensive attention. Accurate energy consumption (EC) models of IRs lay the foundation for energy-saving. However, most dynamic and electrical parameters of IRs are not disclosed by manufacturers, which leads to the invalidity of most model-based EC prediction methods. To bridge this gap, a mechanism-data hybrid-driven method is proposed to predict the EC of IRs in this paper. First, a joint torque prediction model integrating a hybrid-driven parameter identification is developed based on deep reinforcement learning (DRL). The framework for DRL-based parameter identification is constructed through tailored design of interfaces and training mechanisms, wherein the DRL agent can learn to identify the dynamic parameters from the trajectory database. And a deep neural network based on long short-term memory (LSTM) is proposed to predict the EC of IRs according to the joint torques and velocities. The nonlinear item, which is not modeled in the robot dynamic equation, are also encapsulated in the deep neural network with one-dimensional convolutional neural network (1D-CNN) layers to improve the prediction accuracy. To validate the accuracy and efficacy of the proposed method, experiments are conducted on a KUKA KR60-3 industrial robot with different loads. The results demonstrate that the proposed method can predict EC with a mean absolute percentage error of less than 2% under a fixed load and less than 3% under loads not used for agent training.

Evaluating the Impact of Convolutional Neural Network Layer Depth on the Enhancement of Inertial Navigation System Solutions

Secure navigation is pivotal for several applications including autonomous vehicles, robotics, and aviation. The inertial navigation system estimates position, velocity, and attitude through dead reckoning especially when external references like GPS are unavailable. However, the three accelerometers and three gyroscopes that compose the system are exposed to various types of errors including bias errors, scale factor errors, and noise, which can significantly degrade the accuracy of navigation constituting also a key vulnerability of this system. This work aims to adopt a supervised convolutional neural network (ConvNet) to address this vulnerability inherent in inertial navigation systems. In addition to this, this paper evaluates the impact of the ConvNet layer's depth on the accuracy of these corrections. This evaluation aims to determine the optimal layer configuration maximizing the effectiveness of error correction in INS (Inertial Navigation System) leading to precise navigation solutions.

Urban Spatiotemporal Event Prediction Using Convolutional Neural Network and Road Feature Fusion Network

The security challenges faced by smart cities are attracting more attention from more people. Criminal activities and disasters can have a significant impact on the stability of a city, resulting in a loss of safety and property for its residents. Therefore, predicting the occurrence of urban events in advance is of utmost importance. However, current methods fail to consider the impact of road information on the distribution of cases and the fusion of information at different scales. In order to solve the above problems, an urban spatiotemporal event prediction method based on a convolutional neural network (CNN) and road feature fusion network (FFN) named CNN-rFFN is proposed in this paper. The method is divided into two stages: The first stage constructs feature map and structure of CNN then selects the optimal feature map and number of CNN layers. The second stage extracts urban road network information using multiscale convolution and incorporates the extracted road network feature information into the CNN. Some comparison experiments are conducted on the 2018–2019 urban patrol events dataset in Zhengzhou City, China. The CNN-rFFN method has an R2 value of 0.9430, which is higher than the CNN, CNN-LSTM, Dilated-CNN, ResNet, and ST-ResNet algorithms. The experimental results demonstrate that the CNN-rFFN method has better performance than other methods.

A fault detection of aero-engine rolling bearings based on CNN-BiLSTM network integrated cross-attention

Abstract Aero-engine rolling bearings are essential for engine health, in which disruptive failures can be prevented and reduce great losses in air flight. To improve the efficiency of fault detection, an improved network, named CNN- BiLSTM -Cross-Attention (CBLCA) was proposed. The Bidirectional Long Short-Term Memory (BiLSTM) layer captures the temporal features as the input data. The cross-attention mechanism is integrated with the Convolutional Neural Networks (CNN) layer and the BiLSTM layer respectively. More important feature information can be identified with the CBLCA model. The proposed model was also validated with the open-sourced aero-engine rolling bearings data set. To improve the identification accuracy, a novel method that combines Fast Fourier Transform (FFT) and Variational Mode Decomposition (VMD) is used for the data preprocessing. Each original signal sample is transformed into a feature set containing richer information, and the number of features significantly increased in the entire dataset. Compared with some existing LSTM models, such as LSTM, BiLSTM, CNN-BiLSTM, and CNN-LSTM, the classification accuracy was increased by 55%, 54%, 5%, and 7%, respectively. The processing method for vibration signals and the CBLCA model can improve the accuracy and reliability of fault diagnosis for aero-engine rolling bearings.

Layer-Wise Mutual Information Meta-Learning Network for Few-Shot Segmentation.

The goal of few-shot segmentation (FSS) is to segment unlabeled images belonging to previously unseen classes using only a limited number of labeled images. The main objective is to transfer label information effectively from support images to query images. In this study, we introduce a novel meta-learning framework called layer-wise mutual information (LayerMI), which enhances the propagation of label information by maximizing the mutual information (MI) between support and query features at each layer. Our approach involves the utilization of a LayerMI Block based on information-theoretic co-clustering. This block performs online co-clustering on the joint probability distribution obtained from each layer, generating a target-specific attention map. The LayerMI Block can be seamlessly integrated into the meta-learning framework and applied to all convolutional neural network (CNN) layers without altering the training objectives. Notably, the LayerMI Block not only maximizes MI between support and query features but also facilitates internal clustering within the image. Extensive experiments demonstrate that LayerMI significantly enhances the performance of baseline and achieves competitive performance compared to state-of-the-art methods on three challenging benchmarks: PASCAL-5 i , COCO-20 i , and FSS-1000.

Factorizers for distributed sparse block codes

Distributed sparse block codes (SBCs) exhibit compact representations for encoding and manipulating symbolic data structures using fixed-width vectors. One major challenge however is to disentangle, or factorize, the distributed representation of data structures into their constituent elements without having to search through all possible combinations. This factorization becomes more challenging when SBCs vectors are noisy due to perceptual uncertainty and approximations made by modern neural networks to generate the query SBCs vectors. To address these challenges, we first propose a fast and highly accurate method for factorizing a more flexible and hence generalized form of SBCs, dubbed GSBCs. Our iterative factorizer introduces a threshold-based nonlinear activation, conditional random sampling, and an ℓ ∞ -based similarity metric. Its random sampling mechanism, in combination with the search in superposition, allows us to analytically determine the expected number of decoding iterations, which matches the empirical observations up to the GSBC’s bundling capacity. Secondly, the proposed factorizer maintains a high accuracy when queried by noisy product vectors generated using deep convolutional neural networks (CNNs). This facilitates its application in replacing the large fully connected layer (FCL) in CNNs, whereby C trainable class vectors, or attribute combinations, can be implicitly represented by our factorizer having F-factor codebooks, each with C F fixed codevectors. We provide a methodology to flexibly integrate our factorizer in the classification layer of CNNs with a novel loss function. With this integration, the convolutional layers can generate a noisy product vector that our factorizer can still decode, whereby the decoded factors can have different interpretations based on downstream tasks. We demonstrate the feasibility of our method on four deep CNN architectures over CIFAR-100, ImageNet-1K, and RAVEN datasets. In all use cases, the number of parameters and operations are notably reduced compared to the FCL.

Optimized twin spatio-temporal convolutional neural network with squeezenet transfer learning model for cancer miRNA biomarker classification

Cancer MicroRNA (miRNA) biomarkers are small RNA molecules identified in cancer cells that indicate the presence or progression of cancer. miRNA is detected through circulating (liquid biopsy) and tissue-based approaches. Handling the data involves sequencing, followed by bioinformatic analysis to interpret the miRNA profiles. The valuable knowledge is extracted from the cancer genomic datasets using data mining and machine learning techniques, although the accuracy of certain classes has been unsatisfied. To address this issue, an optimized twin spatio-temporal convolutional neural network with SqueezeNet transfer learning model for cancer miRNA biomarker classification (TSTCNN-SNet-AGT-CBC) is proposed. The input imageries are gathered from Cancer Genome Atlas dataset. The noise is reduced and image quality is improved in preprocessing by applying Reliable asynchronous sampled-data filtering (RASDF). Then, features are extracted and the cancer miRNA is effectively classified as a diagnosis, therapy, and prognostic normal region using twin spatio-temporal convolutional neural network with SqueezeNet. The batch normalization layer of the twin Spatio-Temporal Convolutional Neural Network (TSTCNN) is eliminated and added with Squeeze Net Layer (SNet). Then, the Artificial Gorilla Troops Optimization (AGTO) is used for tuning the TSTCNN-SNet optimum weight parameter. The performance of proposed technique is examined under performance metrics like precision, sensitivity, F-scores and accuracy. The proposed TSTCNN-SNet-AGT-CBC method attains 24.52 %, 20.97 % and 16.36 % higher accuracy and 23.68 %, 18.77 %, and 41.36 % greater precision when compared with existing techniques.