Convolutional Neural Network Network Research Articles

CNN-Based Cancer Image Diagnosis: Current Progress and Future Directions

Abstract. With the development of deep learning technology, CNN (Convolutional Neural Network) models have shown great value in medical image analysis, especially in diagnosing early lung cancer, breast cancer, and brain tumors. In this study, we recall and organize the application and progress of CNN models in the field of cancer and tumor diagnosis in the past five years to provide a theoretical basis and reference for related researchers. This article introduces the principles of different CNN cancer diagnostic models and compares and analyzes their results, and ultimately finds that these models have significant advantages in improving the accuracy and efficiency of cancer diagnosis, but at the same time, there are also problems such as too much reliance on large datasets, high model complexity, and poor generalization ability. In the future, we can consider optimizing the performance of CNN network models, enhancing the generalization ability of the models, and developing data enhancement techniques on this basis, so that CNN models can be better applied in the field of cancer diagnosis.

Recognizing human activities using light-weight and effective machine learning methodologies

Background Human activity recognition (HAR) is increasingly important in enhancing healthcare systems by enabling accurate monitoring of individuals' movements through sensor data. This paper is motivated by the need to improve the accuracy of HAR, particularly for applications in e-health systems, where reliable activity detection can lead to better health outcomes. The study explores six prominent machine learning techniques—decision tree, random forest, linear regression, Naïve Bayes, k-nearest neighbor, and neural networks—to determine which methods can most effectively predict activities like walking, sitting, standing, laying, walking upstairs, and walking downstairs. Methods We employed these six machine learning algorithms to analyze a comprehensive dataset derived from various sensors. Each model was rigorously trained and evaluated to compare its effectiveness in recognizing human activities. The experiments aimed to identify strengths and weaknesses in each approach, with particular emphasis on advanced techniques such as random forest, convolutional neural networks (CNNs), and gated recurrent networks (GRNs). Results The experimental evaluation revealed that the random forest classifier, CNN, GRN, and neural networks delivered promising results, achieving high accuracy levels. Notably, the neural network model excelled, attaining an impressive accuracy of 98%. In contrast, the Naïve Bayes model did not meet the performance expectations set by the other algorithms. Conclusions This research effectively classifies activities such as sitting, standing, laying, walking, walking downstairs, and walking upstairs, underscoring the potential of machine learning in HAR. The findings highlight the superior performance of neural networks in enhancing activity recognition, which could lead to advanced applications in e-health systems and improve overall healthcare monitoring strategies.

Design of Chinese painting style classification model based on multi-layer aggregation CNN

This study delves deeply into exploring the artistic value of traditional Chinese painting (TCP) and aims to bridge the gap between its fundamental characteristics and the realm of human emotions. To achieve this, a novel convolutional neural network (CNN)-based classification model for TCP emotions is proposed. By thoroughly analyzing the distinct emotional mapping relationships inherent in TCP, a comprehensive framework is developed. Notably, emotional feature regions are accurately extracted using image saliency, and a multi-layer aggregation recalibrated emotional feature module is seamlessly integrated into the CNN network structure. This integration strengthens the activation of features that significantly influence TCP emotions. Moreover, a method of multi-category weighted activation localization is skillfully employed to classify Chinese painting emotions within the CMYK (cyan magenta yellow black) color space. The empirical results convincingly demonstrate that our algorithm surpasses existing approaches such as VGG-19, GoogLeNet, ResNet-50, and WSCNet in the task of TCP emotion recognition, achieving an impressive accuracy of 92.36%, the largest error is 0.191. This improvement signifies the advancements made by our model in accurately capturing and understanding the emotional nuances within TCP. By outperforming previous methods, our research contributes to the convergence of multimedia technology and cultural education.

ADT: Person re-identification based on efficient attention mechanism and single-channel dual-channel fusion with transformer features aggregation

In person re-identification(Re-ID) tasks, using only convolutional neural network(CNNs) makes it challenging to fully exploit the global key information in person images, while using only the Transformer structure fails to adequately capture the multi-scale local detail information in person images. Therefore, to address the more challenging person Re-ID tasks, effectively integrating CNN and Transformer into an algorithmic framework capable of fusing multi-dimensional features has become one of the current focal points of research. Due to the ability of the attention mechanism to suppress the extraction of non-salient features by convolutional neural networks, and the capability of the dual-channel multi-scale mechanism to facilitate the extraction of finer-grained deep information by the Transformer, we propose a network Based on Efficient Attention Mechanism and Single-channel Dual-channel Fusion with Transformer Features Aggregation (ADT). First, we introduce the Convolutional Block Attention Module(CBAM) and Spatial Reconstruction Unit (SRU) into the CNN network to construct a Multi-dimensional Attention Features Re-fusion Network(MCF). By weighting the local features extracted in the convolutional network, attention to person detail features in the model is enhanced. Then, we propose the Single and Dual-channel Integrated Transformer Features Aggregator(SDT) and overlap it in MCF. This integration is designed to more effectively aggregate the multi-scale high-dimensional global features from each layer of the CNN network. Experimental results on the Market1501, DukeMTMC, MSMT17, and CUHK03 datasets demonstrate that our model significantly enriches features representation compared to most methods.

Deep learning methods for modeling infrasound transmission loss in the middle atmosphere

Accurate modeling of infrasound transmission losses (TLs) is essential to assess the performance of the global International Monitoring System (IMS) infrasound network. Among existing propagation modeling tools, parabolic equation method (PE) enables TLs to be finely modeled, but its computational cost does not allow exploration of a large parameter space for operational monitoring applications. To reduce computation times, Brissaud et al. (2022) explored the potential of convolutional neural networks (CNNs) trained on a large set of regionally simulated wavefields (>1000 km from the source) to predict TLs with negligible computation times compared to PE simulations. However, this new method shows difficulties in upwind conditions, especially at low frequencies, and causal issues with winds at large distances from the source affecting ground TLs close to the source. In this study, we have developed an optimized CNN network designed to minimize prediction errors while predicting TLs from globally simulated combined temperature and wind fields spanning over propagation ranges of 4000 km. Our approach enhances the previously proposed one by implementing key optimizations that improve the overall architecture performances. The implemented model predicts TLs with an average error of 20 dB in the whole frequency band (0.1-4 Hz) and explored realistic atmospheric scenarios.

Hyperspectral Image Denoising: From Model-Driven, Data-Driven, to Model-Data-Driven.

Mixed noise pollution in HSI severely disturbs subsequent interpretations and applications. In this technical review, we first give the noise analysis in different noisy HSIs and conclude crucial points for programming HSI denoising algorithms. Then, a general HSI restoration model is formulated for optimization. Later, we comprehensively review existing HSI denoising methods, from model-driven strategy (nonlocal mean, total variation, sparse representation, low-rank matrix approximation, and low-rank tensor factorization), data-driven strategy [2-D convolutional neural network (CNN), 3-D CNN, hybrid, and unsupervised networks], to model-data-driven strategy. The advantages and disadvantages of each strategy for HSI denoising are summarized and contrasted. Behind this, we present an evaluation of the HSI denoising methods for various noisy HSIs in simulated and real experiments. The classification results of denoised HSIs and execution efficiency are depicted through these HSI denoising methods. Finally, prospects of future HSI denoising methods are listed in this technical review to guide the ongoing road for HSI denoising. The HSI denoising dataset could be found at https://qzhang95.github.io.

Enhanced Malaysian License Plate Recognition System Using an Improved YOLOv2 Model

License Plate Recognition (LPR) has gained popularity among researchers due to its wide range of applications, including law enforcement, monitoring, and toll gate systems. However, existing LPR systems still require improvements to achieve optimum accuracy and speed. The advancements in Convolutional Neural Network (CNN) variants offer potential solutions for these challenges. This primary aim of this system is to ensure accurate and efficient recognition of the vehicle plate characters using CNN techniques. This research utilizes two CNN network architectures for deep object detection to address the Malaysian License Plate Recognition (MLPR) task. The first network is designed to detect the license plate, while the second is responsible for recognizing the characters on the plate. Both networks are cascaded from the architecture of two-stage YOLOv2, providing promising speed and accuracy. The MLPR system achieved an accuracy of 98.75% and a processing speed of 0.0104 seconds, using a total of 2,200 license plate images. In conclusion, the system adapted from deep object detection techniques presents a promising solution for the MLPR problem, based on the achieved accuracy and speed.

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.

IoT-Driven Approaches for Early Detection and Monitoring of Heart Disease: Current Trends and Future Directions

Cardiovascular disease (CVD) is a serious disease having a widespread effect on individuals across the world. Early and accurate detection of cardiac disease is crucial in healthcare, especially in the domain of cardiology. Currently, a non-invasive ultrasound imaging method is used that evaluates the structure, performance, and blood, allowing for the precise identification of a number of cardiac ailments, such as valve problems, heart failure, and congenital anomalies. These traditional techniques have some limitations, including high cost, the need for medical expertise and equipment, and the fact that they often create incorrect results due to human involvement. Furthermore, the traditional method takes more time to predict heart disease. Electrocardiogram (ECG) signals play a critical role in reducing death rates caused by CVDs, and they provide details regarding the heart patient’s health to a medical expert by employing an automated heart failure detection system. Recent developments in deep learning-based health care systems, such as ECG signal analysis, include CNN, LSTM, and other neural networks. In this research, we provide a hybrid deep learning based approach for the timely and accurate diagnosis of cardiovascular disease. The proposed system uses a hybrid of convolutional neural networks (CNN) and long short-term memory (LSTM) and utilizes the MIT-BIH ECG signal dataset for heart disease diagnosis. This study uses two different approaches with MIT-BIH arrhythmia imbalanced and balanced datasets. The first approach uses CNN and CNN-LSTM with an imbalanced dataset, and the second approach uses CNN and CNN-LSTM with a balanced dataset. The performance of both approaches was analyzed. The experimental outcomes show that the overall performance of both CNN, CNN-LSTM was excellent on a balanced dataset compared to imbalanced dataset. The proposed system achieved a better result than the previous suggested methods. Additionally, it is easy to adopt the suggested technique in the field of healthcare in order to identify heart disease.

A 1D-inception-ResNet based global detection model for thin-skinned multifruit spectral quantitative analysis

Near-infrared spectroscopy (NIRS) analysis technology has important research and application value in the inspection of agricultural product quality due to its nondestructive characteristics. Deep learning methods represented by convolutional neural networks (CNNs) reduce the reliance on expert experience and prior knowledge in spectral analysis. They can automatically extract effective feature information from the original spectral data to improve the prediction accuracy of the model. However, the diverse characteristics of fruit varieties present a challenge to the robustness of the model. This study proposes1D-Inception-ResNet, an end-to-end CNN model for multifruit spectral quantitative analysis. The effectiveness of the model is validated on two thin-skinned fruit datasets with similar physicochemical properties. The root mean square error of prediction (RMSEP) for soluble solids content (SSC) and dry matter (DM) were 0.48 °Brix and 0.60 %, and the coefficients of determination of prediction (Rp2) were 0.95 and 0.89, respectively. Compared with traditional partial least squares regression (Rp2 of 0.86 and 0.80, RMSEP of 0.84 °Brix and 0.81 %, respectively) and extreme learning machine (Rp2 of 0.82 and 0.77, RMSEP of 0.95 °Brix and 0.86 %, respectively), our proposed method resulted in a lower RMSEP (46.1% and 28.1% lower, respectively) and higher RP2 (13.1% and 13.4% higher, respectively) for SSC and DM prediction. The results show that compared to other linear and nonlinear algorithms, 1D-Inception-ResNet significantly enhances the modeling effectiveness. Furthermore, external validation using a new sample set showed RMSEP values of 0.62 °Brix and 1.45 % for the SSC and DM models of Inception-ResNet, respectively. In conclusion, using an appropriate CNN network structure to establish a global model for diverse fruits expands the application scope of the model and improves the prediction stability across different species, which is significant in the future application of agricultural products and foods.

Vector to matrix representation for CNN networks for classifying astronomical data

Choosing the right classifier is crucial for effective classification in various astronomical datasets aimed at pattern recognition. While the literature offers numerous solutions, the support vector machine (SVM) continues to be a preferred choice across many scientific fields due to its user-friendliness. In this study, we introduce a novel approach using convolutional neural networks (CNNs) as an alternative to SVMs. CNNs excel at handling image data, which is arranged in a grid pattern. Our research explores converting one-dimensional vector data into two-dimensional matrices so that CNNs pre-trained on large image datasets can be applied. We evaluate different methods to input data into standard CNNs by using two-dimensional feature vector formats. In this work, we propose a new method of data restructuring based on a set of wavelet transforms. The robustness of our approach is tested across two benchmark datasets/problems: brown dwarf identification and threshold crossing event (Kepler data) classification. The proposed ensembles produce promising results on both datasets. The MATLAB code of the proposed ensemble is available at https://github.com/LorisNanni/Vector-to-matrix-representation-for-CNN-networks-for-classifying-astronomical-data

YOLO-HyperVision: A vision transformer backbone-based enhancement of YOLOv5 for detection of dynamic traffic information

With the increase of traffic flow in modern urban areas, traffic congestion has become a serious problem that affects people’s normal production and life. Using target detection technology instead of manual labor can quickly detect the road traffic situation and provide timely information about the traffic flow. However, when using drones to observe the traffic flow in the air, the perspective effect will cause the detected vehicles and pedestrians to be very small, and the scale difference between different categories of targets is large, which increases the detection difficulty of a single convolutional neural network model. In order to solve the problem of low accuracy of traditional single-stage target detection models, this study proposes an improved Yolov5 vehicle target detection model with Vision Transformer (VIT) backbone, You Only Look Once-HyperVision (YOLO-HV), which aims to solve the problem of poor multi-scale target recognition performance caused by the inability of traditional CNN networks to integrate contextual information, and help drones achieve more efficient and accurate traffic flow recognition functions. This study deeply integrates the Vision Transformer (VIT) backbone and Convolutional Neural Network (CNN), effectively combining the multi-scale detection advantages of Vision Transformer and the inductive bias ability of Convolutional Neural Network, and adds multi-scale residual modules and context correlation enhancement modules, which greatly improves the recognition accuracy of single-stage detectors for drone images. Through comparative experiments on the VisDrone dataset, it is found that the detection performance of this model is improved compared with several commonly used detection models. YOLO-HV can increase the mean average precision (mAP) by 3.3% compared with the pure convolutional network of the same model size. YOLO-HV model has achieved excellent performance in the task of traffic flow image detection taken by drones, and can more accurately identify and classify road vehicles than various target detection models.

Building detection based on searching of the optimal kernel shapes pruning method on Res2-Unet

<p>In recent years, advances in remote sensing technology have made it feasible to use satellite data for large-scale building detection. Moreover, the building detection from multispectral satellite photography data is necessary; however, it is difficult to recovery the accurate building footprint from the high-resolution pictures. Because the deep learning networks contains high computational cost and over-parameterized. Therefore, network pruning has been used to reduce the storage and computations of convolutional neural network (CNN) models. In this article, we proposed the pruning technique to prune the CNN network from Res2-Unet model for accurately detecting the buildings. Initially, the CNN network is pruned by utilizing the searching of the optimal kernel shapes technique. It is employed to carry out stripe-wise pruning and automatically find the ideal kernel shapes. Then the data quantification is applied to enhance the proposed model and also reduce the complexity. Finally, the enhanced Res2-Unet model is used for the building detection. Moreover, WHU East Asia Satellite and the Massachusetts building dataset are the two available datasets used to access the suggested framework. Compare to the existing models, the proposed model gives better performance.</p>

Neuro-XAI: Explainable deep learning framework based on deeplabV3+ and bayesian optimization for segmentation and classification of brain tumor in MRI scans

The prevalence of brain tumor disorders is currently a global issue. In general, radiography, which includes a large number of images, is an efficient method for diagnosing these life-threatening disorders. The biggest issue in this area is that it takes a radiologist a long time and is physically strenuous to look at all the images. As a result, research into developing systems based on machine learning to assist radiologists in diagnosis continues to rise daily. Convolutional neural networks (CNNs), one type of deep learning approach, have been pivotal in achieving state-of-the-art results in several medical imaging applications, including the identification of brain tumors. CNN hyperparameters are typically set manually for segmentation and classification, which might take a while and increase the chance of using suboptimal hyperparameters for both tasks. Bayesian optimization is a useful method for updating the deep CNN's optimal hyperparameters. The CNN network, however, can be considered a "black box" model because of how difficult it is to comprehend the information it stores because of its complexity. Therefore, this problem can be solved by using Explainable Artificial Intelligence (XAI) tools, which provide doctors with a realistic explanation of CNN's assessments. Implementation of deep learning-based systems in real-time diagnosis is still rare. One of the causes could be that these methods don't quantify the Uncertainty in the predictions, which could undermine trust in the AI-based diagnosis of diseases. To be used in real-time medical diagnosis, CNN-based models must be realistic and appealing, and uncertainty needs to be evaluated. So, a novel three-phase strategy is proposed for segmenting and classifying brain tumors. Segmentation of brain tumors using the DeeplabV3+ model is first performed with tuning of hyperparameters using Bayesian optimization. For classification, features from state-of-the-art deep learning models Darknet53 and mobilenetv2 are extracted and fed to SVM for classification, and hyperparameters of SVM are also optimized using a Bayesian approach. The second step is to understand whatever portion of the images CNN uses for feature extraction using XAI algorithms. Using confusion entropy, the Uncertainty of the Bayesian optimized classifier is finally quantified. Based on a Bayesian-optimized deep learning framework, the experimental findings demonstrate that the proposed method outperforms earlier techniques, achieving a 97 % classification accuracy and a 0.98 global accuracy.

A new multi-objective hyperparameter optimization algorithm for COVID-19 detection from x-ray images

The coronavirus occurred in Wuhan (China) first and it was declared a global pandemic. To detect coronavirus X-ray images can be used. Convolutional neural networks (CNNs) are used commonly to detect illness from images. There can be lots of different alternative deep CNN models or architectures. To find the best architecture, hyper-parameter optimization can be used. In this study, the problem is modeled as a multi-objective optimization (MOO) problem. Objective functions are multi-class cross entropy, error ratio, and complexity of the CNN network. For the best solutions to the objective functions, multi-objective hyper-parameter optimization is made by NSGA-III, NSGA-II, R-NSGA-II, SMS-EMOA, MOEA/D, and proposed Swarm Genetic Algorithms (SGA). SGA is a swarm-based algorithm with a cross-over process. All six algorithms are run and give Pareto optimal solution sets. When the figures obtained from the algorithms are analyzed and algorithm hypervolume values are compared, SGA outperforms the NSGA-III, NSGA-II, R-NSGA-II, SMS-EMOA, and MOEA/D algorithms. It can be concluded that SGA is better than others for multi-objective hyper-parameter optimization algorithms for COVID-19 detection from X-ray images. Also, a sensitivity analysis has been made to understand the effect of the number of the parameters of CNN on model success.

A comparative study of breast tumour detection using a semantic segmentation network coupled with different pretrained CNNs

ABSTRACT Breast cancer is one of the most prevalent malignancies and the primary origin of cancer-related deaths among females worldwide. Ultrasound image segmentation plays a crucial role in identifying breast tumours by precisely delineating the boundaries of the tumour within the images. Deep learning segmentation networks such as DeepLabV3+ have been used in the literature for this purpose. The coupling of DeepLabV3+ with base convolutional neural networks (CNN) can play a key role in its accuracy in tumour detection. The present study investigates the segmentation performance of four combinations of DeepLabV3+ segmentation network by coupling with the four base decoder CNN networks: DarkNet53, SqueezeNet, EfficientNet-b0 and DarkNet19. The accuracy of segmentation is confirmed using standard segmentation error metrics such as global and mean accuracy, mean and weighted Intersection over union (IoU), Boundary F1 (BF) Score and Dice Score. DeepLabV3+ coupled with DarkNet53 and EfficientNet-b0 as the decoder CNNs performed better than the other two combinations with a global accuracy of 96.50% and 96.18%. Precise tumour delineation can assist in tumour growth monitoring and treatment planning by providing detailed information on tumour size, shape, and location; thereby aiding in patient management and follow-up care.

Pedagogical sentiment analysis based on the BERT-CNN-BiGRU-attention model in the context of intercultural communication barriers.

Amid the wave of globalization, the phenomenon of cultural amalgamation has surged in frequency, bringing to the fore the heightened prominence of challenges inherent in cross-cultural communication. To address these challenges, contemporary research has shifted its focus to human-computer dialogue. Especially in the educational paradigm of human-computer dialogue, analysing emotion recognition in user dialogues is particularly important. Accurately identify and understand users' emotional tendencies and the efficiency and experience of human-computer interaction and play. This study aims to improve the capability of language emotion recognition in human-computer dialogue. It proposes a hybrid model (BCBA) based on bidirectional encoder representations from transformers (BERT), convolutional neural networks (CNN), bidirectional gated recurrent units (BiGRU), and the attention mechanism. This model leverages the BERT model to extract semantic and syntactic features from the text. Simultaneously, it integrates CNN and BiGRU networks to delve deeper into textual features, enhancing the model's proficiency in nuanced sentiment recognition. Furthermore, by introducing the attention mechanism, the model can assign different weights to words based on their emotional tendencies. This enables it to prioritize words with discernible emotional inclinations for more precise sentiment analysis. The BCBA model has achieved remarkable results in emotion recognition and classification tasks through experimental validation on two datasets. The model has significantly improved both accuracy and F1 scores, with an average accuracy of 0.84 and an average F1 score of 0.8. The confusion matrix analysis reveals a minimal classification error rate for this model. Additionally, as the number of iterations increases, the model's recall rate stabilizes at approximately 0.7. This accomplishment demonstrates the model's robust capabilities in semantic understanding and sentiment analysis and showcases its advantages in handling emotional characteristics in language expressions within a cross-cultural context. The BCBA model proposed in this study provides effective technical support for emotion recognition in human-computer dialogue, which is of great significance for building more intelligent and user-friendly human-computer interaction systems. In the future, we will continue to optimize the model's structure, improve its capability in handling complex emotions and cross-lingual emotion recognition, and explore applying the model to more practical scenarios to further promote the development and application of human-computer dialogue technology.

Deep-modified transfer learning-based CNN networks for enhanced breast cancer prediction

Breast cancer (BC) is one of the most fatal forms of cancer, making it a significant contributor to mortality rates worldwide. Early detection and timely treatment of breast cancer are crucial in reducing its mortality rate. To ensure a healthy lifestyle, it is essential to develop systems that can accurately diagnose breast cancer. Recent advances in modern computing and information technologies have enabled significant progress in the early detection and prediction of diseases within healthcare systems. This study proposes a method for precise and automatic breast cancer prediction using deep-modified transfer learning-based Convolutional Neural Networks (CNNs). The CNN architectures employed include ResNet50, MobileNetV2, DenseNet121, and Xception, which serve as feature extractors to capture the most relevant features of breast Ultrasound images (BUSI). These extracted features are then accurately classified as benign or malignant using various high-performance classifiers, including Support Vector Machine (SVM), K-Nearest Neighbors (KNN), XGBoost, and Softmax. The experimental results demonstrate that the proposed deep modified DenseNet121 network with the Softmax classifier outperformed other models and existing techniques. This latter achieved remarkable performance metrics, including an accuracy of 95.34%, a precision of 90.90%, and an F1 score of 93.02%. These results highlight the effectiveness of our approach in enhancing the accuracy of breast cancer prediction. The superior performance of the proposed method provides significant improvements in decision-making speed and reduces the time, effort, and laboratory resources required for healthcare services. Consequently, this method has the potential to significantly enhance early diagnosis and enable more tailored treatment plans, ultimately contributing to better patient outcomes and reducing the overall mortality rates associated with breast cancer.

Cross-correlation-based convolutional neural network with velocity regularization for high-resolution velocimetry of particle images

Particle image velocimetry (PIV) stands as a pivotal experimental technique in fluid dynamics, enabling the visualization and analysis of fluid flows. Traditional methods for extracting velocity fields from particle images often rely on window-cross correlation PIV or, more recently, optical flow techniques rooted in intensity conservation principles. However, the former approach suffers from low resolution, whereas the latter is hampered by computational inefficiency and a high susceptibility to noise. Recent studies have demonstrated the effectiveness of convolutional neural networks (CNNs) in processing particle images to obtain high-resolution and high-accuracy velocity fields, though traditional CNN architectures are still not satisfying in accuracy. The present study introduces an enhanced network, En-FlowNetC, based on the cross correlation-based CNN FlowNetC, specifically designed to process PIV particle images and achieve high-accuracy, high-resolution velocity fields. It incorporates a velocity regularization and is trained and validated on canonical datasets. The results indicate that En-FlowNetC surpasses traditional CNN networks in accuracy and markedly outperforms the classic Horn–Schunck optical flow method in both complex and simple flow scenarios. Furthermore, this study confirms the beneficial impact of velocity regularization, when judiciously applied, on network accuracy. The proposed modifications compared to the original FlowNetC are also examined in the ablation experiments. Overall, En-FlowNetC provides an effective deep-learning solution for PIV analysis, paving the way for future endeavors aimed at achieving highly accurate and resolved velocimetry.

Motor imagery electroencephalography channel selection based on deep learning: A shallow convolutional neural network

Electroencephalography (EEG) motor imagery (MI) signals have recently attracted much attention because of their potential to communicate with the surrounding environment in a specific way without the need for muscular and physical movement. Despite these advantages, these signals are difficult to detect due to their low signal to noise (SNR) rate and non-stationary and dynamic nature. Convolutional neural networks (CNN) can extract appropriate spatial and temporal features without the need for separate feature extraction and classification steps. Clean input data for CNN significantly improves its performance, but sometimes, common interference occurs between multi-channel signals. This challenge, along with the noisy EEG signals, degrades the performance of CNN networks. This paper presents a channel selection method based on convolutional neural networks to obviate these challenges. The proposed shallow convolutional neural network (SCNN) is designed with temporal convolution and pointwise convolution (using 1×1 convolution) to select the best channel with minimal computational load. Following the channel selection step, the multi-layer fusion CNN model was used to classify the signals. Choosing suitable and relevant EEG channels for motor imagery tasks makes it possible to create appropriate inputs for the classification model so that the model can be improved in the end. For High Gamma Dataset and Brian-Computer Interface (BCI) Competition IV-2a, the following accuracies were obtained: 81.15% and 72.01%, which are higher than when the other channel selections such as Mutual Information, Sequential Feature Forward Selection (SFFS) and wrapper methods are used.