Convolutional Neural Network Framework Research Articles

Automated dentition segmentation: 3D UNet-based approach with MIScnn framework

Advancements in technology have led to the adoption of digital workflows in dentistry, which require the segmentation of regions of interest from cone-beam computed tomography (CBCT) scans. These segmentations assist in diagnosis, treatment planning, and research. However, manual segmentation is an expensive and labor-intensive process. Therefore, automated methods, such as convolutional neural networks (CNNs), provide a more efficient way to generate segmentations from CBCT scans. A three-dimensional UNet-based CNN model, utilizing the Medical Image Segmentation CNN framework, was used for training and generating predictions from CBCT scans. A dataset of 351 CBCT scans, with ground-truth labels created through manual segmentation using AI-assisted segmentation software, was prepared. Data preprocessing, augmentation, and model training were performed, and the performance of the proposed CNN model was analyzed. The CNN model achieved high accuracy in segmenting maxillary and mandibular teeth from CBCT scans, with average Dice Similarity Coefficient values of 91.83% and 91.35% for maxillary and mandibular teeth, respectively. Performance metrics, including Intersection over Union, precision, and recall, further confirmed the model's effectiveness. The study demonstrates the efficacy of the three-dimensional UNet-based CNN model within the Medical Image Segmentation CNN framework for automated segmentation of maxillary and mandibular dentition from CBCT scans. Automated segmentation using CNNs has the potential to deliver accurate and efficient results, offering a significant advantage over traditional segmentation methods.

DsRNAPredictor-II: An improved predictor of identifying dsRNA and its silencing efficiency for Tribolium castaneum based on sequence length distribution

RNA interference (RNAi) has been widely utilized to investigate gene functions and has significant potential for control of pest insects. However, recent studies have revealed that the target insect species, dsRNA molecule length, target genes, and other experimental factors can affect the efficiency of RNAi mediated control, restricting the further development and application of this technology. Therefore, the aim of this study was to establish a deep learning model using bioinformatics to help researchers identify dsRNA fragments with the highest RNAi efficiency. In this study, we optimized an existing model, namely, dsRNAPredictor, by designing sub-models based on different sequence lengths. Accordingly, the data were divided into two groups: 130–399 bp and 400–616 bp long sequences. Then, one-hot encoding was employed to extract sequence information. The convolutional neural network framework comprising three convolutional layers, three average pooling layers, a flattened layer, and three dense layers was employed as the classifier. By adjusting the parameters, we established two sub-models for different sequence distributions. Using multiple independent test datasets and conducting hypothesis testing, we demonstrated that our model exhibits superior performance and strong robustness to dsRNAPredictor, respectively. Therefore, our model may help design dsRNAs with pre-screening potential and facilitate further research and applications.

Integrated convolutional kernel based on two-dimensional photonic crystals.

Optical neural networks (ONNs) exhibit significant potential for accelerating artificial intelligence task processing due to their low latency, high bandwidth, and parallel processing capabilities. Photonic crystals (PhCs) are extensively utilized in integrated optoelectronics because of their unique photonic bandgap properties and precise control of light waves. In this study, we propose an optical reconfigurable convolutional kernel based on PhCs. This kernel can perform convolutional operations on weights by constructing a PhC weight bank. The convolutional kernel demonstrates exceptional performance within the developed optical convolutional neural network framework, successfully realizing various image edge processing tasks. It achieves blind recognition accuracies of 97.81% for the MNIST dataset and 80.31% for the Fashion-MNIST dataset. This study not only demonstrates the feasibility of constructing optical neural networks based on PhCs but to our knowledge, also offers new avenues for the future development of optical computing.

A Convolutional Neural Network Model for Crop Disease Detection System

Crop diseases pose a significant challenge to global food security, adversely impacting agricultural output and resulting in considerable economic repercussions. The prompt and precise identification of these diseases is essential for effective intervention and sustainable agricultural practices. This study introduces a model based on Convolutional Neural Networks (CNNs) for the automated detection of crop diseases. The model employs advanced deep learning methodologies to recognize and categorize plant diseases through the analysis of leaf images. Our CNN framework is trained on an extensive dataset comprising both diseased and healthy plant images, employing multiple convolutional layers to extract intricate features, including texture, color variations, and patterns linked to specific diseases. The model demonstrates a high level of accuracy in identifying a variety of diseases across different crop species by learning from both overt symptoms and subtle cues. We evaluate the performance of the system using established metrics such as accuracy and precision, thereby validating its efficacy in practical applications. The proposed system is designed for implementation in low-resource agricultural settings, offering farmers a cost-effective, dependable, and real-time solution for monitoring crop health.

Uncertainty-driven dynamic ensemble framework for rotating machinery fault diagnosis under time-varying working conditions

Multi-scale ensemble learning combines different scales of feature resolution, thereby improving fault diagnostic accuracy. However, the effectiveness of different information scales in characterizing fault features under time-varying speed conditions varies with speed. It is difficult for existing ensemble strategies to ensure the effectiveness of feature information when ensemble multi-scale feature information is involved. Accordingly, we propose an uncertainty-driven dynamic ensemble Bayesian convolutional neural network (DEBCNN) framework. The uncertainty of the results of different scale models was used to dynamically determine their weights in the ensemble framework, which reduced the influence of irrelevant features on the diagnostic results. By employing the proposed dynamic ensemble strategy, the ensemble framework can utilize fault feature information corresponding to different rotational speeds in the final diagnostic results. Experiments on motor and bearing datasets illustrate the superiority of this strategy over other techniques. This study provides useful insights for further research in the field of fault diagnosis of rotating machinery at time-varying speeds.

RT-Cabi: an Internet of Things based framework for anomaly behavior detection with data correction through edge collaboration and dynamic feature fusion

The rapid advancement of Internet of Things (IoT) technologies brings forth new security challenges, particularly in anomaly behavior detection in traffic flow. To address these challenges, this study introduces RT-Cabi (Real-Time Cyber-Intelligence Behavioral Anomaly Identifier), an innovative framework for IoT traffic anomaly detection that leverages edge computing to enhance the data processing and analysis capabilities, thereby improving the accuracy and efficiency of anomaly detection. RT-Cabi incorporates an adaptive edge collaboration mechanism, dynamic feature fusion and selection techniques, and optimized lightweight convolutional neural network (CNN) frameworks to address the limitations of traditional models in resource-constrained edge devices. Experiments conducted on two public datasets, Edge-IIoT and UNSW_NB15, demonstrate that RT-Cabi achieves a detection accuracy of 98.45% and 90.94%, respectively, significantly outperforming existing methods. These contributions not only validate the effectiveness of the RT-Cabi model in identifying anomalous behaviors in IoT traffic but also offer new perspectives and technological pathways for future research in IoT security.

Numerical Weather Prediction of Sea Surface Temperature in South China Sea Using Attention-Based Context Fusion Network

Numerical weather prediction of sea surface temperature (SST) is crucial for regional operational forecasts. Deep learning offers an alternative approach to traditional numerical general circulation models for numerical weather prediction. In our previous work, we developed a sophisticated deep learning model known as the Attention-based Context Fusion Network (ACFN). This model integrates an attention mechanism with a convolutional neural network framework. In this study, we applied the ACFN model to the South China Sea to evaluate its performance in predicting SST. The results indicate that for a 1-day lead time, the ACFN model achieves a Mean Absolute Error of 0.215 °C and a coefficient of determination (R2) of 0.972. In addition, in situ buoy data were utilized to validate the forecast results. The Mean Absolute Error for forecasts using these data increased to 0.500 °C for a 1-day lead time, with a corresponding R2 of 0.590. Comparative analyses show that the ACFN model surpasses traditional models such as ConvLSTM and PredRNN in terms of accuracy and reliability.

The Identification and Quantification of Hidden Hazards in Small Scale Reservoir Engineering Based on Deep Learning: Intelligent Perception for Safety of Small Reservoir Projects in Jiangxi Province

This study aims to enhance the detection and assessment of safety hazards in small-scale reservoir engineering using advanced image processing and deep learning techniques. Given the critical importance of small reservoirs in flood management, water supply, and ecological balance, the effective monitoring of their structural integrity is crucial. This paper developed a fully convolutional semantic segmentation method for hidden danger images of small reservoirs using an encoding–decoding structure, utilizing a deep learning framework of convolutional neural networks (CNNs) to process and analyze high-resolution images captured by unmanned aerial vehicles (UAVs). The method incorporated data augmentation and adaptive learning techniques to improve model accuracy under diverse environmental conditions. Finally, the quantification data of hidden dangers (length, width, area, etc.) were obtained by converting the image pixels to the actual size. Results demonstrate significant improvements in detecting structural deficiencies, such as cracks and seepage areas, with increased precision and recall rates compared to conventional methods, and the HHSN-25 network (Hidden Hazard Segmentation Network with 25 layers) proposed in this paper outperforms other methods. The main evaluation indicator, mIoU of HHSN-25, is higher than other methods, reaching 87.00%, and the Unet is 85.50%, and the Unet++ is 85.55%. The proposed model achieves reliable real-time performance, allowing for early warning and effective management of potential risks. This study contributes to the development of more efficient monitoring systems for small-scale reservoirs, enhancing their safety and operational sustainability.

Accelerating Neural Network Inference in Handwritten Digit Recognition — Comparative Study

In deep learning, one of the popular subclasses of multilayer perceptron is the Convolutional Neural Networks (CNNs), optimized for image and semantic segmentation tasks. With a large number of floating-point operations and relatively less data transfer in the training phase, CNNs are well suited to be handled by parallel architectures. The high accuracy of CNNs comes at the cost of consequential compute and memory demands. In this article, we present a comparative analysis of a pre-trained CNN framework for recognizing a handwritten digit on different processing platforms. The performance of the neural network as the function of the parallel processing platforms offers parametric insights and factors that influence the inference on state-of-the-art Graphics Processing Units (GPUs) and systolic arrays such as Eyeriss and Tensor Processing Unit (TPU). Through inference time analysis, we observed that the systolic arrays can outperform upto 58.7 times better than a Turing architecture powered GPU. We show that while the efficiency with which the available resources are utilized is higher in the existing GPUs (upto 32%) than the TPUs, the efficiency of application-specific systolic arrays can be on par with that of GPUs. We present the results obtained from three different customized systolic array-based platforms which can be adopted by the designers to decide the hardware optimization goal.

A comparative study of early stage Alzheimer’s disease classification using various transfer learning CNN frameworks

ABSTRACT The current research explores the improvements in predictive performance and computational efficiency that machine learning and deep learning methods have made over time. Specifically, the application of transfer learning concepts within Convolutional Neural Networks (CNNs) has proved useful for diagnosing and classifying the various stages of Alzheimer’s disease. Using base architectures such as Xception, InceptionResNetV2, DenseNet201, InceptionV3, ResNet50, and MobileNetV2, this study extends these models by adding batch normalization (BN), dropout, and dense layers. These enhancements improve the model’s effectiveness and precision in addressing the specified medical issue. The proposed model is rigorously validated and evaluated using publicly available Kaggle MRI Alzheimer’s data consisting of 1280 testing images and 5120 patient training images. For comprehensive performance evaluation, precision, recall, F1-score, and accuracy metrics are utilized. The findings indicate that the Xception method is the most promising of those considered. Without employing five K-fold techniques, this model obtains a 99% accuracy and 0.135 loss score. In addition, integrating five K-fold methods enhances the accuracy to 99.68% while decreasing the loss score to 0.120. The research further included the evaluation of the Receiver Operating Characteristic Area Under the Curve (ROC-AUC) for various classes and models. As a result, our model may detect and diagnose Alzheimer’s disease quickly and accurately.

Accelerated 2D radial Look-Locker T1 mapping using a deep learning-based rapid inversion recovery sampling technique.

Efficient abdominal coverage with T1-mapping methods currently available in the clinic is limited by the breath hold period (BHP) and the time needed for T1 recovery. This work develops a T1-mapping framework for efficient abdominal coverage based on rapid T1 recovery curve (T1RC) sampling, slice-selective inversion, optimized slice interleaving, and a convolutional neural network (CNN)-based T1 estimation. The effect of reducing the T1RC sampling was evaluated by comparing T1 estimates from T1RC ranging from 0.63 to 2.0s with reference T1 values obtained from T1RC =2.5-5s. Slice interleaving methodologies were evaluated by comparing the T1 variation in abdominal organs across slices. The repeatability of the proposed framework was demonstrated by performing acquisition on test subjects across imaging sessions. Analysis of in vivo data based on retrospectively shortening the T1RC showed that with the CNN framework, a T1RC =0.84 s yielded T1 estimates without significant changes in mean T1 (p> 0.05) or significant increase in T1 variability (p> 0.48) compared to the reference. Prospectively acquired data using T1RC =0.84 s, an optimized slice interleaving scheme, and the CNN framework enabled 21 slices in a 20 s BHP. Analyses across abdominal organs produced T1 values within 2% of the reference. Repeatability experiments yielded Pearson's correlation, repeatability coefficient, and coefficient of variation of 0.99, 2.5%, and 0.12%, respectively. The proposed T1 mapping framework provides full abdominal coverage within a single BHP.

Leak detection in water supply pipeline with small-size leakage using deep learning networks

Pipeline transportation technology has witnessed remarkable development in recent years, while the issue of leakage poses a serious limitation to this technology’s progress, especially for the small-scale leakage in water pipelines that exhibits weaker leakage signals. To address those challenges, this study investigates leak detection in water supply pipeline with an inner pipe diameter of 100 mm in the operating range of volume flow Q = 40–80 m3/h, leakage flow Ql = 0.65–1.33 m3/h and pressure p = 130–220 kPa. The characteristics and propagation of leakage pressure signals are analyzed. An adaptive improved variational mode decomposition (VMD) based on the white shark optimizer (WSO) algorithm is proposed, utilizing sample entropy as the fitness function. Additionally, tent chaotic mapping is employed for parameter initialization, mean normalization is used to minimize the influence of the DC component, and signal relative energy serves as a boundary condition. The processed data is utilized to establish leakage detection models using long short-term memory network (LSTM), gated recurrent unit (GRU), convolutional neural network (CNN) and SqueezeNet framework. The performance of leakage detection models is evaluated and compared using 962 sets’ data using evaluation indicators and confusion matrix. The results indicate that SqueezeNet-GRU demonstrates superior performance in pipeline leak detection and localization applications, achieving 97.40 % accuracy for the total dataset.

An interpretation-based convolution neural network framework for geophysical data fusion and aquifer structure identification

Identification of 3D realistic aquifer structures is essential for predicting physicochemical processes in groundwater systems. However, the characterization of highly heterogeneous aquifers remains challenging because it relies on the effective fusion of multiple geophysical data sources having wide areal coverage, as well as downhole geophysical data featuring high resolution. This study establishes a novel 3D convolutional neural network model to generate aquifer structure from 3D seismic data, constrained by sparse downhole sonic and lithology logs. In the model, the data fusion procedure is designed to follow the logics of conventional manual interpretation of multiple geophysical data, and to address the 3D spatial relationships between geophysical data and lithology. The method is implemented in a typical fluvial aquifer featuring coarse paleovalley sediments (sandstone) embedded in the tight surrounding rocks (claystone), in order to identify channelized sandstone from low-permeability claystone. It is confirmed that the proposed model reliably generates 3D aquifer structures based on seismic amplitudes, downhole sonic and lithology logs. The method is compared to traditional machine learning models that focus on 1D conversion from geophysical attributes to lithology. The results show that the newly-developed model performs more robustly and accurately because the use of 3D convolution allows considering the relationships between seismic amplitude, sonic velocity and lithology in both vertical and horizontal directions. Moreover, the inclusion of sonic logs constraint in the model, following the logics of manual seismic data interpretation, significantly improves the model accuracy. The method can find broad applications for the characterization of subsurface heterogeneity even featuring non-gaussian permeability distribution like the demonstrated fluvial aquifer.

Predicting the degree of rubber rupture damage using a GAN-enhanced Bayesian-optimized 1DCNN network

Influenced by factors such as temperature, aging, and overloading, rubber bearings may undergo rupture during prolonged usage, leading to severe consequences such as bearing failure and structural damage. To accurately assess the degree of rubber rupture damage, this study proposed a novel generative adversarial network (GAN)-enhanced Bayesian-optimized one-dimensional convolutional neural network (1DCNN) framework (GAN-BCNN). In the GAN-BCNN framework, GAN is used to enhance the proportion of damaged data in the database, while Bayesian optimization is utilized to optimize crucial hyperparameters in the 1DCNN network. The predictive results indicate that the proposed GAN-BCNN model accurately predicts the degree of rubber rupture damage, achieving high classification accuracies of 99.1, 95.5, 92.9, and 98.0% for the categories of no damage, slight damage, moderate damage, and severe damage, respectively. When compared to the linear discriminant analysis model and the BCNN model without GAN enhancement, the proposed GAN-BCNN model exhibits a significant improvement in predictive accuracy, with an increase in Accuracy values of 52.4 and 6.0%, respectively. These results indicate that the GAN-BCNN model holds promising practical applications, and the GAN algorithm contributes to improving the prediction accuracy of the model.

Attention-Guided Sample-Based Feature Enhancement Network for Crowded Pedestrian Detection Using Vision Sensors.

Occlusion presents a major obstacle in the development of pedestrian detection technologies utilizing computer vision. This challenge includes both inter-class occlusion caused by environmental objects obscuring pedestrians, and intra-class occlusion resulting from interactions between pedestrians. In complex and variable urban settings, these compounded occlusion patterns critically limit the efficacy of both one-stage and two-stage pedestrian detectors, leading to suboptimal detection performance. To address this, we introduce a novel architecture termed the Attention-Guided Feature Enhancement Network (AGFEN), designed within the deep convolutional neural network framework. AGFEN improves the semantic information of high-level features by mapping it onto low-level feature details through sampling, creating an effect comparable to mask modulation. This technique enhances both channel-level and spatial-level features concurrently without incurring additional annotation costs. Furthermore, we transition from a traditional one-to-one correspondence between proposals and predictions to a one-to-multiple paradigm, facilitating non-maximum suppression using the prediction set as the fundamental unit. Additionally, we integrate these methodologies by aggregating local features between regions of interest (RoI) through the reuse of classification weights, effectively mitigating false positives. Our experimental evaluations on three widely used datasets demonstrate that AGFEN achieves a 2.38% improvement over the baseline detector on the CrowdHuman dataset, underscoring its effectiveness and potential for advancing pedestrian detection technologies.

A Transparent Pipeline for Identifying Sexism in Social Media: Combining Explainability with Model Prediction

In this study, we present a new approach that combines multiple Bidirectional Encoder Representations from Transformers (BERT) architectures with a Convolutional Neural Network (CNN) framework designed for sexism detection in text at a granular level. Our method relies on the analysis and identification of the most important terms contributing to sexist content using Shapley Additive Explanations (SHAP) values. This approach involves defining a range of Sexism Scores based on both model predictions and explainability, moving beyond binary classification to provide a deeper understanding of the sexism-detection process. Additionally, it enables us to identify specific parts of a sentence and their respective contributions to this range, which can be valuable for decision makers and future research. In conclusion, this study introduces an innovative method for enhancing the clarity of large language models (LLMs), which is particularly relevant in sensitive domains such as sexism detection. The incorporation of explainability into the model represents a significant advancement in this field. The objective of our study is to bridge the gap between advanced technology and human comprehension by providing a framework for creating AI models that are both efficient and transparent. This approach could serve as a pipeline for future studies to incorporate explainability into language models.

Intelligent accounting optimization method based on meta-heuristic algorithm and CNN.

The evolution of social intelligence has led to the adoption of intelligent accounting practices in enterprises. To enhance the efficiency of enterprise accounting operations and improve the capabilities of accountants, we propose an intelligent accounting optimization approach that integrates meta-heuristic algorithms with convolutional neural networks (CNN). First, we enhance the CNN framework by incorporating document and voucher information into accounting audits, creating a multi-modal feature extraction mechanism. Utilizing these multi-modal accounting features, we then introduce a method for assessing accounting quality, which objectively evaluates financial performance. Finally, we propose an optimization technique based on meta-heuristic principles, combining genetic algorithms with annealing models to improve the accounting system. Experimental results validate our approach, demonstrating an accuracy of 0.943 and a mean average precision (mAP) score of 0.812. This method provides technological support for refining accounting audit mechanisms.

Multi-Type Self-Attention-Based Convolutional-Neural-Network Post-Filtering for AV1 Codec

Over the past few years, there has been substantial interest and research activity surrounding the application of Convolutional Neural Networks (CNNs) for post-filtering in video coding. Most current research efforts have focused on using CNNs with various kernel sizes for post-filtering, primarily concentrating on High-Efficiency Video Coding/H.265 (HEVC) and Versatile Video Coding/H.266 (VVC). This narrow focus has limited the exploration and application of these techniques to other video coding standards such as AV1, developed by the Alliance for Open Media, which offers excellent compression efficiency, reducing bandwidth usage and improving video quality, making it highly attractive for modern streaming and media applications. This paper introduces a novel approach that extends beyond traditional CNN methods by integrating three different self-attention layers into the CNN framework. Applied to the AV1 codec, the proposed method significantly improves video quality by incorporating these distinct self-attention layers. This enhancement demonstrates the potential of self-attention mechanisms to revolutionize post-filtering techniques in video coding beyond the limitations of convolution-based methods. The experimental results show that the proposed network achieves an average BD-rate reduction of 10.40% for the Luma component and 19.22% and 16.52% for the Chroma components compared to the AV1 anchor. Visual quality assessments further validated the effectiveness of our approach, showcasing substantial artifact reduction and detail enhancement in videos.

Drone-Based Visible–Thermal Object Detection with Transformers and Prompt Tuning

The use of unmanned aerial vehicles (UAVs) for visible–thermal object detection has emerged as a powerful technique to improve accuracy and resilience in challenging contexts, including dim lighting and severe weather conditions. However, most existing research relies on Convolutional Neural Network (CNN) frameworks, limiting the application of the Transformer’s attention mechanism to mere fusion modules and neglecting its potential for comprehensive global feature modeling. In response to this limitation, this study introduces an innovative dual-modal object detection framework called Visual Prompt multi-modal Detection (VIP-Det) that harnesses the Transformer architecture as the primary feature extractor and integrates vision prompts for refined feature fusion. Our approach begins with the training of a single-modal baseline model to solidify robust model representations, which is then refined through fine-tuning that incorporates additional modal data and prompts. Tests on the DroneVehicle dataset show that our algorithm achieves remarkable accuracy, outperforming comparable Transformer-based methods. These findings indicate that our proposed methodology marks a significant advancement in the realm of UAV-based object detection, holding significant promise for enhancing autonomous surveillance and monitoring capabilities in varied and challenging environments.

Physics-guided interpretable CNN for SAR target recognition

Deep Learning (DL) model has been widely used in the field of Synthetic Aperture Radar Automatic Target Recognition (SAR-ATR) and has achieved excellent performance. However, the black-box nature of DL models has been the focus of criticism, especially in the application of SAR-ATR, which is closely associated with the national defense and security domain. To address these issues, a new interpretable recognition model Physics-Guided BagNet (PGBN) is proposed in this article. The model adopts an interpretable convolutional neural network framework and uses time–frequency analysis to extract physical scattering features in SAR images. Based on the physical scattering features, an unsupervised segmentation method is proposed to distinguish targets from the background in SAR images. On the basis of the segmentation result, a structure is designed, which constrains the model’s spatial attention to focus more on the targets themselves rather than the background, thereby making the model’s decision-making more in line with physical principles. In contrast to previous interpretable research methods, this model combines interpretable structure with physical interpretability, further reducing the model’s risk of error recognition. Experiments on the MSTAR dataset verify that the PGBN model exhibits excellent interpretability and recognition performance, and comparative experiments with heatmaps indicate that the physical feature guidance module presented in this article can constrain the model to focus more on the target itself rather than the background.