Search

End-to-end entity relation extraction can be decomposed into named entity recognition and relation extraction, most recent works model these two subtasks jointly. Existing pipelined approaches validate the importance of fusing entity type information in the relation model and the potential of pipeline models, but they ignore the possibility that certain entities in the text may have multiple types at the same time, which is particularly common in Chinese datasets. This paper proposed an entity cascading type mechanism to address the aforementioned issues and developed a pipeline model named CENTRELINE, which was more suitable for Chinese relation extraction. This pipelined approach incorporated an entity module, which was a wordword relation classification model. It employed BERT and bi-directional LSTM as encoders, introduced dilated convolution after conditional layer normalization, and finally generated outputs for entities and their cascading types using a cascading type predictor. The input of the relation module was only constructed by the entity module. The F₁ values of this method surpass the baseline by 7. 23%, 6.93%, and 8.51% on DulE1.0, DulE2.0, and CMeIE-V2 datasets, respectively. This method achieves state-of-the-art performance on both DuIE1.0 and DuIE2.0 datasets. The results of ablation experiments indicate that both the proposed cascading type mechanism and the pipeline model refined based on Chinese language characteristics can enhance the performance of relation extraction. [ABSTRACT FROM AUTHOR]

The satellite images have been employed in building extraction to aid urban planning, tax assessment, disaster management, etc. The number of buildings and building types is huge in urban areas, which puts more burden on human experts to extract buildings in satellite images. Hence, building extraction from satellite images using deep learning (DL) has become an emerging research domain in recent decades. The performance of the DL model depends on training parameters, the depth of the model, and the memory required to preserve the model. In this work, a Memory-Efficient Residual Dilated Convolutional Network (MRDCN) has been proposed to extract buildings effectively with reduced number of training parameters and with lesser memory consumption. The model is trained using the Massachusetts buildings dataset and implemented using PyTorch in Kaggle platform. The trained model has been tested using both Massachusetts and AIRS Dataset. The simulation results prove that the proposed model uses 31.64% less memory than the existing dilated residual network. It is evident from the results that the MRDCN is able to extract the buildings with better accuracy and an Intersection of Union with minimal memory consumption than the existing standard UNet, SegNet, ResUNet, and Dilated ResUNet models. [ABSTRACT FROM AUTHOR]

Contemporary advancements in deep learning have spurred widespread adoption of spatio-temporal prediction across various scientific disciplines. Nonetheless, traffic flow prediction, as a quintessential spatio-temporal task, continues to present significant challenges, such as the accurate modeling of complex dependencies and dynamic changes over time and space. To address these issues, this paper introduces the Dual-Branch and Multi-Temporal Resolution Convolutional Network with an Adaptive Graph Neural Ordinary Differential Equation (DM-AGODE) model. This innovative approach integrates an optimized graph neural ordinary differential equation with an adaptive correlation adjacency graph, ensuring precise feature propagation across the network. The model incorporates a Dual-Branch Learning (DBL) mechanism to effectively differentiate between short-term dynamics and long-term trends, while the Multi-Temporal Resolution Convolution (MTRC) method enhances the processing of temporal data across multiple scales, critical for capturing the complex behaviors of traffic flow. Furthermore, to demonstrate the effectiveness of our model, we conducted a comprehensive evaluation of our model using six widely recognized real-world datasets, which highlighted its superior adaptability to complex traffic flow patterns. Compared to the leading baseline model, our approach achieves an improvement in prediction accuracy exceeding 8% and significantly enhances efficiency in processing. [ABSTRACT FROM AUTHOR]

Mechanical equipment usually operates in noisy and variable load environments, which presents serious challenges for existing intelligent diagnostic models. In addition, there are few labelled fault samples in real engineering scenarios, which makes it difficult to perform accurate fault identification for mechanical equipment. Thus, to solve the problem of diagnostic model performance degradation under small sample, noisy and variable load environments, this paper proposes a Multiscale Dilated Convolution and Swin-Transformer (MSDC-Swin-T) method for small sample gearbox fault diagnosis. First, we design the Coordinate Reconstruction Attention Mechanism (CRAM), which enhances the capture of impulse information by coordinate reconstruction. In addition, a multiscale convolutional token embedding module is constructed to extract local features at different scales, and its ability for capturing important features is adaptively enhanced by CRAM. Then, Swin-Transformer is utilized for modeling global dependencies, thus mining more subtle fault features. Finally, the effectiveness and stability of the MSDC-Swin-T is proved on two gearbox datasets. The experiments show that MSDC-Swin-T has superior diagnostic performance under small sample with noise and variable load environments. The diagnostic accuracy is better than the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Although pedestrian detection techniques are improving, this task is still challenging due to the problems of target occlusion, small targets, and complex pedestrian backgrounds in images of different scenes. As a result, the You Only Look Once algorithm exhibits lower detection accuracy. In this paper, a multisensor (MS) module that utilizes multiple dilated convolutions is proposed to sample feature images, avoiding information loss caused by repeated sampling, to improve the feature extraction and target detection performance of the algorithm. In addition, a lightweight shuffle-based efficient channel attention mechanism is introduced to conduct grouping in the channel dimension and perform parallel processing for each subfeature map channel. A new branch is introduced to enrich the channel feature information for multiscale feature representation. Finally, a distance intersection over union-based nonmaximum suppression (DIoU-NMS) method is introduced to minimize the occurrence of missed targets due to occlusion by taking the prediction box and ground truth box centroid locations information into account without increasing the computational cost over that of normal NMS. Our method is extensively evaluated on several challenging pedestrian detection datasets, achieving 87.73%, 34.7%, 93.96% and 95.23% mean average precision values on PASCAL VOC, MS COCO, Caltech Pedestrian and INRIA Person, which are respectively. The experimental results demonstrate the effectiveness of the method. [ABSTRACT FROM AUTHOR]

Aiming at the problems of traditional image super-resolution reconstruction algorithms in the image reconstruction process, such as small receptive field, insufficient multi-scale feature extraction, and easy loss of image feature information, a super-resolution reconstruction algorithm of multi-scale dilated convolution network based on dilated convolution is proposed in this paper. First, the algorithm extracts features from the same input image through the dilated convolution kernels of different receptive fields to obtain feature maps with different scales; then, through the residual attention dense block, further obtain the features of the original low resolution images, local residual connections are added to fuse multi-scale feature information between multiple channels, and residual nested networks and jump connections are used at the same time to speed up deep network convergence and avoid network degradation problems. Finally, deep network extraction features, and it is fused with input features to increase the nonlinear expression ability of the network to enhance the superresolution reconstruction effect. Experimental results show that compared with Bicubic, SRCNN, ESPCN, VDSR, DRCN, LapSRN, MemNet, and DSRNet algorithms on the Set5, Set14, BSDS100, and Urban100 test sets, the proposed algorithm has improved peak signal-to-noise ratio and structural similarity, and reconstructed images. The visual effect is better. [ABSTRACT FROM AUTHOR]

Aiming at the problem that face images are easily interfered by occlusion factors in uncontrollable environments, and the complex structure of traditional convolutional neural networks leads to low expression recognition rates, slow network convergence speed, and long network training time, an improved lightweight convolutional neural network is proposed for facial expression recognition algorithm. First, the dilation convolution is introduced into the shortcut connection of the inverted residual structure in the MobileNetV3 network to expand the receptive field of the convolution kernel and reduce the loss of expression features. Then, the channel attention mechanism SENet in the network is replaced by the two-dimensional (channel and spatial) attention mechanism SimAM introduced without parameters to reduce the network parameters. Finally, in the normalization operation, the Batch Normalization of the backbone network is replaced with Group Normalization, which is stable at various batch sizes, to reduce errors caused by processing small batches of data. Experimental results on RaFD, FER2013, and FER2013Plus face expression data sets show that the network reduces the training times while maintaining network accuracy, improves network convergence speed, and has good convergence effects. [ABSTRACT FROM AUTHOR]

When performing remote sensing image segmentation, practitioners often encounter various challenges, such as a strong imbalance in the foreground–background, the presence of tiny objects, high object density, intra-class heterogeneity, and inter-class homogeneity. To overcome these challenges, this paper introduces AerialFormer, a hybrid model that strategically combines the strengths of Transformers and Convolutional Neural Networks (CNNs). AerialFormer features a CNN Stem module integrated to preserve low-level and high-resolution features, enhancing the model's capability to process details of aerial imagery. The proposed AerialFormer is designed with a hierarchical structure, in which a Transformer encoder generates multi-scale features and a multi-dilated CNN (MDC) decoder aggregates the information from the multi-scale inputs. As a result, information is taken into account in both local and global contexts, so that powerful representations and high-resolution segmentation can be achieved. The proposed AerialFormer was benchmarked on three benchmark datasets, including iSAID, LoveDA, and Potsdam. Comprehensive experiments and extensive ablation studies show that the proposed AerialFormer remarkably outperforms state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Underwater segmentation technology achieves effective monitoring of fish biological information through accurate identification of fish species and precise estimation of their quantities. It serves as an effective approach to enhance the informatization level of aquaculture and promote intelligent management in fisheries. However, the complex and diverse underwater environment, coupled with poor visibility, results in blurry and lower-quality underwater fish images. The current image segmentation methods, when applied to fish segmentation, exhibit low accuracy and inadequate generalization capabilities. This paper proposes an image segmentation model based on Atrous Spatial Pyramid Pooling (ASPP) to address these challenges. The model aims to improve fish feature extraction and enhance the segmentation precision of fish images in complex underwater environments, thereby enhancing the accuracy and generalization capabilities of existing fish segmentation models. First, a multiscale feature extraction module (triangular atrous spatial pyramid multifeature fusion) based on dilated convolutional spatial pyramid pooling, which enhances the extraction of high-level semantic features of images through the triangular combination of multilayer dilated convolutional pyramid pooling and the adaptive channel attention module, is proposed. Second, a spatial attention module based on strip pooling (atrous strip pooling) is proposed, which further expands the receptive field of the attention mechanism by combining different expansion rates, enhances the correlations between pixels, and effectively captures spatial information. Finally, a decoder module based on multilayer semantic feature fusion is proposed. Through the processing and fusion of medium-, low-, and high-level semantic features, the model understands image content and performs accurate pixel-level segmentation. The proposed model is evaluated using a VOC-compliant dataset created from underwater fish images and validated against public and specific underwater fish datasets. The results demonstrate the successful application of the feature extraction and feature fusion modules in underwater fish image segmentation, achieving an average Intersection over Union (MIoU) of 85.49% in segmentation tasks. Compared to conventional segmentation models, the proposed model shows significant improvements, with an average increase of 3.8% in MIoU and 2.5% in balanced F-score (F1-score). The accurate segmentation of underwater fish images and extraction of vital biological information by this model provide a solid foundation for intelligent monitoring, including fish length measurement, weight estimation, and analysis of growth and health status. Moreover, the model offers a scientific framework for decision-making in aquaculture, driving advancements in precision and intelligent management practices in the field. [ABSTRACT FROM AUTHOR]

Due to the increasing severity of aging populations in modern society, the accurate and timely identification of, and responses to, sudden abnormal behaviors of the elderly have become an urgent and important issue. In the current research on computer vision-based abnormal behavior recognition, most algorithms have shown poor generalization and recognition abilities in practical applications, as well as issues with recognizing single actions. To address these problems, an MSCS–DenseNet–LSTM model based on a multi-scale attention mechanism is proposed. This model integrates the MSCS (Multi-Scale Convolutional Structure) module into the initial convolutional layer of the DenseNet model to form a multi-scale convolution structure. It introduces the improved Inception X module into the Dense Block to form an Inception Dense structure, and gradually performs feature fusion through each Dense Block module. The CBAM attention mechanism module is added to the dual-layer LSTM to enhance the model's generalization ability while ensuring the accurate recognition of abnormal actions. Furthermore, to address the issue of single-action abnormal behavior datasets, the RGB image dataset RIDS (RGB image dataset) and the contour image dataset CIDS (contour image dataset) containing various abnormal behaviors were constructed. The experimental results validate that the proposed MSCS–DenseNet–LSTM model achieved an accuracy, sensitivity, and specificity of 98.80%, 98.75%, and 98.82% on the two datasets, and 98.30%, 98.28%, and 98.38%, respectively. [ABSTRACT FROM AUTHOR]

Land-cover segmentation, a fundamental task within the domain of remote sensing, boasts a broad spectrum of application potential. We address the challenges in land-cover segmentation of remote-sensing imagery and complete the following work. Firstly, to tackle the issues of foreground–background imbalance and scale variation, a module based on multi-dilated rate convolution fusion was integrated into a decoder. This module extended the receptive field through multi-dilated convolution, enhancing the model's capability to capture global features. Secondly, to address the diversity of scenes and background interference, a hybrid attention module based on large-kernel convolution was employed to improve the performance of the decoder. This module, based on a combination of spatial and channel attention mechanisms, enhanced the extraction of contextual information through large-kernel convolution. A convolution kernel selection mechanism was also introduced to dynamically select the convolution kernel of the appropriate receptive field, suppress irrelevant background information, and improve segmentation accuracy. Ablation studies on the Vaihingen and Potsdam datasets demonstrate that our decoder significantly outperforms the baseline in terms of mean intersection over union and mean F1 score, achieving an increase of up to 1.73% and 1.17%, respectively, compared with the baseline. In quantitative comparisons, the accuracy of our improved decoder also surpasses other algorithms in the majority of categories. The results of this paper indicate that our improved decoder achieves significant performance improvement compared with the old decoder in remote-sensing image-segmentation tasks, which verifies its application potential in the field of land-cover segmentation. [ABSTRACT FROM AUTHOR]

Aiming at the problems of road fracture and detail loss caused by not considering the geometric features in the road extraction method. We proposed an encoder-decoder architecture based on a multi-scale asymmetric dual attention mechanism. Firstly, A multi-scale convolution block in the shape of 'Union Jack' is designed. It includes symmetric convolution and asymmetric convolution along horizontal, vertical, left diagonal, and right diagonal spatial directions, and a multi-scale dilated convolution for extracting features of different scales. Remote dependence relationships are highly converged by using it, and road fracture problems caused by occlusion can be solved effectively. Secondly, a directional dual attention mechanism is proposed, which consists of directional channel attention using strip pooling and a directional spatial attention mechanism using asymmetric convolution along left diagonal and right diagonal spatial directions. It can use the directivity of asymmetric convolution to allocate attention mechanism adaptively in attention mechanism, and effectively avoid the road detail loss problem. Finally, we conducted corresponding experiments on the DeepGlobe and Ottawa road datasets, and the experimental results are superior to the current state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Copyright of Chinese Journal of Liquid Crystal & Displays is the property of Chinese Journal of Liquid Crystal & Displays and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

PANet is widely used in various object detection tasks due to its powerful feature expression ability. However, PANet's performance in complex scenarios is subpar, with frequent object omission or misidentification. We find that the reason for this phenomenon is that the receptive field of PANet can't cover sufficient feature information, to deal with drastic changes of source object size. In order to solve this problem, this paper adopts dilated convolution technology and applies it to each parallel branch directly following the PANet network. This method can effectively represent the feature information of objects at different scales by integrating the information from small and large receptive fields into a new feature output. We also introduce residual structure to circumvent the network degradation caused by excessive convolutions. By combining the above methods, we build a new module named PANetW (PANet with Wider Receptive Fields). Taking YOLOX-S as the baseline, we comprehensively evaluated the proposed module PANetW on two datasets, VOC2007 and MSCOCO2017. The test results show that our PANetW achieves a high level of mean average precision (AP). On the VOC2007 dataset, the AP of our PANetW improves by 4.9% to 43.0%; on the MS COCO2017 dataset, the AP of PANetW is as high as 44.3%, far exceeding the current mainstream modules. The experimental results fully demonstrate the effectiveness of our module. [ABSTRACT FROM AUTHOR]

This paper presents a novel crack size estimation method based on dilated convolution and time series fine tuning using in‐situ Lamb wave. The contributions of this proposed method are twofold: (1) a dilated convolution‐based neural network which is an effective representation learning methods with larger receptive fields for periodic signals and (2) a crack length adjustment approach based on previous estimations with a time series fine tuning method. The time series fine tuning method utilizes the monotonicity of the crack length with respect to time. A new evaluation metric has been adopted, which considers prediction errors, time errors, and monotonicity errors. The proposed method outperforms the previous feature extraction‐based methods with or without time series fine tuning in this new evaluation metric. Highlights: Crack estimation method using deep learning and time series fine tuning was proposed.The neural network using a combination of dilated convolution and long short‐term memory (LSTM) network was designed.The proposed method outperformed prior manual feature extraction‐based approaches. [ABSTRACT FROM AUTHOR]

In order to improve the spatial resolution of brain magnetic resonance imaging (MRI) images, the super resolution (SR) reconstruction method based on deep convolutional neural network (CNN) has achieved remarkable results. Increasing network depth or width usually can expand the receptive field of the network to improve the quality of reconstruction, but it is difficult to train the network due to the large parameter quantity and huge computing requirements. Meanwhile, the network does not pay enough attention to the medium and high frequency features of the image, which affects the quality of reconstruction. To solve the above problems, this paper proposes a cascaded channels-space attention and encode-decode network for 3D-MRI image SR reconstruction. Firstly, feature extraction is carried out in the low resolution space, and the image features are extracted by the symmetric connected encoder-decoder with dilated convolution to alleviate the checkerboard artifacts. Secondly, a serial connected channel-space attention module is constructed to capture the interdependence between the feature channels by channel attention, and the spatial attention is increased to assign weight to information of different positions, effectively enhancing the network' s learning of medium and high frequency information. Finally, subpixel convolution is used to sample the feature map to the target image size and meanwhile reduce memory consumption. Experimental results on public Kirby21 and BraTS datasets show that the proposed method is superior in both quantitative peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and subjective visual quality when compared with traditional SR algorithms and mainstream CNN-based SR algorithms. [ABSTRACT FROM AUTHOR]

The student feedback data possesses to be the fundamental influencers of decision-making process in diverse applications. The performance prediction based on student's feedback about the educational institution helps for better solution recommendation. The automated solution recommendation based on student's feedback extensively support the educational institution to make better decisions for improvisation. In most of the existing research works, the performance can be analysed, but suitable solution recommendation is not provided. Also, the existing recommendation works fail to generate accurate outcomes, consumes more time with higher error rates. Hence on diminishing the existing issues, this research work presents a Data science-based solution Recommendation model based on hybrid deep learning approaches. Pre-processing, feature extraction, feature clustering, performance prediction, and recommendation are the steps in the suggested model. In this research, the student feedback data is collected from Kaggle source and some of the attributes are added manually. Pre-processing techniques for the text data include stop-word-removal, tokenization, case-folding, and stemming. The features are extracted using Enhanced Lexicon bidirectional encoder representations from transformers (ELexBert) model. The significant attributes are selected using Adaptive Coati optimization (ACoaT) algorithm. The selected features are clustered based on feature similarity using Upgraded density based k-means clustering (Uden_KMC) model. A new type of hybrid deep learning model known as Channel Block Densnet with Dilated Convolution BiLSTM (ChaBDBiL) is employed for recommending better decisions. The recommendation performances using feedback data are evaluated using PYTHON where the overall accuracy of 98.19%, specificity of 98.25%, F1 score of 91.28%, sensitivity of 97.41% and Kappa score of 91.28% are obtained. [ABSTRACT FROM AUTHOR]

Copyright of Chinese Journal of Liquid Crystal & Displays is the property of Chinese Journal of Liquid Crystal & Displays and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In complex traffic scenarios, it is crucial to develop a rapid and precise real-time detection system for non-motorized vehicles to ensure safe driving. D-YOLO is a lightweight real-time detection technique for non-motorized vehicles based on an enhanced version of YOLOv4-tiny. Typically, the computing capabilities of mobile devices are constrained, therefore we begin by reducing the number of model parameters. Then, we add dilated convolution and depthwise separable convolution into the network's Cross Stage Partial Connection (CSPNet) in order to produce the DCSPNet with improved performance. Coordinate Attention(CA) is implemented to enhance the network's capability to extract effective features. In the neck network of the model by introducing a spatial pyramid set (SPP) to enhance the feature representation of non-motorized vehicles in the feature layer. Finally, we test this proposed model on dataset, the experimental results show that D-YOLO has a model size of only 6.7MB, which is 16.5MB smaller than YOLOv4-tiny. The detection speed of D-YOLO is about 25% faster than that of YOLOv4-tiny, D-YOLO has approximately 58% fewer model parameters than YOLOv4-tiny, D-YOLO has a mAP of 70.36%, which is 2.01% higher than YOLOv4-tiny. It can be shown that D-YOLO ensures both accuracy and real-time performance to satisfies the demand of real-time detection of non-motorized vehicles in intelligent traffic scenarios. [ABSTRACT FROM AUTHOR]

To overcome the challenges of feature selection in traditional machine learning and enhance the accuracy of deep learning methods for anomaly traffic detection, we propose a novel method called DCGCANet. This model integrates dilated convolution, a GRU, and a Channel Attention Network, effectively combining dilated convolutional structures with GRUs to extract both temporal and spatial features for identifying anomalous patterns in network traffic. The one-dimensional dilated convolution (DC-1D) structure is designed to expand the receptive field, allowing for comprehensive traffic feature extraction while minimizing information loss typically caused by pooling operations. The DC structure captures spatial dependencies in the data, while the GRU processes time series data to capture dynamic traffic changes. Furthermore, the channel attention (CA) module assigns importance-based weights to features in different channels, enhancing the model's representational capacity and improving its ability to detect abnormal traffic. DCGCANet achieved an accuracy rate of 99.6% on the CIC-IDS-2017 dataset, outperforming other algorithms. Additionally, the model attained precision, recall, and F1 score rates of 99%. The generalization capability of DCGCANet was validated on a subset of CIC-IDS-2017, demonstrating superior detection performance and robust generalization potential. [ABSTRACT FROM AUTHOR]

Accurate segmentation of the left ventricle (LV) using echocardiogram (Echo) images is essential for cardiovascular analysis. Conventional techniques are labor-intensive and exhibit inter-observer variability. Deep learning has emerged as a powerful tool for automated medical image segmentation, offering advantages in speed and potentially superior accuracy. This study explores the efficacy of employing a YOLO (You Only Look Once) segmentation model for automated LV segmentation in Echo images. YOLO, a cutting-edge object detection model, achieves exceptional speed–accuracy balance through its well-designed architecture. It utilizes efficient dilated convolutional layers and bottleneck blocks for feature extraction while incorporating innovations like path aggregation and spatial attention mechanisms. These attributes make YOLO a compelling candidate for adaptation to LV segmentation in Echo images. We posit that by fine-tuning a pre-trained YOLO-based model on a well-annotated Echo image dataset, we can leverage the model's strengths in real-time processing and precise object localization to achieve robust LV segmentation. The proposed approach entails fine-tuning a pre-trained YOLO model on a rigorously labeled Echo image dataset. Model performance has been evaluated using established metrics such as mean Average Precision (mAP) at an Intersection over Union (IoU) threshold of 50% (mAP50) with 98.31% and across a range of IoU thresholds from 50% to 95% (mAP50:95) with 75.27%. Successful implementation of YOLO for LV segmentation has the potential to significantly expedite and standardize Echo image analysis. This advancement could translate to improved clinical decision-making and enhanced patient care. [ABSTRACT FROM AUTHOR]

To address the challenges of fracture recognition in the CT scanning images of coal or rock, particularly the interference of gangue and the recognition of fractures at different scales, proposed and implemented a network model for coal-rock fracture extraction based on deep learning (MCSN). According to the U-Net architecture, the model utilized its encoder-decoder structure and skip connections to segment the fracture structure images from complex coal-rock body images. Firstly, the fracture structures in the CT scanning images were annotated manually using the internal scan images captured by a coal industrial CT scanning system. And the annotated original data was augmented to create a coal-rock CT fracture dataset. Subsequently, to make the extraction network of main features have a stronger extraction capability of fracture structure features, the weights of a pre-trained VGG16 model were transferred to the U-Net encoder through a transfer learning technique. Simultaneously, the decoder part of the U-Net model was improved using the deep separable dilated convolutional modules (DCAC) and residual modules to effectively boost the recognition capability of fracture structures in the CT images, demonstrating a superior segmentation accuracy and robustness. To validate the effectiveness of the coal-rock fracture extraction network model proposed, the results obtained by the MCSN were compared with those of classical convolutional neural networks and threshold segmentation methods. Experimental comparisons revealed a significant advantage of the model proposed in both qualitative and quantitative analyses. The proposed model, employing a multi-scale fusion strategy, demonstrated the capability to effectively extract fractures in complex coal-rock images, thereby enhancing the efficiency and accuracy of fracture identification. The model was applied to the identification of fractures in roadway surrounding rock based on borehole imaging. Fracture extraction was performed through the analysis of borehole videos and planar unfolded images collected by a borehole imaging instrument. The results from both sources were cross-validated to obtain an accurate distribution of fractures in roadway surrounding rock. Furthermore, the model provides guidelines for the injection and sealing of boreholes, increasing volume fraction of coal seam gas extraction.

How to use a few defect samples to complete the defect classification is a key challenge in the production of mobile phone screens. An attention-relation network for the mobile phone screen defect classification is proposed in this paper. The architecture of the attention-relation network contains two modules: a feature extract module and a feature metric module. Different from other few-shot models, an attention mechanism is applied to metric learning in our model to measure the distance between features, so as to pay attention to the correlation between features and suppress unwanted information. Besides, we combine dilated convolution and skip connection to extract more feature information for follow-up processing. We validate attention-relation network on the mobile phone screen defect dataset. The experimental results show that the classification accuracy of the attention-relation network is 0.9486 under the 5-way 1-shot training strategy and 0.9039 under the 5-way 5-shot setting. It achieves the excellent effect of classification for mobile phone screen defects and outperforms with dominant advantages.

Abstract Aiming at the problem that face images are easily interfered by occlusion factors in uncontrollable environments, and the complex structure of traditional convolutional neural networks leads to low expression recognition rates, slow network convergence speed, and long network training time, an improved lightweight convolutional neural network is proposed for facial expression recognition algorithm. First, the dilation convolution is introduced into the shortcut connection of the inverted residual structure in the MobileNetV3 network to expand the receptive field of the convolution kernel and reduce the loss of expression features. Then, the channel attention mechanism SENet in the network is replaced by the two-dimensional (channel and spatial) attention mechanism SimAM introduced without parameters to reduce the network parameters. Finally, in the normalization operation, the Batch Normalization of the backbone network is replaced with Group Normalization, which is stable at various batch sizes, to reduce errors caused by processing small batches of data. Experimental results on RaFD, FER2013, and FER2013Plus face expression data sets show that the network reduces the training times while maintaining network accuracy, improves network convergence speed, and has good convergence effects.

Accurate segmentation of skin lesions is essential for physicians to screen in dermoscopy images. However, they commonly face three main limitations: difficulty in accurately processing targets with coarse edges; frequent challenges in recovering detailed feature data; and a lack of adequate capability for the effective amalgamation of multi-scale features. To overcome these problems, we propose a skin lesion segmentation network (SFCC Net) that combines an attention mechanism and a redundancy reduction strategy. The initial step involved the design of a downsampling encoder and an encoder composed of Receptive Field (REFC) Blocks, aimed at supplementing lost details and extracting latent features. Subsequently, the Spatial-Frequency-Channel (SF) Block was employed to minimize feature redundancy and restore fine-grained information. To fully leverage previously learned features, an Up-sampling Convolution (UpC) Block was designed for information integration. The network’s performance was compared with state-of-the-art models on four public datasets. Experimental results demonstrate significant improvements in the network’s performance. On the ISIC datasets, the proposed network outperformed D-LKA Net by 4.19%, 0.19%, and 7.75% in F1, and by 2.14%, 0.51%, and 12.20% in IoU. The frame rate (FPS) of the proposed network when processing skin lesion images underscores its suitability for real-time image analysis. Additionally, the network’s generalization capability was validated on a lung dataset. [ABSTRACT FROM AUTHOR]

This study addresses the limitations of Transformer models in image feature extraction, particularly their lack of inductive bias for visual structures. Compared to Convolutional Neural Networks (CNNs), the Transformers are more sensitive to different hyperparameters of optimizers, which leads to a lack of stability and slow convergence. To tackle these challenges, we propose the Convolution-based Efficient Transformer Image Feature Extraction Network (CEFormer) as an enhancement of the Transformer architecture. Our model incorporates E-Attention, depthwise separable convolution, and dilated convolution to introduce crucial inductive biases, such as translation invariance, locality, and scale invariance, into the Transformer framework. Additionally, we implement a lightweight convolution module to process the input images, resulting in faster convergence and improved stability. This results in an efficient convolution combined Transformer image feature extraction network. Experimental results on the ImageNet1k Top-1 dataset demonstrate that the proposed network achieves better accuracy while maintaining high computational speed. It achieves up to 85.0% accuracy across various model sizes on image classification, outperforming various baseline models. When integrated into the Mask Region-Convolutional Neural Network (R-CNN) framework as a backbone network, CEFormer outperforms other models and achieves the highest mean Average Precision (mAP) scores. This research presents a significant advancement in Transformer-based image feature extraction, balancing performance and computational efficiency. [ABSTRACT FROM AUTHOR]

The automatic monitoring and detection of maritime targets hold paramount significance in safeguarding national sovereignty, ensuring maritime rights, and advancing national development. Among the principal means of maritime surveillance, infrared (IR) small ship detection technology stands out. However, due to their minimal pixel occupancy and lack of discernible color and texture information, IR small ships have persistently posed a formidable challenge in the realm of target detection. Additionally, the intricate maritime backgrounds often exacerbate the issue by inducing high false alarm rates. In an effort to surmount these challenges, this paper proposes a flexible convolutional network (FCNet), integrating dilated convolutions and deformable convolutions to achieve flexible variations in convolutional receptive fields. Firstly, a feature enhancement module (FEM) is devised to enhance input features by fusing standard convolutions with dilated convolutions, thereby obtaining precise feature representations. Subsequently, a context fusion module (CFM) is designed to integrate contextual information during the downsampling process, mitigating information loss. Furthermore, a semantic fusion module (SFM) is crafted to fuse shallow features with deep semantic information during the upsampling process. Additionally, squeeze-and-excitation (SE) blocks are incorporated during upsampling to bolster channel information. Experimental evaluations conducted on two datasets demonstrate that FCNet outperforms other algorithms in the detection of IR small ships on maritime surfaces. Moreover, to propel research in deep learning-based IR small ship detection on maritime surfaces, we introduce the IR small ship dataset (Maritime-SIRST). [ABSTRACT FROM AUTHOR]

Image denoising is regarded as an ill-posed problem in computer vision tasks that removes additive noise from imaging sensors. Recently, several convolution neural network-based image-denoising methods have achieved remarkable advances. However, it is difficult for a simple denoising network to recover aesthetically pleasing images owing to the complexity of image content. Therefore, this study proposes a multi-branch network to improve the performance of the denoising method. First, the proposed network is designed based on a conventional autoencoder to learn multi-level contextual features from input images. Subsequently, we integrate two modules into the network, including the Pyramid Context Module (PCM) and the Residual Bottleneck Attention Module (RBAM), to extract salient information for the training process. More specifically, PCM is applied at the beginning of the network to enlarge the receptive field and successfully address the loss of global information using dilated convolution. Meanwhile, RBAM is inserted into the middle of the encoder and decoder to eliminate degraded features and reduce undesired artifacts. Finally, extensive experimental results prove the superiority of the proposed method over state-of-the-art deep-learning methods in terms of objective and subjective performances. [ABSTRACT FROM AUTHOR]

Lung nodules occupy fewer pixels in CT (Computed Tomography), which brings great difficulty to detection. In view of small target detection of lung nodule, this study proposes YOLOV3 (You Only Look Once version 3) lung nodule detection algorithm based on coordinate attention. The backbone network adopts the improved YOLOv3, reduces the number of residual blocks and introduces the dilated convolution module to sense context information around the target. In the feature utilization, coordinate attention is introduced to capture the position, direction and cross channel information, so as to locate lung nodules accurately. The loss function of YOLOv3 is improved, the boundary box is modeled as Gaussian distribution. Wasserstein distance is used to calculate the similarity between the two distributions instead of IoU (Intersection over Union), so as to improve the sensitivity of the target scale. The results on LUNA16 show that the average precision is 89.96% and the sensitivity is 95.37%. Compared with mainstream target detection algorithms, the precision and sensitivity of the proposed method are improved by 11.33% and 9.03%, respectively. [ABSTRACT FROM AUTHOR]

With the successful application and breakthrough of deep learning technology in image segmentation, there has been continuous development in the field of seismic facies interpretation using convolutional neural networks. These intelligent and automated methods significantly reduce manual labor, particularly in the laborious task of manually labeling seismic facies. However, the extensive demand for training data imposes limitations on their wider application. To overcome this challenge, we adopt the UNet architecture as the foundational network structure for seismic facies classification, which has demonstrated effective segmentation results even with small-sample training data. Additionally, we integrate spatial pyramid pooling and dilated convolution modules into the network architecture to enhance the perception of spatial information across a broader range. The seismic facies classification test on the public data from the F3 block verifies the superior performance of our proposed improved network structure in delineating seismic facies boundaries. Comparative analysis against the traditional UNet model reveals that our method achieves more accurate predictive classification results, as evidenced by various evaluation metrics for image segmentation. Obviously, the classification accuracy reaches an impressive 96%. Furthermore, the results of seismic facies classification in the seismic slice dimension provide further confirmation of the superior performance of our proposed method, which accurately defines the range of different seismic facies. This approach holds significant potential for analyzing geological patterns and extracting valuable depositional information. [ABSTRACT FROM AUTHOR]

Copyright of Chinese Journal of Liquid Crystal & Displays is the property of Chinese Journal of Liquid Crystal & Displays and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

UAVs are commonly used in forest fire detection, but the captured fire images often suffer from blurring due to the rapid motion between the airborne camera and the fire target. In this study, a multi-input, multi-output U-Net architecture that combines spatial domain and frequency domain information is proposed for image deblurring. The architecture includes a multi-branch dilated convolution attention residual module in the encoder to enhance receptive fields and address local features and texture detail limitations. A feature-fusion module integrating spatial frequency domains is also included in the skip connection structure to reduce feature loss and enhance deblurring performance. Additionally, a multi-channel convolution attention residual module in the decoders improves the reconstruction of local and contextual information. A weighted loss function is utilized to enhance network stability and generalization. Experimental results demonstrate that the proposed model outperforms popular models in terms of subjective perception and quantitative evaluation, achieving a PSNR of 32.26 dB, SSIM of 0.955, LGF of 10.93, and SMD of 34.31 on the self-built forest fire datasets and reaching 86% of the optimal PSNR and 87% of the optimal SSIM. In experiments without reference images, the model performs well in terms of LGF and SMD. The results obtained by this model are superior to the currently popular SRN and MPRNet models. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Graphics is the property of Journal of Graphics Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Recent work has demonstrated the Transformer model is effective for computer vision tasks. However, the global self-attention mechanism utilized in Transformer models does not adequately consider the local structure and details of images, which may result in the loss of information and local details, causing decreased estimation accuracy in gaze estimation tasks when compared to convolution or sequential stacking methods. To address this issue, we propose a parallel CNNs-Transformer aggregation network (CTA-Net) for gaze estimation, which fully leverages the advantages of the Transformer model in modeling global context while the convolutional neural networks (CNNs) model in retaining local details. Specifically, Transformer and ResNet are deployed to extract facial and eye information, respectively. Additionally, an attention cross fusion (ACFusion) Block is embedded with CNN branch, which decomposes features in space and channels to supplement lost features, suppress noise, and help extract eye features more effectively. Finally, a dual-feature aggregation (DFA) module is proposed to effectively fuse the output features of both branches with the help feature a selection mechanism and a residual structure. Experimental results on the MPIIGaze and Gaze360 datasets demonstrate that our CTA-Net achieves state-of-the-art results. [ABSTRACT FROM AUTHOR]

Classical YOLOv4 object detector transcends some famous object detectors in speed and accuracy. However, despite its superior performance, it still has some limitations such as the insufficient for extracting the feature. Therefore, we propose an improved YOLOv4 method for object detection in this paper. Specifically, we introduce a dilated coordinate attention module for improving the accuracy of object detection, which combine the dilated convolutional neural network with coordinate attention. At the same time, the multi-scale training strategy is also introduced to enhance the performance of object detection via using the training data with different scales. Experiments on PASCAL VOC2007 and VOC2012 datasets demonstrate that our proposed improved YOLOv4 object detector is superior to other state of the art object detection detector. Specially, its performance is better than the traditional YOLOv4, which is 1.84% and 1.91% higher on the mean average precision (mAP) for the two datasets, respectively. [ABSTRACT FROM AUTHOR]

To address the various challenges in aluminum surface defect detection, such as multiscale intricacies, sensitivity to lighting variations, occlusion, and noise, this study proposes the AluDef-ClassNet model. Firstly, a Gaussian difference pyramid is utilized to capture multiscale image features. Secondly, a self-attention mechanism is introduced to enhance feature representation. Additionally, an improved residual network structure incorporating dilated convolutions is adopted to increase the receptive field, thereby enhancing the network's ability to learn from extensive information. A small-scale dataset of high-quality aluminum surface defect images is acquired using a CCD camera. To better tackle the challenges in surface defect detection, advanced deep learning techniques and data augmentation strategies are employed. To address the difficulty of data labeling, a transfer learning approach based on fine-tuning is utilized, leveraging prior knowledge to enhance the efficiency and accuracy of model training. In dataset testing, our model achieved a classification accuracy of 97.6%, demonstrating significant advantages over other classification models. [ABSTRACT FROM AUTHOR]

The vast volumes of videos produced daily require highly efficient measures to ensure that key information is reported for effective review and storage, which leads to the popularity of video summarization techniques. Deep learning has shown its advantages in video summarization, especially convolutional neural network, which are effective in extracting features for video summarization. However, the deep network layers and the limited range of temporal dependence make it challenging to deploy the network and thus affect the accuracy of identifying important video frames. To tackle these issues, we present a knowledge distillation-based attentive network (KDAN) for supervised video summarization in this paper. The proposed method separates the full convolutional network from the attention mechanism based on the idea of education and learning processes in biology and uses a full convolutional network as a teacher network to guide the learning of the student network consisting of an attention mechanism. The obtained lightweight network considers the knowledge learned from both networks, thus solving the problems of explosion in the number of participants and slow training. We have conducted experiments on two widely used benchmarks SumMe and TVSum. DANtea achieves F-scores 53.09 and 60.30, and DAN achieves F-scores 51.26 and 61.55 in Canonical settings on the SumMe and TVSum datasets, respectively. Experiments on two public benchmarks SumMe and TVSum demonstrate the effectiveness and superiority of the proposed network over existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The presence of small lesions is an important marker for determining whether a patient will develop malignant tumors. Clinical practitioners could easily overlook the presence of small lesions, meaning automated approaches are essential for screening test results. The use of deep learning-based detectors for this purpose has so far been suboptimal as small lesions easily lose the spatial information during the convolution operation, resulting in unsatisfactory detection accuracy and limited application in clinical decision making. In this paper, we propose a Cyclic Pyramid-based Small lesion detection Network (CPSNet), which iteratively enhances the features in the parallel layer of the Feature Parallel Network (FPN), the features learned in the loop are fused again with the initial FPN to compensate for the inadequacy problem in the initial training. In addition, we propose an aggregated dilation block (ADB) to capture small variations at different scales and a global attention block (GAB) to adaptively recalibrate the channel-based feature responses while focusing on the target spatial information and highlighting the most relevant feature channels. Extensive experiments on eight organs included in the DeepLesion dataset show that our method has a high detection accuracy(mAP=60.4) and a high overall sensitivity(80.5%), which is superior to the state-of-art methods. [ABSTRACT FROM AUTHOR]

The development of deep learning greatly promotes the progress of speaker verification (SV). Studies show that both convolutional neural networks (CNNs) and dilated time-delay neural networks (TDNNs) achieve advanced performance in text-independent SV, due to their ability to sufficiently extract the local feature and the temporal contextual information, respectively. Also, the combination of the above two has achieved better results. However, we found a serious gridding effect when we apply the 1D-Res2Net-based dilated TDNN proposed in ECAPA-TDNN for SV, which indicates discontinuity and local information losses of frame-level features. To achieve high-resolution process for speaker embedding, we improve the CNN–TDNN structure with proposed repeated multi-scale feature fusions. Through the proposed structure, we can effectively improve the channel utilization of TDNN and achieve higher performance under the same TDNN channel. And, unlike previous studies that have all converted CNN features to TDNN features directly, we also studied the latent space transformation between CNN and TDNN to achieve efficient conversion. Our best method obtains 0.72 EER and 0.0672 MinDCF on VoxCeleb-O test set, and the proposed method performs better in cross-domain SV without additional parameters and computational complexity. [ABSTRACT FROM AUTHOR]

Copyright of Cotton Textile Technology is the property of Cotton Textile Technology Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Mobile phone backplanes are an important part of mobile phones, and are often affected by a wide range of factors during the manufacturing process, resulting in defects of various scales and similar backgrounds. Therefore, accurately identifying these defects is crucial for improving mobile phone quality. To address this challenge, this paper proposes a multi-scale and dynamic attention fusion UNet (MDAF-UNet) model. The model innovatively combines normal convolution with dilated convolution. This allows the model to capture subtle features of defects and to perceive a larger range of feature variations. Moreover, an improved attention mechanism is introduced in this paper. It fuses channel attention and spatial attention, and dynamically adjusts the feature fusion strategy with learnable weights. This allows the model to increase the attention of important features and improve the effectiveness of feature representation. Experimental results on a publicly available dataset show that the MDAF-UNet model achieves 66.9% Mean Intersection over Union (MIoU), outperforming other state-of-the-art models. This result provides an effective solution to the mobile phone backplane defect segmentation problem. [ABSTRACT FROM AUTHOR]

Deep learning technologies have led to substantial breakthroughs in the field of biomedical image analysis. Accurate brain tumor segmentation is an essential aspect of treatment planning. Radiologists agree that manual segmentation is a difficult and time-consuming task that frequently causes delays in the diagnosing process. While U-Net-based methods have been widely used for brain tumor segmentation, many challenges persist, particularly when dealing with tumors of varying sizes, locations, and shapes. Additionally, segmenting tumor regions with structures requires a comprehensive model, which can increase computational complexity and potentially cause gradient vanishing issues. This study presents a novel method called 3D Guided Attention-based deep Inception Residual U-Net (GAIR-U-Net) to address these challenges. This model combines attention mechanisms, an inception module, and residual blocks with dilated convolution to enhance feature representation and spatial context understanding. The backbone of the model is the U-Net model, which leverages the power of inception and residual connections to capture intricate patterns and hierarchical features while expanding the model’s width in three-dimensional space without significantly increasing computational complexity. The attention mechanisms play a role in focusing on important regions and areas while downgrading irrelevant details. The dilated convolutions in the network help in learning both local and global information, improving accuracy and adaptability in segmenting tumors. All the experiments in this study were carried out on multimodal MRI scans that include (T1-weighted, T1-ce, T2-weighted, and FLAIR sequences) from the BraTS 2020 dataset. The presented model is trained and tested on the same dataset, which exhibited promising performance compared to previous methods. On the BraTS 2020 validation dataset, the proposed model obtained a dice score of 0.8796, 0.8634, and 0.8441 for whole tumor (WT), tumor core (TC), and enhancing tumor (ET), respectively. These results demonstrate the model’s efficacy in precisely segmenting brain tumors across various modalities. Comparative analyses underscore the model’s versatility in handling tumor shape variations, size, and location, making it a promising solution for clinical applications.