Long-range Dependence Research Articles

2S-SGCN: A two-stage stratified graph convolutional network model for facial landmark detection on 3D data

Facial Landmark Detection (FLD) algorithms play a crucial role in numerous computer vision applications, particularly in tasks such as face recognition, head pose estimation, and facial expression analysis. While FLD on images has long been the focus, the emergence of 3D data has led to a surge of interest in FLD on it due to its potential applications in various fields, including medical research. However, automating FLD in this context presents significant challenges, such as selecting suitable network architectures, refining outputs for precise landmark localization and optimizing computational efficiency. In response, this paper presents a novel approach, the 2-Stage Stratified Graph Convolutional Network (2S-SGCN), which addresses these challenges comprehensively. The first stage aims to detect landmark regions using heatmap regression, which leverages both local and long-range dependencies through a stratified approach. In the second stage, 3D landmarks are precisely determined using a new post-processing technique, namely MSE-over-mesh. 2S-SGCN ensures both efficiency and suitability for resource-constrained devices. Experimental results on 3D scans from the public Facescape and Headspace datasets, as well as on point clouds derived from FLAME meshes collected in the DAD-3DHeads dataset, demonstrate that the proposed method achieves state-of-the-art performance across various conditions. Source code is accessible at https://github.com/gfacchi-dev/CVIU-2S-SGCN.

PFEI-Net: A profound feature exploration and interaction network for ceramic substrate surface defect detection

Ceramic substrates serve as the foundational material for numerous electronic devices, and their surface quality directly affects performance and longevity. Therefore, surface defect detection on ceramic substrates is an indispensable task during the manufacturing process. Ceramic substrate surface defects exhibit low contrast, along with intraclass difference and interclass similarity, posing substantial challenges for automated visual inspection. Existing deep learning-based methods utilize carefully crafted backbone networks and feature pyramid structures that perceive multilevel information to deal with the above challenges. However, these methods still suffer from insufficient extraction of discriminative features and low effectiveness in feature fusion, thereby hindering the detection performance. To address these issues, we propose the profound feature exploration and interaction network (PFEI-Net). Initially, we propose the discriminative feature mining backbone (DFM-backbone) to progressively explore the effective discriminative features. In the DFM-backbone, the global information grated perception module is devised to construct long-range feature dependencies and aggregate rich features, whereas the contextual semantic flexible extraction module is designed to meticulously perceive contextual semantic information in a targeted manner, enhancing the comprehensive representation of the discriminative features. Then, we propose the multipath semantic interaction guided-feature pyramid network (MSIG-FPN) to facilitate the fusion and interaction of the valuable information. In the MSIG-FPN, the hierarchical refocus–reaggregation module is constructed to refocus on task-relevant features and provide reliable guidance for the network, whereas the cross-stage semantic deep fusion module is designed to integrate multi-level contextual information deeply across stages and is employed to construct the semantic complementary paths, thereby preventing the loss of crucial features. Finally, built upon the backbone of DFM-backbone and MSIG-FPN, we establish the PFEI-Net, achieving accurate detection of surface defects on ceramic substrates. Experimental results highlight that the PFEI-Net achieves excellent performance with a remarkable 89.3 % mean average precision (mAP) and an inference time of 26.1 ms, which offers a promising and viable solution for automated surface defect detection of ceramic substrate.

HTD-TS3: Weakly Supervised Hyperspectral Target Detection Based on Transformer via Spectral-Spatial Similarity.

As an advanced technique in remote sensing, hyperspectral target detection (HTD) is widely concerned in civilian and military applications. However, the limitation of prior and heterogeneous backgrounds makes HTD models sensitive to data corruption under various interference from the environment. In this article, a novel united HTD framework based on the concept of transformer is proposed to extract [HTD based on transformer via spectral-spatial similarity (HTD-TS3)] under weak supervision, which opens up more flexible ways to study HTD. For the first time, the transformer mechanism is introduced into the HTD task to extract spectral and spatial features in a unified optimization procedure. By modeling long-range dependence among spectra, it realizes spectral-spatial joint inference based on long-range context, which addresses the issues of insufficient utilization of spatial information. To provide samples for weakly supervised learning (WSL), the coarse sample selection and spectral sequence construction in an efficient way are proposed, which makes full use of limited prior information. Finally, an exponential constrained nonlinear function is adopted to acquire pixel-level prediction via combining discriminative spectral-spatial features and coarse spatial information. Experiments on real hyperspectral images (HSIs) captured by different sensors at various scenes verify the effectiveness and efficiency of HTD-TS3.

A frequency channel-attention based vision Transformer method for bearing fault identification across different working conditions

Fault identification of rolling bearings plays a crucial role in maintaining the efficient and stable operation of equipment. Although traditional fault identification methods have made certain progress, they still lack in model feature extraction capabilities and generalization ability. In this paper, a frequency channel-attention based vision Transformer method is proposed for rolling bearings intelligent fault identification. Using frequency domain channel-attention mechanism, the proposed method is able to preserve fundamental fault information and integrate the frequency characteristics of the vibration signals. The proposed method also leverages the inherent self-attention mechanism of vision Transformer to recognize long-range dependencies within the signal data. This integration of attention not only enhances the model’s sensitivity to signal frequency characteristics but also enables the visualization of the attention mechanism, thereby increasing the model’s interpretability. Additionally, a shift linear layer is proposed to reduce the model’s computational demands while maintaining its robust feature extraction capabilities. This proposed method directly uses the collected vibration raw signals to achieve precise fault identification of rolling bearings, and experimental validation on two datasets demonstrates the model’s diagnostic accuracy under across working conditions

A flexible 2.5D medical image segmentation approach with in-slice and cross-slice attention

Deep learning has become the de facto method for medical image segmentation, with 3D segmentation models excelling in capturing complex 3D structures and 2D models offering high computational efficiency. However, segmenting 2.5D images, characterized by high in-plane resolution but lower through-plane resolution, presents significant challenges. While applying 2D models to individual slices of a 2.5D image is feasible, it fails to capture the spatial relationships between slices. On the other hand, 3D models face challenges such as resolution inconsistencies in 2.5D images, along with computational complexity and susceptibility to overfitting when trained with limited data. In this context, 2.5D models, which capture inter-slice correlations using only 2D neural networks, emerge as a promising solution due to their reduced computational demand and simplicity in implementation. In this paper, we introduce CSA-Net, a flexible 2.5D segmentation model capable of processing 2.5D images with an arbitrary number of slices. CSA-Net features an innovative Cross-Slice Attention (CSA) module that effectively captures 3D spatial information by learning long-range dependencies between the center slice (for segmentation) and its neighboring slices. Moreover, CSA-Net utilizes the self-attention mechanism to learn correlations among pixels within the center slice. We evaluated CSA-Net on three 2.5D segmentation tasks: (1) multi-class brain MR image segmentation, (2) binary prostate MR image segmentation, and (3) multi-class prostate MR image segmentation. CSA-Net outperformed leading 2D, 2.5D, and 3D segmentation methods across all three tasks, achieving average Dice coefficients and HD95 values of 0.897 and 1.40 mm for the brain dataset, 0.921 and 1.06 mm for the prostate dataset, and 0.659 and 2.70 mm for the ProstateX dataset, demonstrating its efficacy and superiority. Our code is publicly available at: https://github.com/mirthAI/CSA-Net.

Exploring drug-target interaction prediction on cold-start scenarios via meta-learning-based graph transformer.

Predicting drug-target interaction (DTI) is of great importance for drug discovery and development. With the rapid development of biological and chemical technologies, computational methods for DTI prediction are becoming a promising approach. However, there are few solutions to the cold-start problem in DTI prediction scenarios, as these methods rely on existing interaction information to support their modeling. Consequently, they are unable to effectively predict DTIs for new drugs or targets with limited interaction data in the existing work. To this end, we propose a graph transformer method based on meta-learning named MGDTI (short for Meta-learning-based Graph Transformer for Drug-Target Interaction prediction) to fill this gap. Technically, we employ drug-drug similarity and target-target similarity as additional information to mitigate the scarcity of interactions. Besides, we trained MGDTI via meta-learning to be adaptive to cold-start tasks. Moreover, we employed graph transformer to prevent over-smoothing by capturing long-range dependencies. Extensive results on the benchmark dataset demonstrate that MGDTI is effective on DTI prediction under cold-start scenarios.

Graph-enhanced visual representations and question-guided dual attention for visual question answering

Visual Question Answering (VQA) has witnessed significant advancements recently, due to the application of deep learning in the field of vision-language research. Most current VQA models focus on merging visual and text features, but it is essential for these models to also consider the relationships between different parts of an image and use question information to highlight important features. This study proposes a method to enhance neighboring image region features and learn question-aware visual representations. First, we construct a region graph to represent spatial relationships between objects in the image. Then, graph convolutional networks (GCNs) are used to propagate information across neighboring regions, enriching each region’s feature representation by integrating contextual information. To capture long-range dependencies, the graph is enhanced with random walk with restart (RWR), enabling multi-hop reasoning across distant regions. Furthermore, a question-aware dual attention mechanism is introduced to further refine region features at both region and feature levels, ensuring that the model emphasizes key regions that are critical for answering the question. The enhanced region representations are then combined with the encoded question to predict an answer. Through extensive experiments on VQA benchmarks, the study demonstrates state-of-the-art performance by leveraging regional dependencies and question guidance. The integration of GCNs and random walks in the graph helps capture contextual information to focus visual attention selectively, resulting in significant improvements over existing methods on VQA 1.0 and VQA 2.0 benchmark datasets.

A method based on hybrid cross-multiscale spectral-spatial transformer network for hyperspectral and multispectral image fusion

Convolutional neural networks (CNNs) have made a significant contribution to hyperspectral image (HSI) generation. However, capturing long-range dependencies can be challenging with CNNs due to the limitations of their local receptive fields, which can lead to distortions in fused images. Transformers excel at capturing long-range dependencies but have limited capacity for handling fine details. Additionally, priorwork has often overlooked the extraction of global features during the image preprocessing stage, resulting in the potential loss of fine details. To address these issues, we propose a hybrid cross-multiscale spectral-spatial Transformer (HCMSST) that combines the advantages of CNNs in feature extraction and Transformers in capturing long-range dependencies. To fully extract and retain local and global information in the shallow feature extraction phase, the network incorporatesCNNs with a staggered cascade-dense residual block (SCDRB). This block employs staggered residuals to establish direct connections bothwithin and between branches and integrates attention modules to enhance the response to important features. This approach facilitates unrestricted information exchange and fosters deeper feature representations. To address the limitationsof Transformer in processing fine details, we introduce multiscale spatial-spectral coding-decoding structures to obtain comprehensive spatial-spectral features, which are utilized to capture the long-range dependencies via the cross-multiscale spectral-spatial Transformer (CMSST). Further, the CMSST incorporates a cross-level dual-stream feature interaction strategy that integrates spatial and spectral features from different levels and then feeds the fused features back to their corresponding branches for information interaction. Experimental results indicate that the proposed HCMSST achieves superior performance compared to many state-of-the-art (SOTA) methods. Specifically, HCMSST reduces the ERGAS metric by 3.05% compared to the SOTA methods on the CAVE dataset, while on the Harvard dataset, it achieves a 2.69% reduction in ERGAS compared to the SOTA results.

Pan-Mamba: Effective pan-sharpening with state space model

Pan-sharpening involves integrating information from low-resolution multi-spectral and high-resolution panchromatic images to generate high-resolution multi-spectral counterparts. While recent advancements in the state space model, particularly the efficient long-range dependency modeling achieved by Mamba, have revolutionized computer vision community, its untapped potential in pan-sharpening motivates our exploration. Our contribution, Pan-Mamba, represents a novel pan-sharpening network that leverages the efficiency of the Mamba model in global information modeling. In Pan-Mamba, we customize two core components: channel swapping Mamba and cross-modal Mamba, strategically designed for efficient cross-modal information exchange and fusion. The former initiates a lightweight cross-modal interaction through the exchange of partial panchromatic and multi-spectral channels, while the latter facilities the information representation capability by exploiting inherent cross-modal relationships. Through extensive experiments across diverse datasets, our proposed approach surpasses state-of-the-art methods, showcasing superior fusion results in pan-sharpening. To the best of our knowledge, this work is the first attempt in exploring the potential of the Mamba model and establishes a new frontier in the pan-sharpening techniques. The source code is available at https://github.com/alexhe101/Pan-Mamba.

MIMO-Uformer: A Transformer-Based Image Deblurring Network for Vehicle Surveillance Scenarios

Motion blur is a common problem in the field of surveillance scenarios, and it obstructs the acquisition of valuable information. Thanks to the success of deep learning, a sequence of CNN-based architecture has been designed for image deblurring and has made great progress. As another type of neural network, transformers have exhibited powerful deep representation learning and impressive performance based on high-level vision tasks. Transformer-based networks leverage self-attention to capture the long-range dependencies in the data, yet the computational complexity is quadratic to the spatial resolution, which makes transformers infeasible for the restoration of high-resolution images. In this article, we propose an efficient transformer-based deblurring network, named MIMO-Uformer, for vehicle-surveillance scenarios. The distinct feature of the MIMO-Uformer is that the basic-window-based multi-head self-attention (W-MSA) of the Swin transformer is employed to reduce the computational complexity and then incorporated into a multi-input and multi-output U-shaped network (MIMO-UNet). The performance can benefit from the operation of multi-scale images by MIMO-UNet. However, most deblurring networks are designed for global blur, while local blur is more common under vehicle-surveillance scenarios since the motion blur is primarily caused by local moving vehicles. Based on this observation, we further propose an Intersection over Patch (IoP) factor and a supervised morphological loss to improve the performance based on local blur. Extensive experiments on a public and a self-established dataset are carried out to verify the effectiveness. As a result, the deblurring behavior based on PSNR is improved at least 0.21 dB based on GOPRO and 0.74 dB based on the self-established datasets compared to the existing benchmarks.

ACMamba: A State Space Model-Based Approach for Multi-Weather Degraded Image Restoration

In computer vision, eliminating the effects of adverse weather conditions such as rain, snow, and fog on images is a key research challenge. Existing studies primarily focus on image restoration for single weather types, while methods addressing image restoration under multiple combined weather conditions remain relatively scarce. Furthermore, current mainstream restoration networks, mostly based on Transformer and CNN architectures, struggle to achieve an effective balance between global receptive field and computational efficiency, limiting their performance in practical applications. This study proposes ACMamba, an end-to-end lightweight network based on selective state space models, aimed at achieving image restoration under multiple weather conditions using a unified set of parameters. Specifically, we design a novel Visual State Space Module (VSSM) and a Spatially Aware Feed-Forward Network (SAFN), which organically combine the local feature extraction capabilities of convolutions with the long-range dependency modeling capabilities of selective state space models (SSMs). This combination significantly improves computational efficiency while maintaining a global receptive field, enabling effective application of the Mamba architecture to multi-weather image restoration tasks. Comprehensive experiments demonstrate that our proposed approach significantly outperforms existing methods for both specific and multi-weather tasks across multiple benchmark datasets, showcasing its efficient long-range modeling potential in multi-weather image restoration tasks.

TTMGNet: Tree Topology Mamba-Guided Network Collaborative Hierarchical Incremental Aggregation for Change Detection

Change detection (CD) identifies surface changes by analyzing bi-temporal remote sensing (RS) images of the same region and is essential for effective urban planning, ensuring the optimal allocation of resources, and supporting disaster management efforts. However, deep-learning-based CD methods struggle with background noise and pseudo-changes due to local receptive field limitations or computing resource constraints, which limits long-range dependency capture and feature integration, normally resulting in fragmented detections and high false positive rates. To address these challenges, we propose a tree topology Mamba-guided network (TTMGNet) based on Mamba architecture, which combines the Mamba architecture for effectively capturing global features, a unique tree topology structure for retaining fine local details, and a hierarchical feature fusion mechanism that enhances multi-scale feature integration and robustness against noise. Specifically, the a Tree Topology Mamba Feature Extractor (TTMFE) leverages the similarity of pixels to generate minimum spanning tree (MST) topology sequences, guiding information aggregation and transmission. This approach utilizes a Tree Topology State Space Model (TTSSM) to embed spatial and positional information while preserving the global feature extraction capability, thereby retaining local features. Subsequently, the Hierarchical Incremental Aggregation Module is utilized to gradually align and merge features from deep to shallow layers to facilitate hierarchical feature integration. Through residual connections and cross-channel attention (CCA), HIAM enhances the interaction between neighboring feature maps, ensuring that critical features are retained and effectively utilized during the fusion process, thereby enabling more accurate detection results in CD. The proposed TTMGNet achieved F1 scores of 92.31% on LEVIR-CD, 90.94% on WHU-CD, and 77.25% on CL-CD, outperforming current mainstream methods in suppressing the impact of background noise and pseudo-change and more accurately identifying change regions.

A Feature-Driven Inception Dilated Network for Infrared Image Super-Resolution Reconstruction

Image super-resolution (SR) algorithms based on deep learning yield good visual performances on visible images. Due to the blurred edges and low contrast of infrared (IR) images, methods transferred directly from visible images to IR images have a poor performance and ignore the demands of downstream detection tasks. Therefore, an Inception Dilated Super-Resolution (IDSR) network with multiple branches is proposed. A dilated convolutional branch captures high-frequency information to reconstruct edge details, while a non-local operation branch captures long-range dependencies between any two positions to maintain the global structure. Furthermore, deformable convolution is utilized to fuse features extracted from different branches, enabling adaptation to targets of various shapes. To enhance the detection performance of low-resolution (LR) images, we crop the images into patches based on target labels before feeding them to the network. This allows the network to focus on learning the reconstruction of the target areas only, reducing the interference of background areas in the target areas’ reconstruction. Additionally, a feature-driven module is cascaded at the end of the IDSR network to guide the high-resolution (HR) image reconstruction with feature prior information from a detection backbone. This method has been tested on the FLIR Thermal Dataset and the M3FD Dataset and compared with five mainstream SR algorithms. The final results demonstrate that our method effectively maintains image texture details. More importantly, our method achieves 80.55% mAP, outperforming other methods on FLIR Dataset detection accuracy, and with 74.7% mAP outperforms other methods on M3FD Dataset detection accuracy.

A spatial hierarchical network learning framework for drug repositioning allowing interpretation from macro to micro scale

Biomedical network learning offers fresh prospects for expediting drug repositioning. However, traditional network architectures struggle to quantify the relationship between micro-scale drug spatial structures and corresponding macro-scale biomedical networks, limiting their ability to capture key pharmacological properties and complex biomedical information crucial for drug screening and therapeutic discovery. Moreover, challenges such as difficulty in capturing long-range dependencies hinder current network-based approaches. To address these limitations, we introduce the Spatial Hierarchical Network, modeling molecular 3D structures and biological associations into a unified network. We propose an end-to-end framework, SpHN-VDA, integrating spatial hierarchical information through triple attention mechanisms to enhance machine understanding of molecular functionality and improve the accuracy of virus-drug association identification. SpHN-VDA outperforms leading models across three datasets, particularly excelling in out-of-distribution and cold-start scenarios. It also exhibits enhanced robustness against data perturbation, ranging from 20% to 40%. It accurately identifies critical motifs for binding sites, even without protein residue annotations. Leveraging reliability of SpHN-VDA, we have identified 25 potential candidate drugs through gene expression analysis and CMap. Molecular docking experiments with the SARS-CoV-2 spike protein further corroborate the predictions. This research highlights the broad potential of SpHN-VDA to enhance drug repositioning and identify effective treatments for various diseases.

Efficient Metal Corrosion Area Detection Model Combining Convolution and Transformer

In the context of rapid industrialization, efficiently detecting metal corrosion areas has become a critical task in preventing material damage. Unlike conventional semantic segmentation targets, metal corrosion characteristics vary significantly in color, texture, and size. Traditional image segmentation methods need improvement in scenarios involving occlusions, shadows, and defects. This paper proposes a convolution and sequence encoding combined network, MCD-Net, for metal corrosion area segmentation. First, a visual Transformer sequence encoder is introduced into the convolutional encoder–decoder network to enhance global information processing capabilities and establish long-range feature dependencies. A feature fusion method based on an attention module is proposed to enhance the model’s ability to recognize corrosion boundaries, thereby enhancing segmentation accuracy and model robustness. Finally, in the model’s decoding stage, a score-based multi-scale feature enhancement method is employed to emphasize significant features in the corrosion areas. Experimental results indicate that this method attained an F1 score of 84.53% on a public corrosion dataset, demonstrating the model’s deeper understanding and reasoning capabilities for shadow and defect features, as well as excellent noise resistance performance.

Multi-Modal LiDAR Point Cloud Semantic Segmentation with Salience Refinement and Boundary Perception

Point cloud segmentation is essential for scene understanding, which provides advanced information for many applications, such as autonomous driving, robots, and virtual reality. To improve the accuracy and robustness of point cloud segmentation, many researchers have attempted to fuze camera images to complement the color and texture information. The common fusion strategy is the combination of convolutional operations with concatenation, element-wise addition or element-wise multiplication. However, conventional convolutional operators tend to confine the fusion of modal features within their receptive fields, which can be incomplete and limited. In addition, the inability of encoder–decoder segmentation networks to explicitly perceive segmentation boundary information results in semantic ambiguity and classification errors at object edges. These errors are further amplified in point cloud segmentation tasks, significantly affecting the accuracy of point cloud segmentation. To address the above issues, we propose a novel self-attention multi-modal fusion semantic segmentation network for point cloud semantic segmentation. Firstly, to effectively fuze different modal features, we propose a self-cross fusion module (SCF), which models long-range modality dependencies and transfers complementary image information to the point cloud to fully leverage the modality-specific advantages. Secondly, we design the salience refinement module (SR), which calculates the importance of channels in the feature maps and global descriptors to enhance the representation capability of salient modal features. Finally, we propose the local-aware anisotropy loss measure the element-level importance in the data and explicitly provide boundary information for the model, which alleviates the inherent semantic ambiguity problem in segmentation networks. Extensive experiments on two benchmark datasets demonstrate that our proposed method surpasses current state-of-the-art methods.

A newton interpolation network for smoke semantic segmentation

Smoke has large variances of visual appearances that are very adverse to visual segmentation. Furthermore, its semi-transparency often produces highly complicated mixtures of smoke and backgrounds. These factors lead to great difficulties in labelling and segmenting smoke regions. To improve accuracy of smoke segmentation, we propose a Newton Interpolation Network (NINet) for visual smoke semantic segmentation. Unlike simply concatenating or point-wisely adding multi-scale encoded feature maps for information fusion or re-usage, we design a Newton Interpolation Module (NIM) to extract structured information by analyzing the feature values in the same position but from encoded feature maps with different scales. Interpolated features by our NIM contain long-range dependency and semantic structures across different levels, but traditional fusion of multi-scale feature maps cannot model intrinsic structures embedded in these maps. To obtain multi-scale structured information, we repeatedly use the proposed NIM at different levels of the decoding stages. In addition, we use more encoded feature maps to construct a higher order Newton interpolation polynomial for extracting higher order information. Extensive experiments validate that our method significantly outperforms existing state-of-the-art algorithms on virtual and real smoke datasets, and ablation experiments also validate the effectiveness of our NIMs.

CNN-Informer: A hybrid deep learning model for seizure detection on long-term EEG

Timely detecting epileptic seizures can significantly reduce accidental injuries of epilepsy patients and offer a novel intervention approach to improve their quality of life. Investigation on seizure detection based on deep learning models has achieved great success. However, there still remain challenging issues, such as the high computational complexity of the models and overfitting caused by the scarce availability of ictal electroencephalogram (EEG) signals for training. Therefore, we propose a novel end-to-end automatic seizure detection model named CNN-Informer, which leverages the capability of Convolutional Neural Network (CNN) to extract EEG local features of multi-channel EEGs, and the low computational complexity and memory usage ability of the Informer to capture the long-range dependencies. In view of the existence of various artifacts in long-term EEGs, we filter those raw EEGs using Discrete Wavelet Transform (DWT) before feeding them into the proposed CNN-Informer model for feature extraction and classification. Post-processing operations are further employed to achieve the final detection results. Our method is extensively evaluated on the CHB-MIT dataset and SH-SDU dataset with both segment-based and event-based criteria. The experimental outcomes demonstrate the superiority of the proposed CNN-Informer model and its strong generalization ability across two EEG datasets. In addition, the lightweight architecture of CNN-Informer makes it suitable for real-time implementation.

Global induced local network for infrared: dim small target detection

Abstract It is challenging to detect infrared dim targets submerged in complicated backgrounds due to their small size and faint intensity. The previous attention-based detection networks frequently require global long-range dependence. Significant calculations are required to determine the target's sparse but meaningful position. To prevent wasting calculations on the background, this paper offers a detection network guided by global context for local feature learning, named GILNet (Global Induced Local Network). It designs a global location module (GLM) and a local feature interaction module (LFIM) to capture the global position and features of targets, respectively. More specifically, using global context interaction, the GLM finds the region that might contain dim small targets, that is, the coarse location. In the coarsely located regions, the LFIM further acquires feature information about targets. Next, we also design an eight-directional attention operation to obtain the contour information of targets in the low feature map. It is fused with the high feature map in the multi-directional feature fusion module (MFFM), which retains more semantic and spatial information about targets. Finally, quantitative and qualitative analysis show that the GILNet performs better than eight comparison methods on two public datasets.

Multi-label dental disorder diagnosis based on MobileNetV2 and swin transformer using bagging ensemble classifier

Dental disorders are common worldwide, causing pain or infections and limiting mouth opening, so dental conditions impact productivity, work capability, and quality of life. Manual detection and classification of oral diseases is time-consuming and requires dentists’ evaluation and examination. The dental disease detection and classification system based on machine learning and deep learning will aid in early dental disease diagnosis. Hence, this paper proposes a new diagnosis system for dental diseases using X-ray imaging. The framework includes a robust pre-processing phase that uses image normalization and adaptive histogram equalization to improve image quality and reduce variation. A dual-stream approach is used for feature extraction, utilizing the advantages of Swin Transformer for capturing long-range dependencies and global context and MobileNetV2 for effective local feature extraction. A thorough representation of dental anomalies is produced by fusing the extracted features. To obtain reliable and broadly applicable classification results, a bagging ensemble classifier is utilized in the end. We evaluate our model on a benchmark dental radiography dataset. The experimental results and comparisons show the superiority of the proposed system with 95.7% for precision, 95.4% for sensitivity, 95.7% for specificity, 95.5% for Dice similarity coefficient, and 95.6% for accuracy. The results demonstrate the effectiveness of our hybrid model integrating MoileNetv2 and Swin Transformer architectures, outperforming state-of-the-art techniques in classifying dental diseases using dental panoramic X-ray imaging. This framework presents a promising method for robustly and accurately diagnosing dental diseases automatically, which may help dentists plan treatments and identify dental diseases early on.