Regular inspection of bridge cracks is crucial to bridge maintenance and repair.The traditional manual crack detection methods are timeconsuming,dangerous and subjective.At the same time,for the existing mainstream vi...Regular inspection of bridge cracks is crucial to bridge maintenance and repair.The traditional manual crack detection methods are timeconsuming,dangerous and subjective.At the same time,for the existing mainstream vision-based automatic crack detection algorithms,it is challenging to detect fine cracks and balance the detection accuracy and speed.Therefore,this paper proposes a new bridge crack segmentationmethod based on parallel attention mechanism and multi-scale features fusion on top of the DeeplabV3+network framework.First,the improved lightweight MobileNetv2 network and dilated separable convolution are integrated into the original DeeplabV3+network to improve the original backbone network Xception and atrous spatial pyramid pooling(ASPP)module,respectively,dramatically reducing the number of parameters in the network and accelerates the training and prediction speed of the model.Moreover,we introduce the parallel attention mechanism into the encoding and decoding stages.The attention to the crack regions can be enhanced from the aspects of both channel and spatial parts and significantly suppress the interference of various noises.Finally,we further improve the detection performance of the model for fine cracks by introducing a multi-scale features fusion module.Our research results are validated on the self-made dataset.The experiments show that our method is more accurate than other methods.Its intersection of union(IoU)and F1-score(F1)are increased to 77.96%and 87.57%,respectively.In addition,the number of parameters is only 4.10M,which is much smaller than the original network;also,the frames per second(FPS)is increased to 15 frames/s.The results prove that the proposed method fits well the requirements of rapid and accurate detection of bridge cracks and is superior to other methods.展开更多
Convolutional neural networks(CNNs)-based medical image segmentation technologies have been widely used in medical image segmentation because of their strong representation and generalization abilities.However,due to ...Convolutional neural networks(CNNs)-based medical image segmentation technologies have been widely used in medical image segmentation because of their strong representation and generalization abilities.However,due to the inability to effectively capture global information from images,CNNs can easily lead to loss of contours and textures in segmentation results.Notice that the transformer model can effectively capture the properties of long-range dependencies in the image,and furthermore,combining the CNN and the transformer can effectively extract local details and global contextual features of the image.Motivated by this,we propose a multi-branch and multi-scale attention network(M2ANet)for medical image segmentation,whose architecture consists of three components.Specifically,in the first component,we construct an adaptive multi-branch patch module for parallel extraction of image features to reduce information loss caused by downsampling.In the second component,we apply residual block to the well-known convolutional block attention module to enhance the network’s ability to recognize important features of images and alleviate the phenomenon of gradient vanishing.In the third component,we design a multi-scale feature fusion module,in which we adopt adaptive average pooling and position encoding to enhance contextual features,and then multi-head attention is introduced to further enrich feature representation.Finally,we validate the effectiveness and feasibility of the proposed M2ANet method through comparative experiments on four benchmark medical image segmentation datasets,particularly in the context of preserving contours and textures.展开更多
The fusion of infrared and visible images should emphasize the salient targets in the infrared image while preserving the textural details of the visible images.To meet these requirements,an autoencoder-based method f...The fusion of infrared and visible images should emphasize the salient targets in the infrared image while preserving the textural details of the visible images.To meet these requirements,an autoencoder-based method for infrared and visible image fusion is proposed.The encoder designed according to the optimization objective consists of a base encoder and a detail encoder,which is used to extract low-frequency and high-frequency information from the image.This extraction may lead to some information not being captured,so a compensation encoder is proposed to supplement the missing information.Multi-scale decomposition is also employed to extract image features more comprehensively.The decoder combines low-frequency,high-frequency and supplementary information to obtain multi-scale features.Subsequently,the attention strategy and fusion module are introduced to perform multi-scale fusion for image reconstruction.Experimental results on three datasets show that the fused images generated by this network effectively retain salient targets while being more consistent with human visual perception.展开更多
Semantic segmentation plays a foundational role in biomedical image analysis, providing precise information about cellular, tissue, and organ structures in both biological and medical imaging modalities. Traditional a...Semantic segmentation plays a foundational role in biomedical image analysis, providing precise information about cellular, tissue, and organ structures in both biological and medical imaging modalities. Traditional approaches often fail in the face of challenges such as low contrast, morphological variability, and densely packed structures. Recent advancements in deep learning have transformed segmentation capabilities through the integration of fine-scale detail preservation, coarse-scale contextual modeling, and multi-scale feature fusion. This work provides a comprehensive analysis of state-of-the-art deep learning models, including U-Net variants, attention-based frameworks, and Transformer-integrated networks, highlighting innovations that improve accuracy, generalizability, and computational efficiency. Key architectural components such as convolution operations, shallow and deep blocks, skip connections, and hybrid encoders are examined for their roles in enhancing spatial representation and semantic consistency. We further discuss the importance of hierarchical and instance-aware segmentation and annotation in interpreting complex biological scenes and multiplexed medical images. By bridging methodological developments with diverse application domains, this paper outlines current trends and future directions for semantic segmentation, emphasizing its critical role in facilitating annotation, diagnosis, and discovery in biomedical research.展开更多
This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualiz...This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualization enhancement.However,fine structure registration of complex thoracoabdominal organs and large deformation registration caused by respiratory motion is challenging.To deal with this problem,we propose a 3D multi-scale attention VoxelMorph(MAVoxelMorph)registration network.To alleviate the large deformation problem,a multi-scale axial attention mechanism is utilized by using a residual dilated pyramid pooling for multi-scale feature extraction,and position-aware axial attention for long-distance dependencies between pixels capture.To further improve the large deformation and fine structure registration results,a multi-scale context channel attention mechanism is employed utilizing content information via adjacent encoding layers.Our method was evaluated on four public lung datasets(DIR-Lab dataset,Creatis dataset,Learn2Reg dataset,OASIS dataset)and a local dataset.Results proved that the proposed method achieved better registration performance than current state-of-the-art methods,especially in handling the registration of large deformations and fine structures.It also proved to be fast in 3D image registration,using about 1.5 s,and faster than most methods.Qualitative and quantitative assessments proved that the proposed MA-VoxelMorph has the potential to realize precise and fast tumor localization in clinical interventional surgeries.展开更多
Image captioning,the task of generating descriptive sentences for images,has advanced significantly with the integration of semantic information.However,traditional models still rely on static visual features that do ...Image captioning,the task of generating descriptive sentences for images,has advanced significantly with the integration of semantic information.However,traditional models still rely on static visual features that do not evolve with the changing linguistic context,which can hinder the ability to form meaningful connections between the image and the generated captions.This limitation often leads to captions that are less accurate or descriptive.In this paper,we propose a novel approach to enhance image captioning by introducing dynamic interactions where visual features continuously adapt to the evolving linguistic context.Our model strengthens the alignment between visual and linguistic elements,resulting in more coherent and contextually appropriate captions.Specifically,we introduce two innovative modules:the Visual Weighting Module(VWM)and the Enhanced Features Attention Module(EFAM).The VWM adjusts visual features using partial attention,enabling dynamic reweighting of the visual inputs,while the EFAM further refines these features to improve their relevance to the generated caption.By continuously adjusting visual features in response to the linguistic context,our model bridges the gap between static visual features and dynamic language generation.We demonstrate the effectiveness of our approach through experiments on the MS-COCO dataset,where our method outperforms state-of-the-art techniques in terms of caption quality and contextual relevance.Our results show that dynamic visual-linguistic alignment significantly enhances image captioning performance.展开更多
The application of image super-resolution(SR)has brought significant assistance in the medical field,aiding doctors to make more precise diagnoses.However,solely relying on a convolutional neural network(CNN)for image...The application of image super-resolution(SR)has brought significant assistance in the medical field,aiding doctors to make more precise diagnoses.However,solely relying on a convolutional neural network(CNN)for image SR may lead to issues such as blurry details and excessive smoothness.To address the limitations,we proposed an algorithm based on the generative adversarial network(GAN)framework.In the generator network,three different sizes of convolutions connected by a residual dense structure were used to extract detailed features,and an attention mechanism combined with dual channel and spatial information was applied to concentrate the computing power on crucial areas.In the discriminator network,using InstanceNorm to normalize tensors sped up the training process while retaining feature information.The experimental results demonstrate that our algorithm achieves higher peak signal-to-noise ratio(PSNR)and structural similarity index measure(SSIM)compared to other methods,resulting in an improved visual quality.展开更多
We propose a hierarchical multi-scale attention mechanism-based model in response to the low accuracy and inefficient manual classification of existing oceanic biological image classification methods. Firstly, the hie...We propose a hierarchical multi-scale attention mechanism-based model in response to the low accuracy and inefficient manual classification of existing oceanic biological image classification methods. Firstly, the hierarchical efficient multi-scale attention(H-EMA) module is designed for lightweight feature extraction, achieving outstanding performance at a relatively low cost. Secondly, an improved EfficientNetV2 block is used to integrate information from different scales better and enhance inter-layer message passing. Furthermore, introducing the convolutional block attention module(CBAM) enhances the model's perception of critical features, optimizing its generalization ability. Lastly, Focal Loss is introduced to adjust the weights of complex samples to address the issue of imbalanced categories in the dataset, further improving the model's performance. The model achieved 96.11% accuracy on the intertidal marine organism dataset of Nanji Islands and 84.78% accuracy on the CIFAR-100 dataset, demonstrating its strong generalization ability to meet the demands of oceanic biological image classification.展开更多
Nuclearmagnetic resonance imaging of breasts often presents complex backgrounds.Breast tumors exhibit varying sizes,uneven intensity,and indistinct boundaries.These characteristics can lead to challenges such as low a...Nuclearmagnetic resonance imaging of breasts often presents complex backgrounds.Breast tumors exhibit varying sizes,uneven intensity,and indistinct boundaries.These characteristics can lead to challenges such as low accuracy and incorrect segmentation during tumor segmentation.Thus,we propose a two-stage breast tumor segmentation method leveraging multi-scale features and boundary attention mechanisms.Initially,the breast region of interest is extracted to isolate the breast area from surrounding tissues and organs.Subsequently,we devise a fusion network incorporatingmulti-scale features and boundary attentionmechanisms for breast tumor segmentation.We incorporate multi-scale parallel dilated convolution modules into the network,enhancing its capability to segment tumors of various sizes through multi-scale convolution and novel fusion techniques.Additionally,attention and boundary detection modules are included to augment the network’s capacity to locate tumors by capturing nonlocal dependencies in both spatial and channel domains.Furthermore,a hybrid loss function with boundary weight is employed to address sample class imbalance issues and enhance the network’s boundary maintenance capability through additional loss.Themethod was evaluated using breast data from 207 patients at RuijinHospital,resulting in a 6.64%increase in Dice similarity coefficient compared to the benchmarkU-Net.Experimental results demonstrate the superiority of the method over other segmentation techniques,with fewer model parameters.展开更多
With the rapid expansion of drone applications,accurate detection of objects in aerial imagery has become crucial for intelligent transportation,urban management,and emergency rescue missions.However,existing methods ...With the rapid expansion of drone applications,accurate detection of objects in aerial imagery has become crucial for intelligent transportation,urban management,and emergency rescue missions.However,existing methods face numerous challenges in practical deployment,including scale variation handling,feature degradation,and complex backgrounds.To address these issues,we propose Edge-enhanced and Detail-Capturing You Only Look Once(EHDC-YOLO),a novel framework for object detection in Unmanned Aerial Vehicle(UAV)imagery.Based on the You Only Look Once version 11 nano(YOLOv11n)baseline,EHDC-YOLO systematically introduces several architectural enhancements:(1)a Multi-Scale Edge Enhancement(MSEE)module that leverages multi-scale pooling and edge information to enhance boundary feature extraction;(2)an Enhanced Feature Pyramid Network(EFPN)that integrates P2-level features with Cross Stage Partial(CSP)structures and OmniKernel convolutions for better fine-grained representation;and(3)Dynamic Head(DyHead)with multi-dimensional attention mechanisms for enhanced cross-scale modeling and perspective adaptability.Comprehensive experiments on the Vision meets Drones for Detection(VisDrone-DET)2019 dataset demonstrate that EHDC-YOLO achieves significant improvements,increasing mean Average Precision(mAP)@0.5 from 33.2%to 46.1%(an absolute improvement of 12.9 percentage points)and mAP@0.5:0.95 from 19.5%to 28.0%(an absolute improvement of 8.5 percentage points)compared with the YOLOv11n baseline,while maintaining a reasonable parameter count(2.81 M vs the baseline’s 2.58 M).Further ablation studies confirm the effectiveness of each proposed component,while visualization results highlight EHDC-YOLO’s superior performance in detecting objects and handling occlusions in complex drone scenarios.展开更多
Audio-visual scene classification(AVSC)poses a formidable challenge owing to the intricate spatial-temporal relationships exhibited by audio-visual signals,coupled with the complex spatial patterns of objects and text...Audio-visual scene classification(AVSC)poses a formidable challenge owing to the intricate spatial-temporal relationships exhibited by audio-visual signals,coupled with the complex spatial patterns of objects and textures found in visual images.The focus of recent studies has predominantly revolved around extracting features from diverse neural network structures,inadvertently neglecting the acquisition of semantically meaningful regions and crucial components within audio-visual data.The authors present a feature pyramid attention network(FPANet)for audio-visual scene understanding,which extracts semantically significant characteristics from audio-visual data.The authors’approach builds multi-scale hierarchical features of sound spectrograms and visual images using a feature pyramid representation and localises the semantically relevant regions with a feature pyramid attention module(FPAM).A dimension alignment(DA)strategy is employed to align feature maps from multiple layers,a pyramid spatial attention(PSA)to spatially locate essential regions,and a pyramid channel attention(PCA)to pinpoint significant temporal frames.Experiments on visual scene classification(VSC),audio scene classification(ASC),and AVSC tasks demonstrate that FPANet achieves performance on par with state-of-the-art(SOTA)approaches,with a 95.9 F1-score on the ADVANCE dataset and a relative improvement of 28.8%.Visualisation results show that FPANet can prioritise semantically meaningful areas in audio-visual signals.展开更多
Globally,diabetic retinopathy(DR)is the primary cause of blindness,affecting millions of people worldwide.This widespread impact underscores the critical need for reliable and precise diagnostic techniques to ensure p...Globally,diabetic retinopathy(DR)is the primary cause of blindness,affecting millions of people worldwide.This widespread impact underscores the critical need for reliable and precise diagnostic techniques to ensure prompt diagnosis and effective treatment.Deep learning-based automated diagnosis for diabetic retinopathy can facilitate early detection and treatment.However,traditional deep learning models that focus on local views often learn feature representations that are less discriminative at the semantic level.On the other hand,models that focus on global semantic-level information might overlook critical,subtle local pathological features.To address this issue,we propose an adaptive multi-scale feature fusion network called(AMSFuse),which can adaptively combine multi-scale global and local features without compromising their individual representation.Specifically,our model incorporates global features for extracting high-level contextual information from retinal images.Concurrently,local features capture fine-grained details,such as microaneurysms,hemorrhages,and exudates,which are critical for DR diagnosis.These global and local features are adaptively fused using a fusion block,followed by an Integrated Attention Mechanism(IAM)that refines the fused features by emphasizing relevant regions,thereby enhancing classification accuracy for DR classification.Our model achieves 86.3%accuracy on the APTOS dataset and 96.6%RFMiD,both of which are comparable to state-of-the-art methods.展开更多
To solve the problems of redundant feature information,the insignificant difference in feature representation,and low recognition accuracy of the fine-grained image,based on the ResNeXt50 model,an MSFResNet network mo...To solve the problems of redundant feature information,the insignificant difference in feature representation,and low recognition accuracy of the fine-grained image,based on the ResNeXt50 model,an MSFResNet network model is proposed by fusing multi-scale feature information.Firstly,a multi-scale feature extraction module is designed to obtain multi-scale information on feature images by using different scales of convolution kernels.Meanwhile,the channel attention mechanism is used to increase the global information acquisition of the network.Secondly,the feature images processed by the multi-scale feature extraction module are fused with the deep feature images through short links to guide the full learning of the network,thus reducing the loss of texture details of the deep network feature images,and improving network generalization ability and recognition accuracy.Finally,the validity of the MSFResNet model is verified using public datasets and applied to wild mushroom identification.Experimental results show that compared with ResNeXt50 network model,the accuracy of the MSFResNet model is improved by 6.01%on the FGVC-Aircraft common dataset.It achieves 99.13%classification accuracy on the wild mushroom dataset,which is 0.47%higher than ResNeXt50.Furthermore,the experimental results of the thermal map show that the MSFResNet model significantly reduces the interference of background information,making the network focus on the location of the main body of wild mushroom,which can effectively improve the accuracy of wild mushroom identification.展开更多
Accurate prediction of landslide displacement is crucial for effective early warning of landslide disasters.While most existing prediction methods focus on time-series forecasting for individual monitoring points,ther...Accurate prediction of landslide displacement is crucial for effective early warning of landslide disasters.While most existing prediction methods focus on time-series forecasting for individual monitoring points,there is limited research on the spatiotemporal characteristics of landslide deformation.This paper proposes a novel Multi-Relation Spatiotemporal Graph Residual Network with Multi-Level Feature Attention(MFA-MRSTGRN)that effectively improves the prediction performance of landslide displacement through spatiotemporal fusion.This model integrates internal seepage factors as data feature enhancements with external triggering factors,allowing for accurate capture of the complex spatiotemporal characteristics of landslide displacement and the construction of a multi-source heterogeneous dataset.The MFA-MRSTGRN model incorporates dynamic graph theory and four key modules:multilevel feature attention,temporal-residual decomposition,spatial multi-relational graph convolution,and spatiotemporal fusion prediction.This comprehensive approach enables the efficient analyses of multi-source heterogeneous datasets,facilitating adaptive exploration of the evolving multi-relational,multi-dimensional spatiotemporal complexities in landslides.When applying this model to predict the displacement of the Liangshuijing landslide,we demonstrate that the MFA-MRSTGRN model surpasses traditional models,such as random forest(RF),long short-term memory(LSTM),and spatial temporal graph convolutional networks(ST-GCN)models in terms of various evaluation metrics including mean absolute error(MAE=1.27 mm),root mean square error(RMSE=1.49 mm),mean absolute percentage error(MAPE=0.026),and R-squared(R^(2)=0.88).Furthermore,feature ablation experiments indicate that incorporating internal seepage factors improves the predictive performance of landslide displacement models.This research provides an advanced and reliable method for landslide displacement prediction.展开更多
In the robotic welding process with thick steel plates,laser vision sensors are widely used to profile the weld seam to implement automatic seam tracking.The weld seam profile extraction(WSPE)result is a crucial step ...In the robotic welding process with thick steel plates,laser vision sensors are widely used to profile the weld seam to implement automatic seam tracking.The weld seam profile extraction(WSPE)result is a crucial step for identifying the feature points of the extracted profile to guide the welding torch in real time.The visual information processing system may collapse when interference data points in the image survive during the phase of feature point identification,which results in low tracking accuracy and poor welding quality.This paper presents a visual attention featurebased method to extract the weld seam profile(WSP)from the strong arc background using clustering results.First,a binary image is obtained through the preprocessing stage.Second,all data points with a gray value 255 are clustered with the nearest neighborhood clustering algorithm.Third,a strategy is developed to discern one cluster belonging to the WSP from the appointed candidate clusters in each loop,and a scheme is proposed to extract the entire WSP using visual continuity.Compared with the previous methods the proposed method in this paper can extract more useful details of the WSP and has better stability in terms of removing the interference data.Considerable WSPE tests with butt joints and T-joints show the anti-interference ability of the proposed method,which contributes to smoothing the welding process and shows its practical value in robotic automated welding with thick steel plates.展开更多
Multi-label image classification is a challenging task due to the diverse sizes and complex backgrounds of objects in images.Obtaining class-specific precise representations at different scales is a key aspect of feat...Multi-label image classification is a challenging task due to the diverse sizes and complex backgrounds of objects in images.Obtaining class-specific precise representations at different scales is a key aspect of feature representation.However,existing methods often rely on the single-scale deep feature,neglecting shallow and deeper layer features,which poses challenges when predicting objects of varying scales within the same image.Although some studies have explored multi-scale features,they rarely address the flow of information between scales or efficiently obtain class-specific precise representations for features at different scales.To address these issues,we propose a two-stage,three-branch Transformer-based framework.The first stage incorporates multi-scale image feature extraction and hierarchical scale attention.This design enables the model to consider objects at various scales while enhancing the flow of information across different feature scales,improving the model’s generalization to diverse object scales.The second stage includes a global feature enhancement module and a region selection module.The global feature enhancement module strengthens interconnections between different image regions,mitigating the issue of incomplete represen-tations,while the region selection module models the cross-modal relationships between image features and labels.Together,these components enable the efficient acquisition of class-specific precise feature representations.Extensive experiments on public datasets,including COCO2014,VOC2007,and VOC2012,demonstrate the effectiveness of our proposed method.Our approach achieves consistent performance gains of 0.3%,0.4%,and 0.2%over state-of-the-art methods on the three datasets,respectively.These results validate the reliability and superiority of our approach for multi-label image classification.展开更多
Detecting abnormal cervical cells is crucial for early identification and timely treatment of cervical cancer.However,this task is challenging due to the morphological similarities between abnormal and normal cells an...Detecting abnormal cervical cells is crucial for early identification and timely treatment of cervical cancer.However,this task is challenging due to the morphological similarities between abnormal and normal cells and the significant variations in cell size.Pathologists often refer to surrounding cells to identify abnormalities.To emulate this slide examination behavior,this study proposes a Multi-Scale Feature Fusion Network(MSFF-Net)for detecting cervical abnormal cells.MSFF-Net employs a Cross-Scale Pooling Model(CSPM)to effectively capture diverse features and contextual information,ranging from local details to the overall structure.Additionally,a Multi-Scale Fusion Attention(MSFA)module is introduced to mitigate the impact of cell size variations by adaptively fusing local and global information at different scales.To handle the complex environment of cervical cell images,such as cell adhesion and overlapping,the Inner-CIoU loss function is utilized to more precisely measure the overlap between bounding boxes,thereby improving detection accuracy in such scenarios.Experimental results on the Comparison detector dataset demonstrate that MSFF-Net achieves a mean average precision(mAP)of 63.2%,outperforming state-of-the-art methods while maintaining a relatively small number of parameters(26.8 M).This study highlights the effectiveness of multi-scale feature fusion in enhancing the detection of cervical abnormal cells,contributing to more accurate and efficient cervical cancer screening.展开更多
The detection of Ochotona Curzoniae serves as a fundamental component for estimating the population size of this species and for analyzing the dynamics of its population fluctuations.In natural environments,the pixels...The detection of Ochotona Curzoniae serves as a fundamental component for estimating the population size of this species and for analyzing the dynamics of its population fluctuations.In natural environments,the pixels representing Ochotona Curzoniae constitute a small fraction of the total pixels,and their distinguishing features are often subtle,complicating the target detection process.To effectively extract the characteristics of these small targets,a feature fusion approach that utilizes up-sampling and channel integration from various layers within a CNN can significantly enhance the representation of target features,ultimately improving detection accuracy.However,the top-down fusion of features from different layers may lead to information duplication and semantic bias,resulting in redundancy and high-frequency noise.To address the challenges of information redundancy and high-frequency noise during the feature fusion process in CNN,we have developed a target detection model for Ochotona Curzoniae.This model is based on a spatial-channel reconfiguration convolutional(SCConv)attentional mechanism and feature fusion(FFBCA),integrated with the Faster R-CNN framework.It consists of a feature extraction network,an attention mechanism-based feature fusion module,and a jump residual connection fusion module.Initially,we designed a dual attention mechanism feature fusion module that employs spatial-channel reconstruction convolution.In the spatial dimension,the attention mechanism adopts a separation-reconstruction approach,calculating a weight matrix for the spatial information within the feature map through group normalization.This process directs the model to concentrate on feature information assigned varying weights,thereby reducing redundancy during feature fusion.In the channel dimension,the attention mechanism utilizes a partition-transpose-fusion method,segmenting the input feature map into high-noise and low-noise components based on the variance of the feature information.The high-noise segment is processed through a low-pass filter constructed from pointwise convolution(PWC)to eliminate some high-frequency noise,while the low-noise segment employs a bottleneck structure with global average pooling(GAP)to generate a weight matrix that emphasizes the significance of channel dimension feature information.This approach diminishes the model’s focus on low-weight feature information,thereby preserving low-frequency semantic information while reducing information redundancy.Furthermore,we have developed a novel feature extraction network,ResNeXt-S,by integrating the Sim attention mechanism into ResNeXt50.This configuration assigns three-dimensional attention weights to each position within the feature map,thereby enhancing the local feature information of small targets while reducing background noise.Finally,we constructed a jump residual connection fusion module to minimize the loss of high-level semantic information during the feature fusion process.Experiments on Ochotona Curzoniae target detection on the Ochotona Curzoniae dataset show that the detection accuracy of the model in this paper is 92.3%,which is higher than that of FSSD512(84.6%),TDFSSD512(81.3%),FPN(86.5%),FFBAM(88.5%),Faster R-CNN(89.6%),and SSD512(88.6%)detection accuracies.展开更多
Pest detection techniques are helpful in reducing the frequency and scale of pest outbreaks;however,their application in the actual agricultural production process is still challenging owing to the problems of intersp...Pest detection techniques are helpful in reducing the frequency and scale of pest outbreaks;however,their application in the actual agricultural production process is still challenging owing to the problems of interspecies similarity,multi-scale,and background complexity of pests.To address these problems,this study proposes an FD-YOLO pest target detection model.The FD-YOLO model uses a Fully Connected Feature Pyramid Network(FC-FPN)instead of a PANet in the neck,which can adaptively fuse multi-scale information so that the model can retain small-scale target features in the deep layer,enhance large-scale target features in the shallow layer,and enhance the multiplexing of effective features.A dual self-attention module(DSA)is then embedded in the C3 module of the neck,which captures the dependencies between the information in both spatial and channel dimensions,effectively enhancing global features.We selected 16 types of pests that widely damage field crops in the IP102 pest dataset,which were used as our dataset after data supplementation and enhancement.The experimental results showed that FD-YOLO’s mAP@0.5 improved by 6.8%compared to YOLOv5,reaching 82.6%and 19.1%–5%better than other state-of-the-art models.This method provides an effective new approach for detecting similar or multiscale pests in field crops.展开更多
基金This work was supported by the High-Tech Industry Science and Technology Innovation Leading Plan Project of Hunan Provincial under Grant 2020GK2026,author B.Y,http://kjt.hunan.gov.cn/.
文摘Regular inspection of bridge cracks is crucial to bridge maintenance and repair.The traditional manual crack detection methods are timeconsuming,dangerous and subjective.At the same time,for the existing mainstream vision-based automatic crack detection algorithms,it is challenging to detect fine cracks and balance the detection accuracy and speed.Therefore,this paper proposes a new bridge crack segmentationmethod based on parallel attention mechanism and multi-scale features fusion on top of the DeeplabV3+network framework.First,the improved lightweight MobileNetv2 network and dilated separable convolution are integrated into the original DeeplabV3+network to improve the original backbone network Xception and atrous spatial pyramid pooling(ASPP)module,respectively,dramatically reducing the number of parameters in the network and accelerates the training and prediction speed of the model.Moreover,we introduce the parallel attention mechanism into the encoding and decoding stages.The attention to the crack regions can be enhanced from the aspects of both channel and spatial parts and significantly suppress the interference of various noises.Finally,we further improve the detection performance of the model for fine cracks by introducing a multi-scale features fusion module.Our research results are validated on the self-made dataset.The experiments show that our method is more accurate than other methods.Its intersection of union(IoU)and F1-score(F1)are increased to 77.96%and 87.57%,respectively.In addition,the number of parameters is only 4.10M,which is much smaller than the original network;also,the frames per second(FPS)is increased to 15 frames/s.The results prove that the proposed method fits well the requirements of rapid and accurate detection of bridge cracks and is superior to other methods.
基金supported by the Natural Science Foundation of the Anhui Higher Education Institutions of China(Grant Nos.2023AH040149 and 2024AH051915)the Anhui Provincial Natural Science Foundation(Grant No.2208085MF168)+1 种基金the Science and Technology Innovation Tackle Plan Project of Maanshan(Grant No.2024RGZN001)the Scientific Research Fund Project of Anhui Medical University(Grant No.2023xkj122).
文摘Convolutional neural networks(CNNs)-based medical image segmentation technologies have been widely used in medical image segmentation because of their strong representation and generalization abilities.However,due to the inability to effectively capture global information from images,CNNs can easily lead to loss of contours and textures in segmentation results.Notice that the transformer model can effectively capture the properties of long-range dependencies in the image,and furthermore,combining the CNN and the transformer can effectively extract local details and global contextual features of the image.Motivated by this,we propose a multi-branch and multi-scale attention network(M2ANet)for medical image segmentation,whose architecture consists of three components.Specifically,in the first component,we construct an adaptive multi-branch patch module for parallel extraction of image features to reduce information loss caused by downsampling.In the second component,we apply residual block to the well-known convolutional block attention module to enhance the network’s ability to recognize important features of images and alleviate the phenomenon of gradient vanishing.In the third component,we design a multi-scale feature fusion module,in which we adopt adaptive average pooling and position encoding to enhance contextual features,and then multi-head attention is introduced to further enrich feature representation.Finally,we validate the effectiveness and feasibility of the proposed M2ANet method through comparative experiments on four benchmark medical image segmentation datasets,particularly in the context of preserving contours and textures.
基金Supported by the Henan Province Key Research and Development Project(231111211300)the Central Government of Henan Province Guides Local Science and Technology Development Funds(Z20231811005)+2 种基金Henan Province Key Research and Development Project(231111110100)Henan Provincial Outstanding Foreign Scientist Studio(GZS2024006)Henan Provincial Joint Fund for Scientific and Technological Research and Development Plan(Application and Overcoming Technical Barriers)(242103810028)。
文摘The fusion of infrared and visible images should emphasize the salient targets in the infrared image while preserving the textural details of the visible images.To meet these requirements,an autoencoder-based method for infrared and visible image fusion is proposed.The encoder designed according to the optimization objective consists of a base encoder and a detail encoder,which is used to extract low-frequency and high-frequency information from the image.This extraction may lead to some information not being captured,so a compensation encoder is proposed to supplement the missing information.Multi-scale decomposition is also employed to extract image features more comprehensively.The decoder combines low-frequency,high-frequency and supplementary information to obtain multi-scale features.Subsequently,the attention strategy and fusion module are introduced to perform multi-scale fusion for image reconstruction.Experimental results on three datasets show that the fused images generated by this network effectively retain salient targets while being more consistent with human visual perception.
基金Open Access funding provided by the National Institutes of Health(NIH)The funding for this project was provided by NCATS Intramural Fund.
文摘Semantic segmentation plays a foundational role in biomedical image analysis, providing precise information about cellular, tissue, and organ structures in both biological and medical imaging modalities. Traditional approaches often fail in the face of challenges such as low contrast, morphological variability, and densely packed structures. Recent advancements in deep learning have transformed segmentation capabilities through the integration of fine-scale detail preservation, coarse-scale contextual modeling, and multi-scale feature fusion. This work provides a comprehensive analysis of state-of-the-art deep learning models, including U-Net variants, attention-based frameworks, and Transformer-integrated networks, highlighting innovations that improve accuracy, generalizability, and computational efficiency. Key architectural components such as convolution operations, shallow and deep blocks, skip connections, and hybrid encoders are examined for their roles in enhancing spatial representation and semantic consistency. We further discuss the importance of hierarchical and instance-aware segmentation and annotation in interpreting complex biological scenes and multiplexed medical images. By bridging methodological developments with diverse application domains, this paper outlines current trends and future directions for semantic segmentation, emphasizing its critical role in facilitating annotation, diagnosis, and discovery in biomedical research.
基金supported in part by the National Natural Science Foundation of China[62301374]Hubei Provincial Natural Science Foundation of China[2022CFB804]+2 种基金Hubei Provincial Education Research Project[B2022057]the Youths Science Foundation of Wuhan Institute of Technology[K202240]the 15th Graduate Education Innovation Fund of Wuhan Institute of Technology[CX2023295].
文摘This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualization enhancement.However,fine structure registration of complex thoracoabdominal organs and large deformation registration caused by respiratory motion is challenging.To deal with this problem,we propose a 3D multi-scale attention VoxelMorph(MAVoxelMorph)registration network.To alleviate the large deformation problem,a multi-scale axial attention mechanism is utilized by using a residual dilated pyramid pooling for multi-scale feature extraction,and position-aware axial attention for long-distance dependencies between pixels capture.To further improve the large deformation and fine structure registration results,a multi-scale context channel attention mechanism is employed utilizing content information via adjacent encoding layers.Our method was evaluated on four public lung datasets(DIR-Lab dataset,Creatis dataset,Learn2Reg dataset,OASIS dataset)and a local dataset.Results proved that the proposed method achieved better registration performance than current state-of-the-art methods,especially in handling the registration of large deformations and fine structures.It also proved to be fast in 3D image registration,using about 1.5 s,and faster than most methods.Qualitative and quantitative assessments proved that the proposed MA-VoxelMorph has the potential to realize precise and fast tumor localization in clinical interventional surgeries.
基金supported by the National Natural Science Foundation of China(Nos.U22A2034,62177047)High Caliber Foreign Experts Introduction Plan funded by MOST,and Central South University Research Programme of Advanced Interdisciplinary Studies(No.2023QYJC020).
文摘Image captioning,the task of generating descriptive sentences for images,has advanced significantly with the integration of semantic information.However,traditional models still rely on static visual features that do not evolve with the changing linguistic context,which can hinder the ability to form meaningful connections between the image and the generated captions.This limitation often leads to captions that are less accurate or descriptive.In this paper,we propose a novel approach to enhance image captioning by introducing dynamic interactions where visual features continuously adapt to the evolving linguistic context.Our model strengthens the alignment between visual and linguistic elements,resulting in more coherent and contextually appropriate captions.Specifically,we introduce two innovative modules:the Visual Weighting Module(VWM)and the Enhanced Features Attention Module(EFAM).The VWM adjusts visual features using partial attention,enabling dynamic reweighting of the visual inputs,while the EFAM further refines these features to improve their relevance to the generated caption.By continuously adjusting visual features in response to the linguistic context,our model bridges the gap between static visual features and dynamic language generation.We demonstrate the effectiveness of our approach through experiments on the MS-COCO dataset,where our method outperforms state-of-the-art techniques in terms of caption quality and contextual relevance.Our results show that dynamic visual-linguistic alignment significantly enhances image captioning performance.
文摘The application of image super-resolution(SR)has brought significant assistance in the medical field,aiding doctors to make more precise diagnoses.However,solely relying on a convolutional neural network(CNN)for image SR may lead to issues such as blurry details and excessive smoothness.To address the limitations,we proposed an algorithm based on the generative adversarial network(GAN)framework.In the generator network,three different sizes of convolutions connected by a residual dense structure were used to extract detailed features,and an attention mechanism combined with dual channel and spatial information was applied to concentrate the computing power on crucial areas.In the discriminator network,using InstanceNorm to normalize tensors sped up the training process while retaining feature information.The experimental results demonstrate that our algorithm achieves higher peak signal-to-noise ratio(PSNR)and structural similarity index measure(SSIM)compared to other methods,resulting in an improved visual quality.
基金supported by the National Natural Science Foundation of China (Nos.61806107 and 61702135)。
文摘We propose a hierarchical multi-scale attention mechanism-based model in response to the low accuracy and inefficient manual classification of existing oceanic biological image classification methods. Firstly, the hierarchical efficient multi-scale attention(H-EMA) module is designed for lightweight feature extraction, achieving outstanding performance at a relatively low cost. Secondly, an improved EfficientNetV2 block is used to integrate information from different scales better and enhance inter-layer message passing. Furthermore, introducing the convolutional block attention module(CBAM) enhances the model's perception of critical features, optimizing its generalization ability. Lastly, Focal Loss is introduced to adjust the weights of complex samples to address the issue of imbalanced categories in the dataset, further improving the model's performance. The model achieved 96.11% accuracy on the intertidal marine organism dataset of Nanji Islands and 84.78% accuracy on the CIFAR-100 dataset, demonstrating its strong generalization ability to meet the demands of oceanic biological image classification.
基金funded by the National Natural Foundation of China under Grant No.61172167the Science Fund Project of Heilongjiang Province(LH2020F035).
文摘Nuclearmagnetic resonance imaging of breasts often presents complex backgrounds.Breast tumors exhibit varying sizes,uneven intensity,and indistinct boundaries.These characteristics can lead to challenges such as low accuracy and incorrect segmentation during tumor segmentation.Thus,we propose a two-stage breast tumor segmentation method leveraging multi-scale features and boundary attention mechanisms.Initially,the breast region of interest is extracted to isolate the breast area from surrounding tissues and organs.Subsequently,we devise a fusion network incorporatingmulti-scale features and boundary attentionmechanisms for breast tumor segmentation.We incorporate multi-scale parallel dilated convolution modules into the network,enhancing its capability to segment tumors of various sizes through multi-scale convolution and novel fusion techniques.Additionally,attention and boundary detection modules are included to augment the network’s capacity to locate tumors by capturing nonlocal dependencies in both spatial and channel domains.Furthermore,a hybrid loss function with boundary weight is employed to address sample class imbalance issues and enhance the network’s boundary maintenance capability through additional loss.Themethod was evaluated using breast data from 207 patients at RuijinHospital,resulting in a 6.64%increase in Dice similarity coefficient compared to the benchmarkU-Net.Experimental results demonstrate the superiority of the method over other segmentation techniques,with fewer model parameters.
文摘With the rapid expansion of drone applications,accurate detection of objects in aerial imagery has become crucial for intelligent transportation,urban management,and emergency rescue missions.However,existing methods face numerous challenges in practical deployment,including scale variation handling,feature degradation,and complex backgrounds.To address these issues,we propose Edge-enhanced and Detail-Capturing You Only Look Once(EHDC-YOLO),a novel framework for object detection in Unmanned Aerial Vehicle(UAV)imagery.Based on the You Only Look Once version 11 nano(YOLOv11n)baseline,EHDC-YOLO systematically introduces several architectural enhancements:(1)a Multi-Scale Edge Enhancement(MSEE)module that leverages multi-scale pooling and edge information to enhance boundary feature extraction;(2)an Enhanced Feature Pyramid Network(EFPN)that integrates P2-level features with Cross Stage Partial(CSP)structures and OmniKernel convolutions for better fine-grained representation;and(3)Dynamic Head(DyHead)with multi-dimensional attention mechanisms for enhanced cross-scale modeling and perspective adaptability.Comprehensive experiments on the Vision meets Drones for Detection(VisDrone-DET)2019 dataset demonstrate that EHDC-YOLO achieves significant improvements,increasing mean Average Precision(mAP)@0.5 from 33.2%to 46.1%(an absolute improvement of 12.9 percentage points)and mAP@0.5:0.95 from 19.5%to 28.0%(an absolute improvement of 8.5 percentage points)compared with the YOLOv11n baseline,while maintaining a reasonable parameter count(2.81 M vs the baseline’s 2.58 M).Further ablation studies confirm the effectiveness of each proposed component,while visualization results highlight EHDC-YOLO’s superior performance in detecting objects and handling occlusions in complex drone scenarios.
基金Shenzhen Institute of Artificial Intelligence and Robotics for Society,Grant/Award Number:AC01202201003-02GuangDong Basic and Applied Basic Research Foundation,Grant/Award Number:2024A1515010252Longgang District Shenzhen's“Ten Action Plan”for Supporting Innovation Projects,Grant/Award Number:LGKCSDPT2024002。
文摘Audio-visual scene classification(AVSC)poses a formidable challenge owing to the intricate spatial-temporal relationships exhibited by audio-visual signals,coupled with the complex spatial patterns of objects and textures found in visual images.The focus of recent studies has predominantly revolved around extracting features from diverse neural network structures,inadvertently neglecting the acquisition of semantically meaningful regions and crucial components within audio-visual data.The authors present a feature pyramid attention network(FPANet)for audio-visual scene understanding,which extracts semantically significant characteristics from audio-visual data.The authors’approach builds multi-scale hierarchical features of sound spectrograms and visual images using a feature pyramid representation and localises the semantically relevant regions with a feature pyramid attention module(FPAM).A dimension alignment(DA)strategy is employed to align feature maps from multiple layers,a pyramid spatial attention(PSA)to spatially locate essential regions,and a pyramid channel attention(PCA)to pinpoint significant temporal frames.Experiments on visual scene classification(VSC),audio scene classification(ASC),and AVSC tasks demonstrate that FPANet achieves performance on par with state-of-the-art(SOTA)approaches,with a 95.9 F1-score on the ADVANCE dataset and a relative improvement of 28.8%.Visualisation results show that FPANet can prioritise semantically meaningful areas in audio-visual signals.
基金supported by the National Natural Science Foundation of China(No.62376287)the International Science and Technology Innovation Joint Base of Machine Vision and Medical Image Processing in Hunan Province(2021CB1013)the Natural Science Foundation of Hunan Province(Nos.2022JJ30762,2023JJ70016).
文摘Globally,diabetic retinopathy(DR)is the primary cause of blindness,affecting millions of people worldwide.This widespread impact underscores the critical need for reliable and precise diagnostic techniques to ensure prompt diagnosis and effective treatment.Deep learning-based automated diagnosis for diabetic retinopathy can facilitate early detection and treatment.However,traditional deep learning models that focus on local views often learn feature representations that are less discriminative at the semantic level.On the other hand,models that focus on global semantic-level information might overlook critical,subtle local pathological features.To address this issue,we propose an adaptive multi-scale feature fusion network called(AMSFuse),which can adaptively combine multi-scale global and local features without compromising their individual representation.Specifically,our model incorporates global features for extracting high-level contextual information from retinal images.Concurrently,local features capture fine-grained details,such as microaneurysms,hemorrhages,and exudates,which are critical for DR diagnosis.These global and local features are adaptively fused using a fusion block,followed by an Integrated Attention Mechanism(IAM)that refines the fused features by emphasizing relevant regions,thereby enhancing classification accuracy for DR classification.Our model achieves 86.3%accuracy on the APTOS dataset and 96.6%RFMiD,both of which are comparable to state-of-the-art methods.
基金supported by National Natural Science Foundation of China(No.61862037)Lanzhou Jiaotong University Tianyou Innovation Team Project(No.TY202002)。
文摘To solve the problems of redundant feature information,the insignificant difference in feature representation,and low recognition accuracy of the fine-grained image,based on the ResNeXt50 model,an MSFResNet network model is proposed by fusing multi-scale feature information.Firstly,a multi-scale feature extraction module is designed to obtain multi-scale information on feature images by using different scales of convolution kernels.Meanwhile,the channel attention mechanism is used to increase the global information acquisition of the network.Secondly,the feature images processed by the multi-scale feature extraction module are fused with the deep feature images through short links to guide the full learning of the network,thus reducing the loss of texture details of the deep network feature images,and improving network generalization ability and recognition accuracy.Finally,the validity of the MSFResNet model is verified using public datasets and applied to wild mushroom identification.Experimental results show that compared with ResNeXt50 network model,the accuracy of the MSFResNet model is improved by 6.01%on the FGVC-Aircraft common dataset.It achieves 99.13%classification accuracy on the wild mushroom dataset,which is 0.47%higher than ResNeXt50.Furthermore,the experimental results of the thermal map show that the MSFResNet model significantly reduces the interference of background information,making the network focus on the location of the main body of wild mushroom,which can effectively improve the accuracy of wild mushroom identification.
基金the funding support from the National Natural Science Foundation of China(Grant No.52308340)Chongqing Talent Innovation and Entrepreneurship Demonstration Team Project(Grant No.cstc2024ycjh-bgzxm0012)the Science and Technology Projects supported by China Coal Technology and Engineering Chongqing Design and Research Institute(Group)Co.,Ltd.(Grant No.H20230317).
文摘Accurate prediction of landslide displacement is crucial for effective early warning of landslide disasters.While most existing prediction methods focus on time-series forecasting for individual monitoring points,there is limited research on the spatiotemporal characteristics of landslide deformation.This paper proposes a novel Multi-Relation Spatiotemporal Graph Residual Network with Multi-Level Feature Attention(MFA-MRSTGRN)that effectively improves the prediction performance of landslide displacement through spatiotemporal fusion.This model integrates internal seepage factors as data feature enhancements with external triggering factors,allowing for accurate capture of the complex spatiotemporal characteristics of landslide displacement and the construction of a multi-source heterogeneous dataset.The MFA-MRSTGRN model incorporates dynamic graph theory and four key modules:multilevel feature attention,temporal-residual decomposition,spatial multi-relational graph convolution,and spatiotemporal fusion prediction.This comprehensive approach enables the efficient analyses of multi-source heterogeneous datasets,facilitating adaptive exploration of the evolving multi-relational,multi-dimensional spatiotemporal complexities in landslides.When applying this model to predict the displacement of the Liangshuijing landslide,we demonstrate that the MFA-MRSTGRN model surpasses traditional models,such as random forest(RF),long short-term memory(LSTM),and spatial temporal graph convolutional networks(ST-GCN)models in terms of various evaluation metrics including mean absolute error(MAE=1.27 mm),root mean square error(RMSE=1.49 mm),mean absolute percentage error(MAPE=0.026),and R-squared(R^(2)=0.88).Furthermore,feature ablation experiments indicate that incorporating internal seepage factors improves the predictive performance of landslide displacement models.This research provides an advanced and reliable method for landslide displacement prediction.
基金Supported by National Natural Science Foundation of China(Grant Nos.51575349,51665037,51575348)State Key Laboratory of Smart Manufacturing for Special Vehicles and Transmission System(Grant No.GZ2016KF002).
文摘In the robotic welding process with thick steel plates,laser vision sensors are widely used to profile the weld seam to implement automatic seam tracking.The weld seam profile extraction(WSPE)result is a crucial step for identifying the feature points of the extracted profile to guide the welding torch in real time.The visual information processing system may collapse when interference data points in the image survive during the phase of feature point identification,which results in low tracking accuracy and poor welding quality.This paper presents a visual attention featurebased method to extract the weld seam profile(WSP)from the strong arc background using clustering results.First,a binary image is obtained through the preprocessing stage.Second,all data points with a gray value 255 are clustered with the nearest neighborhood clustering algorithm.Third,a strategy is developed to discern one cluster belonging to the WSP from the appointed candidate clusters in each loop,and a scheme is proposed to extract the entire WSP using visual continuity.Compared with the previous methods the proposed method in this paper can extract more useful details of the WSP and has better stability in terms of removing the interference data.Considerable WSPE tests with butt joints and T-joints show the anti-interference ability of the proposed method,which contributes to smoothing the welding process and shows its practical value in robotic automated welding with thick steel plates.
基金supported by the National Natural Science Foundation of China(62302167,62477013)Natural Science Foundation of Shanghai(No.24ZR1456100)+1 种基金Science and Technology Commission of Shanghai Municipality(No.24DZ2305900)the Shanghai Municipal Special Fund for Promoting High-Quality Development of Industries(2211106).
文摘Multi-label image classification is a challenging task due to the diverse sizes and complex backgrounds of objects in images.Obtaining class-specific precise representations at different scales is a key aspect of feature representation.However,existing methods often rely on the single-scale deep feature,neglecting shallow and deeper layer features,which poses challenges when predicting objects of varying scales within the same image.Although some studies have explored multi-scale features,they rarely address the flow of information between scales or efficiently obtain class-specific precise representations for features at different scales.To address these issues,we propose a two-stage,three-branch Transformer-based framework.The first stage incorporates multi-scale image feature extraction and hierarchical scale attention.This design enables the model to consider objects at various scales while enhancing the flow of information across different feature scales,improving the model’s generalization to diverse object scales.The second stage includes a global feature enhancement module and a region selection module.The global feature enhancement module strengthens interconnections between different image regions,mitigating the issue of incomplete represen-tations,while the region selection module models the cross-modal relationships between image features and labels.Together,these components enable the efficient acquisition of class-specific precise feature representations.Extensive experiments on public datasets,including COCO2014,VOC2007,and VOC2012,demonstrate the effectiveness of our proposed method.Our approach achieves consistent performance gains of 0.3%,0.4%,and 0.2%over state-of-the-art methods on the three datasets,respectively.These results validate the reliability and superiority of our approach for multi-label image classification.
基金funded by the China Chongqing Municipal Science and Technology Bureau,grant numbers 2024TIAD-CYKJCXX0121,2024NSCQ-LZX0135Chongqing Municipal Commission of Housing and Urban-Rural Development,grant number CKZ2024-87+3 种基金the Chongqing University of Technology graduate education high-quality development project,grant number gzlsz202401the Chongqing University of Technology-Chongqing LINGLUE Technology Co.,Ltd.,Electronic Information(Artificial Intelligence)graduate joint training basethe Postgraduate Education and Teaching Reform Research Project in Chongqing,grant number yjg213116the Chongqing University of Technology-CISDI Chongqing Information Technology Co.,Ltd.,Computer Technology graduate joint training base.
文摘Detecting abnormal cervical cells is crucial for early identification and timely treatment of cervical cancer.However,this task is challenging due to the morphological similarities between abnormal and normal cells and the significant variations in cell size.Pathologists often refer to surrounding cells to identify abnormalities.To emulate this slide examination behavior,this study proposes a Multi-Scale Feature Fusion Network(MSFF-Net)for detecting cervical abnormal cells.MSFF-Net employs a Cross-Scale Pooling Model(CSPM)to effectively capture diverse features and contextual information,ranging from local details to the overall structure.Additionally,a Multi-Scale Fusion Attention(MSFA)module is introduced to mitigate the impact of cell size variations by adaptively fusing local and global information at different scales.To handle the complex environment of cervical cell images,such as cell adhesion and overlapping,the Inner-CIoU loss function is utilized to more precisely measure the overlap between bounding boxes,thereby improving detection accuracy in such scenarios.Experimental results on the Comparison detector dataset demonstrate that MSFF-Net achieves a mean average precision(mAP)of 63.2%,outperforming state-of-the-art methods while maintaining a relatively small number of parameters(26.8 M).This study highlights the effectiveness of multi-scale feature fusion in enhancing the detection of cervical abnormal cells,contributing to more accurate and efficient cervical cancer screening.
基金funded by the National Natural Science Foundation of China(Grant Nos.62161019,62061024).
文摘The detection of Ochotona Curzoniae serves as a fundamental component for estimating the population size of this species and for analyzing the dynamics of its population fluctuations.In natural environments,the pixels representing Ochotona Curzoniae constitute a small fraction of the total pixels,and their distinguishing features are often subtle,complicating the target detection process.To effectively extract the characteristics of these small targets,a feature fusion approach that utilizes up-sampling and channel integration from various layers within a CNN can significantly enhance the representation of target features,ultimately improving detection accuracy.However,the top-down fusion of features from different layers may lead to information duplication and semantic bias,resulting in redundancy and high-frequency noise.To address the challenges of information redundancy and high-frequency noise during the feature fusion process in CNN,we have developed a target detection model for Ochotona Curzoniae.This model is based on a spatial-channel reconfiguration convolutional(SCConv)attentional mechanism and feature fusion(FFBCA),integrated with the Faster R-CNN framework.It consists of a feature extraction network,an attention mechanism-based feature fusion module,and a jump residual connection fusion module.Initially,we designed a dual attention mechanism feature fusion module that employs spatial-channel reconstruction convolution.In the spatial dimension,the attention mechanism adopts a separation-reconstruction approach,calculating a weight matrix for the spatial information within the feature map through group normalization.This process directs the model to concentrate on feature information assigned varying weights,thereby reducing redundancy during feature fusion.In the channel dimension,the attention mechanism utilizes a partition-transpose-fusion method,segmenting the input feature map into high-noise and low-noise components based on the variance of the feature information.The high-noise segment is processed through a low-pass filter constructed from pointwise convolution(PWC)to eliminate some high-frequency noise,while the low-noise segment employs a bottleneck structure with global average pooling(GAP)to generate a weight matrix that emphasizes the significance of channel dimension feature information.This approach diminishes the model’s focus on low-weight feature information,thereby preserving low-frequency semantic information while reducing information redundancy.Furthermore,we have developed a novel feature extraction network,ResNeXt-S,by integrating the Sim attention mechanism into ResNeXt50.This configuration assigns three-dimensional attention weights to each position within the feature map,thereby enhancing the local feature information of small targets while reducing background noise.Finally,we constructed a jump residual connection fusion module to minimize the loss of high-level semantic information during the feature fusion process.Experiments on Ochotona Curzoniae target detection on the Ochotona Curzoniae dataset show that the detection accuracy of the model in this paper is 92.3%,which is higher than that of FSSD512(84.6%),TDFSSD512(81.3%),FPN(86.5%),FFBAM(88.5%),Faster R-CNN(89.6%),and SSD512(88.6%)detection accuracies.
基金funded by Liaoning Provincial Department of Education Project,Award number JYTMS20230418.
文摘Pest detection techniques are helpful in reducing the frequency and scale of pest outbreaks;however,their application in the actual agricultural production process is still challenging owing to the problems of interspecies similarity,multi-scale,and background complexity of pests.To address these problems,this study proposes an FD-YOLO pest target detection model.The FD-YOLO model uses a Fully Connected Feature Pyramid Network(FC-FPN)instead of a PANet in the neck,which can adaptively fuse multi-scale information so that the model can retain small-scale target features in the deep layer,enhance large-scale target features in the shallow layer,and enhance the multiplexing of effective features.A dual self-attention module(DSA)is then embedded in the C3 module of the neck,which captures the dependencies between the information in both spatial and channel dimensions,effectively enhancing global features.We selected 16 types of pests that widely damage field crops in the IP102 pest dataset,which were used as our dataset after data supplementation and enhancement.The experimental results showed that FD-YOLO’s mAP@0.5 improved by 6.8%compared to YOLOv5,reaching 82.6%and 19.1%–5%better than other state-of-the-art models.This method provides an effective new approach for detecting similar or multiscale pests in field crops.
