Despite its remarkable performance on natural images,the segment anything model(SAM)lacks domain-specific information in medical imaging.and faces the challenge of losing local multi-scale information in the encoding ...Despite its remarkable performance on natural images,the segment anything model(SAM)lacks domain-specific information in medical imaging.and faces the challenge of losing local multi-scale information in the encoding phase.This paper presents a medical image segmentation model based on SAM with a local multi-scale feature encoder(LMSFE-SAM)to address the issues above.Firstly,based on the SAM,a local multi-scale feature encoder is introduced to improve the representation of features within local receptive field,thereby supplying the Vision Transformer(ViT)branch in SAM with enriched local multi-scale contextual information.At the same time,a multiaxial Hadamard product module(MHPM)is incorporated into the local multi-scale feature encoder in a lightweight manner to reduce the quadratic complexity and noise interference.Subsequently,a cross-branch balancing adapter is designed to balance the local and global information between the local multi-scale feature encoder and the ViT encoder in SAM.Finally,to obtain smaller input image size and to mitigate overlapping in patch embeddings,the size of the input image is reduced from 1024×1024 pixels to 256×256 pixels,and a multidimensional information adaptation component is developed,which includes feature adapters,position adapters,and channel-spatial adapters.This component effectively integrates the information from small-sized medical images into SAM,enhancing its suitability for clinical deployment.The proposed model demonstrates an average enhancement ranging from 0.0387 to 0.3191 across six objective evaluation metrics on BUSI,DDTI,and TN3K datasets compared to eight other representative image segmentation models.This significantly enhances the performance of the SAM on medical images,providing clinicians with a powerful tool in clinical diagnosis.展开更多
Detailed individual tree crown segmentation is highly relevant for the detection and monitoring of Fraxinus excelsior L.trees affected by ash dieback,a major threat to common ash populations across Europe.In this stud...Detailed individual tree crown segmentation is highly relevant for the detection and monitoring of Fraxinus excelsior L.trees affected by ash dieback,a major threat to common ash populations across Europe.In this study,both fine and coarse crown segmentation methods were applied to close-range multispectral UAV imagery.The fine tree crown segmentation method utilized a novel unsupervised machine learning approach based on a blended NIR-NDVI image,whereas the coarse segmentation relied on the segment anything model(SAM).Both methods successfully delineated tree crown outlines,however,only the fine segmentation accurately captured internal canopy gaps.Despite these structural differences,mean NDVI values calculated per tree crown revealed no significant differences between the two approaches,indicating that coarse segmentation is sufficient for mean vegetation index assessments.Nevertheless,the fine segmentation revealed increased heterogeneity in NDVI values in more severely damaged trees,underscoring its value for detailed structural and health analyses.Furthermore,the fine segmentation workflow proved transferable to both individual UAV images and orthophotos from broader UAV surveys.For applications focused on structural integrity and spatial variation in canopy health,the fine segmentation approach is recommended.展开更多
AIM:To construct an intelligent segmentation scheme for precise localization of central serous chorioretinopathy(CSC)leakage points,thereby enabling ophthalmologists to deliver accurate laser treatment without navigat...AIM:To construct an intelligent segmentation scheme for precise localization of central serous chorioretinopathy(CSC)leakage points,thereby enabling ophthalmologists to deliver accurate laser treatment without navigational laser equipment.METHODS:A dataset with dual labels(point-level and pixel-level)was first established based on fundus fluorescein angiography(FFA)images of CSC and subsequently divided into training(102 images),validation(40 images),and test(40 images)datasets.An intelligent segmentation method was then developed,based on the You Only Look Once version 8 Pose Estimation(YOLOv8-Pose)model and segment anything model(SAM),to segment CSC leakage points.Next,the YOLOv8-Pose model was trained for 200 epochs,and the best-performing model was selected to form the optimal combination with SAM.Additionally,the classic five types of U-Net series models[i.e.,U-Net,recurrent residual U-Net(R2U-Net),attention U-Net(AttU-Net),recurrent residual attention U-Net(R2AttUNet),and nested U-Net(UNet^(++))]were initialized with three random seeds and trained for 200 epochs,resulting in a total of 15 baseline models for comparison.Finally,based on the metrics including Dice similarity coefficient(DICE),intersection over union(IoU),precision,recall,precisionrecall(PR)curve,and receiver operating characteristic(ROC)curve,the proposed method was compared with baseline models through quantitative and qualitative experiments for leakage point segmentation,thereby demonstrating its effectiveness.RESULTS:With the increase of training epochs,the mAP50-95,Recall,and precision of the YOLOv8-Pose model showed a significant increase and tended to stabilize,and it achieved a preliminary localization success rate of 90%(i.e.,36 images)for CSC leakage points in 40 test images.Using manually expert-annotated pixel-level labels as the ground truth,the proposed method achieved outcomes with a DICE of 57.13%,an IoU of 45.31%,a precision of 45.91%,a recall of 93.57%,an area under the PR curve(AUC-PR)of 0.78 and an area under the ROC curve(AUC-ROC)of 0.97,which enables more accurate segmentation of CSC leakage points.CONCLUSION:By combining the precise localization capability of the YOLOv8-Pose model with the robust and flexible segmentation ability of SAM,the proposed method not only demonstrates the effectiveness of the YOLOv8-Pose model in detecting keypoint coordinates of CSC leakage points from the perspective of application innovation but also establishes a novel approach for accurate segmentation of CSC leakage points through the“detect-then-segment”strategy,thereby providing a potential auxiliary means for the automatic and precise realtime localization of leakage points during traditional laser photocoagulation for CSC.展开更多
Data augmentation plays an important role in training deep neural model by expanding the size and diversity of the dataset.Initially,data augmentation mainly involved some simple transformations of images.Later,in ord...Data augmentation plays an important role in training deep neural model by expanding the size and diversity of the dataset.Initially,data augmentation mainly involved some simple transformations of images.Later,in order to increase the diversity and complexity of data,more advanced methods appeared and evolved to sophisticated generative models.However,these methods required a mass of computation of training or searching.In this paper,a novel training-free method that utilises the Pre-Trained Segment Anything Model(SAM)model as a data augmentation tool(PTSAM-DA)is proposed to generate the augmented annotations for images.Without the need for training,it obtains prompt boxes from the original annotations and then feeds the boxes to the pre-trained SAM to generate diverse and improved annotations.In this way,annotations are augmented more ingenious than simple manipulations without incurring huge computation for training a data augmentation model.Multiple comparative experiments on three datasets are conducted,including an in-house dataset,ADE20K and COCO2017.On this in-house dataset,namely Agricultural Plot Segmentation Dataset,maximum improvements of 3.77%and 8.92%are gained in two mainstream metrics,mIoU and mAcc,respectively.Consequently,large vision models like SAM are proven to be promising not only in image segmentation but also in data augmentation.展开更多
Existing sandstone rock structure evaluation methods rely on visual inspection,with low efficiency,semi-quantitative analysis of roundness,and inability to perform classified statistics in particle size analysis.This ...Existing sandstone rock structure evaluation methods rely on visual inspection,with low efficiency,semi-quantitative analysis of roundness,and inability to perform classified statistics in particle size analysis.This study presents an intelligent evaluation method for sandstone rock structure based on the Segment Anything Model(SAM).By developing a lightweight SAM fine-tuning method with rank-decomposition matrix adapters,a multispectral rock particle segmentation model named CoreSAM is constructed,which achieves rock particle edge extraction and type identification.Building upon this,we propose a comprehensive quantitative evaluation system for rock structure,assessing parameters including particle size,sorting,roundness,particle contact and cementation types.The experimental results demonstrate that CoreSAM outperforms existing methods in rock particle segmentation accuracy while showing excellent generalization across different image types such as CT scans and core photographs.The proposed method enables full-sample,classified particle size analysis and quantitative characterization of parameters like roundness,advancing reservoir evaluation towards more precise,quantitative,intuitive,and comprehensive development.展开更多
Efficient segmentation of oiled pixels in optical remotely sensed images is the precondition of optical identification and classification of different spilled oils,which remains one of the keys to optical remote sensi...Efficient segmentation of oiled pixels in optical remotely sensed images is the precondition of optical identification and classification of different spilled oils,which remains one of the keys to optical remote sensing of oil spills.Optical remotely sensed images of oil spills are inherently multidimensional and embedded with a complex knowledge framework.This complexity often hinders the effectiveness of mechanistic algorithms across varied scenarios.Although optical remote-sensing theory for oil spills has advanced,the scarcity of curated datasets and the difficulty of collecting them limit their usefulness for training deep learning models.This study introduces a data expansion strategy that utilizes the Segment Anything Model(SAM),effectively bridging the gap between traditional mechanism algorithms and emergent self-adaptive deep learning models.Optical dimension reduction is achieved through standardized preprocessing processes that address the decipherable properties of the input image.After preprocessing,SAM can swiftly and accurately segment spilled oil in images.The unified AI-based workflow significantly accelerates labeled-dataset creation and has proven effective for both rapid emergency intelligence during spill incidents and the rapid mapping and classification of oil footprints across China’s coastal waters.Our results show that coupling a remote sensing mechanism with a foundation model enables near-real-time,large-scale monitoring of complex surface slicks and offers guidance for the next generation of detection and quantification algorithms.展开更多
Deep learning-based methods have become alternatives to traditional numerical weather prediction systems,offering faster computation and the ability to utilize large historical datasets.However,the application of deep...Deep learning-based methods have become alternatives to traditional numerical weather prediction systems,offering faster computation and the ability to utilize large historical datasets.However,the application of deep learning to medium-range regional weather forecasting with limited data remains a significant challenge.In this work,three key solutions are proposed:(1)motivated by the need to improve model performance in data-scarce regional forecasting scenarios,the authors innovatively apply semantic segmentation models,to better capture spatiotemporal features and improve prediction accuracy;(2)recognizing the challenge of overfitting and the inability of traditional noise-based data augmentation methods to effectively enhance model robustness,a novel learnable Gaussian noise mechanism is introduced that allows the model to adaptively optimize perturbations for different locations,ensuring more effective learning;and(3)to address the issue of error accumulation in autoregressive prediction,as well as the challenge of learning difficulty and the lack of intermediate data utilization in one-shot prediction,the authors propose a cascade prediction approach that effectively resolves these problems while significantly improving model forecasting performance.The method achieves a competitive result in The East China Regional AI Medium Range Weather Forecasting Competition.Ablation experiments further validate the effectiveness of each component,highlighting their contributions to enhancing prediction performance.展开更多
Three model polyurethane hard segments based on dimethylol butanoic acid (DMBA) and 1,6-hexane diisocyanate (HDI), toluene diisocyanate (TDI) and 4,4'-diphenylmethane diisocyanate (MDI) were prepared by the soluti...Three model polyurethane hard segments based on dimethylol butanoic acid (DMBA) and 1,6-hexane diisocyanate (HDI), toluene diisocyanate (TDI) and 4,4'-diphenylmethane diisocyanate (MDI) were prepared by the solution method. Fourier Infrared (FTIR) spectroscopy was employed to study the H-bonds in these model polyurethanes. The model polyurethane hard segment prepared from HDI and 1,4-butanodiol (BDO) was used for comparison. It was found that the incorporation of the pendent carboxyl through DMBA into the model hard segments weakens the original NH…O = C H-bond but gives more H-bond patterns based on the two H-bond donors, urethane NH and carboxylic OH. The carboxylic dimer is one of the main H-bond types and is stronger than another main H-bond type NH…O=C. In addition, the H-bond in aromatic model hard segments is stronger than that of aliphatic hard segments. The appearance of the free C=O and the fact that almost all N—H is H-bonded suggest that there possibly exist either the third H-bond acceptor or the H-bond formed by one acceptor with two donors.展开更多
The texture analysis is often discussed in image processing domain, but most methods are limited within gray-level image or color image, and the present conception of texture is defined mainly based on gray-level imag...The texture analysis is often discussed in image processing domain, but most methods are limited within gray-level image or color image, and the present conception of texture is defined mainly based on gray-level image of single band. One of the essential characters of remote sensing image is multidimensional or even high-dimensional, and the traditional texture conception cannot contain enough information for these. Therefore, it is necessary to pursuit a proper texture definition based on remote sensing images, which is the first discussion in this paper. This paper describes the mapping model of spectral vector in two-dimensional image space using Markov random field (MRF), establishes a texture model of multiband remote sensing image based on MRF, and analyzes the calculations of Gibbs potential energy and Gibbs parameters. Further, this paper also analyzes the limitations of the traditional Gibbs model, prefers a new Gibbs model avoiding estimation of parameters, and presents a new texture segmentation algorithm for hy-perspectral remote sensing image later.展开更多
Background:The greater trochanter marker is commonly used in 3-dimensional(3D) models;however,its influence on hip and knee kinematics during gait is unclear.Understanding the influence of the greater trochanter marke...Background:The greater trochanter marker is commonly used in 3-dimensional(3D) models;however,its influence on hip and knee kinematics during gait is unclear.Understanding the influence of the greater trochanter marker is important when quantifying frontal and transverse plane hip and knee kinematics,parameters which are particularly relevant to investigate in individuals with conditions such as patellofemoral pain,knee osteoarthritis,anterior cruciate ligament(ACL) injury,and hip pain.The aim of this study was to evaluate the effect of including the greater trochanter in the construction of the thigh segment on hip and knee kinematics during gait.Methods:3D kinematics were collected in 19 healthy subjects during walking using a surface marker system.Hip and knee angles were compared across two thigh segment definitions(with and without greater trochanter) at two time points during stance:peak knee flexion(PKF) and minimum knee flexion(Min KF).Results:Hip and knee angles differed in magnitude and direction in the transverse plane at both time points.In the thigh model with the greater trochanter the hip was more externally rotated than in the thigh model without the greater trochanter(PKF:-9.34°± 5.21° vs.1.40°± 5.22°,Min KF:-5.68°± 4.24° vs.5.01°± 4.86°;p < 0.001).In the thigh model with the greater trochanter,the knee angle was more internally rotated compared to the knee angle calculated using the thigh definition without the greater trochanter(PKF:14.67°± 6.78° vs.4.33°± 4.18°,Min KF:10.54°± 6.71° vs.-0.01°± 2.69°;p < 0.001).Small but significant differences were detected in the sagittal and frontal plane angles at both time points(p < 0.001).Conclusion:Hip and knee kinematics differed across different segment definitions including or excluding the greater trochanter marker,especially in the transverse plane.Therefore when considering whether to include the greater trochanter in the thigh segment model when using a surface markers to calculate 3D kinematics for movement assessment,it is important to have a clear understanding of the effect of different marker sets and segment models in use.展开更多
The quantitative analysis of dispersed phases(bubbles,droplets,and particles)in multiphase flow systems represents a persistent technological challenge in petroleum engineering applications,including CO2-enhanced oil ...The quantitative analysis of dispersed phases(bubbles,droplets,and particles)in multiphase flow systems represents a persistent technological challenge in petroleum engineering applications,including CO2-enhanced oil recovery,foam flooding,and unconventional reservoir development.Current characterization methods remain constrained by labor-intensive manual workflows and limited dynamic analysis capabilities,particularly for processing large-scale microscopy data and video sequences that capture critical transient behavior like gas cluster migration and droplet coalescence.These limitations hinder the establishment of robust correlations between pore-scale flow patterns and reservoir-scale production performance.This study introduces a novel computer vision framework that integrates foundation models with lightweight neural networks to address these industry challenges.Leveraging the segment anything model's zero-shot learning capability,we developed an automated workflow that achieves an efficiency improvement of approximately 29 times in bubble labeling compared to manual methods while maintaining less than 2%deviation from expert annotations.Engineering-oriented optimization ensures lightweight deployment with 94%segmentation accuracy,while the integrated quantification system precisely resolves gas saturation,shape factors,and interfacial dynamics,parameters critical for optimizing gas injection strategies and predicting phase redistribution patterns.Validated through microfluidic gas-liquid displacement experiments for discontinuous phase segmentation accuracy,this methodology enables precise bubble morphology quantification with broad application potential in multiphase systems,including emulsion droplet dynamics characterization and particle transport behavior analysis.This work bridges the critical gap between pore-scale dynamics characterization and reservoir-scale simulation requirements,providing a foundational framework for intelligent flow diagnostics and predictive modeling in next-generation digital oilfield systems.展开更多
CO_(2) storage capacity is significantly influenced by the saturation levels of reservoir rocks,with underground fluid saturation typically evaluated using resistivity data.The conductive pathways of fluids in various...CO_(2) storage capacity is significantly influenced by the saturation levels of reservoir rocks,with underground fluid saturation typically evaluated using resistivity data.The conductive pathways of fluids in various states within rock pores differ,alongside variations in conductive mechanisms.To clarify the conductivity of water in rocks across different states,this study employed a three-pore segment saturation model,which corrected for the additional conductivity of clay by categorizing water into large-pore segment,medium-pore segment,and small-pore segment types.Addressing the heterogeneity of tight sandstone reservoirs,we classified distinct pore structures and inverted Archie equation parameters from NMR logging data using a segmented characterization approach,yielding dynamic Archie parameters that vary with depth.Ultimately,we established an improved saturation parameter method based on joint inversion of NMR and resistivity data,which was validated through laboratory experiments and practical downhole applications.The results indicate that this saturation parameter inversion method has been effectively applied in both settings.Furthermore,we discussed the varying conductive behaviors of fluids in large and medium pore segment under saturated and drained states.Lastly,we proposed a workflow for inverting saturation based on downhole data,providing a robust foundation for CO_(2) storage and predicting underground fluid saturation.展开更多
Large-scale unsupervised semantic segmentation(LUSS)is a sophisticated process that aims to segment similar areas within an image without relying on labeled training data.While existing methodologies have made substan...Large-scale unsupervised semantic segmentation(LUSS)is a sophisticated process that aims to segment similar areas within an image without relying on labeled training data.While existing methodologies have made substantial progress in this area,there is ample scope for enhancement.We thus introduce the PASS-SAM model,a comprehensive solution that amalgamates the benefits of various models to improve segmentation performance.展开更多
Although the genetic algorithm (GA) has very powerful robustness and fitness, it needs a large size of population and a large number of iterations to reach the optimum result. Especially when GA is used in complex str...Although the genetic algorithm (GA) has very powerful robustness and fitness, it needs a large size of population and a large number of iterations to reach the optimum result. Especially when GA is used in complex structural optimization problems, if the structural reanalysis technique is not adopted, the more the number of finite element analysis (FEA) is, the more the consuming time is. In the conventional structural optimization the number of FEA can be reduced by the structural reanalysis technique based on the approximation techniques and sensitivity analysis. With these techniques, this paper provides a new approximation model-segment approximation model, adopted for the GA application. This segment approximation model can decrease the number of FEA and increase the convergence rate of GA. So it can apparently decrease the computation time of GA. Two examples demonstrate the availability of the new segment approximation model.展开更多
AIM:To develop a deep learning-based model for automatic retinal vascular segmentation,analyzing and comparing parameters under diverse glucose metabolic status(normal,prediabetes,diabetes)and to assess the potential ...AIM:To develop a deep learning-based model for automatic retinal vascular segmentation,analyzing and comparing parameters under diverse glucose metabolic status(normal,prediabetes,diabetes)and to assess the potential of artificial intelligence(AI)in image segmentation and retinal vascular parameters for predicting prediabetes and diabetes.METHODS:Retinal fundus photos from 200 normal individuals,200 prediabetic patients,and 200 diabetic patients(600 eyes in total)were used.The U-Net network served as the foundational architecture for retinal arteryvein segmentation.An automatic segmentation and evaluation system for retinal vascular parameters was trained,encompassing 26 parameters.RESULTS:Significant differences were found in retinal vascular parameters across normal,prediabetes,and diabetes groups,including artery diameter(P=0.008),fractal dimension(P=0.000),vein curvature(P=0.003),C-zone artery branching vessel count(P=0.049),C-zone vein branching vessel count(P=0.041),artery branching angle(P=0.005),vein branching angle(P=0.001),artery angle asymmetry degree(P=0.003),vessel length density(P=0.000),and vessel area density(P=0.000),totaling 10 parameters.CONCLUSION:The deep learning-based model facilitates retinal vascular parameter identification and quantification,revealing significant differences.These parameters exhibit potential as biomarkers for prediabetes and diabetes.展开更多
Dear Editor,This letter presents techniques to simplify dataset generation for instance segmentation of raw meat products,a critical step toward automating food production lines.Accurate segmentation is essential for ...Dear Editor,This letter presents techniques to simplify dataset generation for instance segmentation of raw meat products,a critical step toward automating food production lines.Accurate segmentation is essential for addressing challenges such as occlusions,indistinct edges,and stacked configurations,which demand large,diverse datasets.To meet these demands,we propose two complementary approaches:a semi-automatic annotation interface using tools like the segment anything model(SAM)and GrabCut and a synthetic data generation pipeline leveraging 3D-scanned models.These methods reduce reliance on real meat,mitigate food waste,and improve scalability.Experimental results demonstrate that incorporating synthetic data enhances segmentation model performance and,when combined with real data,further boosts accuracy,paving the way for more efficient automation in the food industry.展开更多
基金supported by Natural Science Foundation Programme of Gansu Province(No.24JRRA231)National Natural Science Foundation of China(No.62061023)Gansu Provincial Science and Technology Plan Key Research and Development Program Project(No.24YFFA024).
文摘Despite its remarkable performance on natural images,the segment anything model(SAM)lacks domain-specific information in medical imaging.and faces the challenge of losing local multi-scale information in the encoding phase.This paper presents a medical image segmentation model based on SAM with a local multi-scale feature encoder(LMSFE-SAM)to address the issues above.Firstly,based on the SAM,a local multi-scale feature encoder is introduced to improve the representation of features within local receptive field,thereby supplying the Vision Transformer(ViT)branch in SAM with enriched local multi-scale contextual information.At the same time,a multiaxial Hadamard product module(MHPM)is incorporated into the local multi-scale feature encoder in a lightweight manner to reduce the quadratic complexity and noise interference.Subsequently,a cross-branch balancing adapter is designed to balance the local and global information between the local multi-scale feature encoder and the ViT encoder in SAM.Finally,to obtain smaller input image size and to mitigate overlapping in patch embeddings,the size of the input image is reduced from 1024×1024 pixels to 256×256 pixels,and a multidimensional information adaptation component is developed,which includes feature adapters,position adapters,and channel-spatial adapters.This component effectively integrates the information from small-sized medical images into SAM,enhancing its suitability for clinical deployment.The proposed model demonstrates an average enhancement ranging from 0.0387 to 0.3191 across six objective evaluation metrics on BUSI,DDTI,and TN3K datasets compared to eight other representative image segmentation models.This significantly enhances the performance of the SAM on medical images,providing clinicians with a powerful tool in clinical diagnosis.
基金This study was conducted within the project FraxVir“Detection,characterisation and analyses of the occurrence of viruses and ash dieback in special stands of Fraxinus excelsior-a supplementary study to the FraxForFuture demonstration project”and receives funding via the Waldklimafonds(WKF)funded by the German Federal Ministry of Food and Agriculture(BMEL)and Federal Ministry for the Environment,Nature Conservation,Nuclear Safety and Consumer Protection(BMUV)administrated by the Agency for Renewable Resources(FNR)under grant agreement 2220WK40A4.
文摘Detailed individual tree crown segmentation is highly relevant for the detection and monitoring of Fraxinus excelsior L.trees affected by ash dieback,a major threat to common ash populations across Europe.In this study,both fine and coarse crown segmentation methods were applied to close-range multispectral UAV imagery.The fine tree crown segmentation method utilized a novel unsupervised machine learning approach based on a blended NIR-NDVI image,whereas the coarse segmentation relied on the segment anything model(SAM).Both methods successfully delineated tree crown outlines,however,only the fine segmentation accurately captured internal canopy gaps.Despite these structural differences,mean NDVI values calculated per tree crown revealed no significant differences between the two approaches,indicating that coarse segmentation is sufficient for mean vegetation index assessments.Nevertheless,the fine segmentation revealed increased heterogeneity in NDVI values in more severely damaged trees,underscoring its value for detailed structural and health analyses.Furthermore,the fine segmentation workflow proved transferable to both individual UAV images and orthophotos from broader UAV surveys.For applications focused on structural integrity and spatial variation in canopy health,the fine segmentation approach is recommended.
基金Supported by the Shenzhen Science and Technology Program(No.JCYJ20240813152704006)the National Natural Science Foundation of China(No.62401259)+2 种基金the Fundamental Research Funds for the Central Universities(No.NZ2024036)the Postdoctoral Fellowship Program of CPSF(No.GZC20242228)High Performance Computing Platform of Nanjing University of Aeronautics and Astronautics。
文摘AIM:To construct an intelligent segmentation scheme for precise localization of central serous chorioretinopathy(CSC)leakage points,thereby enabling ophthalmologists to deliver accurate laser treatment without navigational laser equipment.METHODS:A dataset with dual labels(point-level and pixel-level)was first established based on fundus fluorescein angiography(FFA)images of CSC and subsequently divided into training(102 images),validation(40 images),and test(40 images)datasets.An intelligent segmentation method was then developed,based on the You Only Look Once version 8 Pose Estimation(YOLOv8-Pose)model and segment anything model(SAM),to segment CSC leakage points.Next,the YOLOv8-Pose model was trained for 200 epochs,and the best-performing model was selected to form the optimal combination with SAM.Additionally,the classic five types of U-Net series models[i.e.,U-Net,recurrent residual U-Net(R2U-Net),attention U-Net(AttU-Net),recurrent residual attention U-Net(R2AttUNet),and nested U-Net(UNet^(++))]were initialized with three random seeds and trained for 200 epochs,resulting in a total of 15 baseline models for comparison.Finally,based on the metrics including Dice similarity coefficient(DICE),intersection over union(IoU),precision,recall,precisionrecall(PR)curve,and receiver operating characteristic(ROC)curve,the proposed method was compared with baseline models through quantitative and qualitative experiments for leakage point segmentation,thereby demonstrating its effectiveness.RESULTS:With the increase of training epochs,the mAP50-95,Recall,and precision of the YOLOv8-Pose model showed a significant increase and tended to stabilize,and it achieved a preliminary localization success rate of 90%(i.e.,36 images)for CSC leakage points in 40 test images.Using manually expert-annotated pixel-level labels as the ground truth,the proposed method achieved outcomes with a DICE of 57.13%,an IoU of 45.31%,a precision of 45.91%,a recall of 93.57%,an area under the PR curve(AUC-PR)of 0.78 and an area under the ROC curve(AUC-ROC)of 0.97,which enables more accurate segmentation of CSC leakage points.CONCLUSION:By combining the precise localization capability of the YOLOv8-Pose model with the robust and flexible segmentation ability of SAM,the proposed method not only demonstrates the effectiveness of the YOLOv8-Pose model in detecting keypoint coordinates of CSC leakage points from the perspective of application innovation but also establishes a novel approach for accurate segmentation of CSC leakage points through the“detect-then-segment”strategy,thereby providing a potential auxiliary means for the automatic and precise realtime localization of leakage points during traditional laser photocoagulation for CSC.
基金Natural Science Foundation of Zhejiang Province,Grant/Award Number:LY23F020025Science and Technology Commissioner Program of Huzhou,Grant/Award Number:2023GZ42Sichuan Provincial Science and Technology Support Program,Grant/Award Numbers:2023ZHCG0005,2023ZHCG0008。
文摘Data augmentation plays an important role in training deep neural model by expanding the size and diversity of the dataset.Initially,data augmentation mainly involved some simple transformations of images.Later,in order to increase the diversity and complexity of data,more advanced methods appeared and evolved to sophisticated generative models.However,these methods required a mass of computation of training or searching.In this paper,a novel training-free method that utilises the Pre-Trained Segment Anything Model(SAM)model as a data augmentation tool(PTSAM-DA)is proposed to generate the augmented annotations for images.Without the need for training,it obtains prompt boxes from the original annotations and then feeds the boxes to the pre-trained SAM to generate diverse and improved annotations.In this way,annotations are augmented more ingenious than simple manipulations without incurring huge computation for training a data augmentation model.Multiple comparative experiments on three datasets are conducted,including an in-house dataset,ADE20K and COCO2017.On this in-house dataset,namely Agricultural Plot Segmentation Dataset,maximum improvements of 3.77%and 8.92%are gained in two mainstream metrics,mIoU and mAcc,respectively.Consequently,large vision models like SAM are proven to be promising not only in image segmentation but also in data augmentation.
基金Supported by the National Natural Science Foundation of China(42372175,72088101)PetroChina Science and Technology Project of(2023DJ84)Basic Research Cooperation Project between China National Petroleum Corporation and Peking University.
文摘Existing sandstone rock structure evaluation methods rely on visual inspection,with low efficiency,semi-quantitative analysis of roundness,and inability to perform classified statistics in particle size analysis.This study presents an intelligent evaluation method for sandstone rock structure based on the Segment Anything Model(SAM).By developing a lightweight SAM fine-tuning method with rank-decomposition matrix adapters,a multispectral rock particle segmentation model named CoreSAM is constructed,which achieves rock particle edge extraction and type identification.Building upon this,we propose a comprehensive quantitative evaluation system for rock structure,assessing parameters including particle size,sorting,roundness,particle contact and cementation types.The experimental results demonstrate that CoreSAM outperforms existing methods in rock particle segmentation accuracy while showing excellent generalization across different image types such as CT scans and core photographs.The proposed method enables full-sample,classified particle size analysis and quantitative characterization of parameters like roundness,advancing reservoir evaluation towards more precise,quantitative,intuitive,and comprehensive development.
基金The National Natural Science Foundation of China under contract No.42371380the National Key Research and Development Program of China under contract No.2023YFC2811800the Fundamental Research Funds for the Central Universities under contract No.0904-14380035.
文摘Efficient segmentation of oiled pixels in optical remotely sensed images is the precondition of optical identification and classification of different spilled oils,which remains one of the keys to optical remote sensing of oil spills.Optical remotely sensed images of oil spills are inherently multidimensional and embedded with a complex knowledge framework.This complexity often hinders the effectiveness of mechanistic algorithms across varied scenarios.Although optical remote-sensing theory for oil spills has advanced,the scarcity of curated datasets and the difficulty of collecting them limit their usefulness for training deep learning models.This study introduces a data expansion strategy that utilizes the Segment Anything Model(SAM),effectively bridging the gap between traditional mechanism algorithms and emergent self-adaptive deep learning models.Optical dimension reduction is achieved through standardized preprocessing processes that address the decipherable properties of the input image.After preprocessing,SAM can swiftly and accurately segment spilled oil in images.The unified AI-based workflow significantly accelerates labeled-dataset creation and has proven effective for both rapid emergency intelligence during spill incidents and the rapid mapping and classification of oil footprints across China’s coastal waters.Our results show that coupling a remote sensing mechanism with a foundation model enables near-real-time,large-scale monitoring of complex surface slicks and offers guidance for the next generation of detection and quantification algorithms.
基金supported by the National Natural Science Foundation of China[grant number 62376217]the Young Elite Scientists Sponsorship Program by CAST[grant number 2023QNRC001]the Joint Research Project for Meteorological Capacity Improvement[grant number 24NLTSZ003]。
文摘Deep learning-based methods have become alternatives to traditional numerical weather prediction systems,offering faster computation and the ability to utilize large historical datasets.However,the application of deep learning to medium-range regional weather forecasting with limited data remains a significant challenge.In this work,three key solutions are proposed:(1)motivated by the need to improve model performance in data-scarce regional forecasting scenarios,the authors innovatively apply semantic segmentation models,to better capture spatiotemporal features and improve prediction accuracy;(2)recognizing the challenge of overfitting and the inability of traditional noise-based data augmentation methods to effectively enhance model robustness,a novel learnable Gaussian noise mechanism is introduced that allows the model to adaptively optimize perturbations for different locations,ensuring more effective learning;and(3)to address the issue of error accumulation in autoregressive prediction,as well as the challenge of learning difficulty and the lack of intermediate data utilization in one-shot prediction,the authors propose a cascade prediction approach that effectively resolves these problems while significantly improving model forecasting performance.The method achieves a competitive result in The East China Regional AI Medium Range Weather Forecasting Competition.Ablation experiments further validate the effectiveness of each component,highlighting their contributions to enhancing prediction performance.
基金This work was supported by the Natural Science Foundation of Henan Province (004030600)
文摘Three model polyurethane hard segments based on dimethylol butanoic acid (DMBA) and 1,6-hexane diisocyanate (HDI), toluene diisocyanate (TDI) and 4,4'-diphenylmethane diisocyanate (MDI) were prepared by the solution method. Fourier Infrared (FTIR) spectroscopy was employed to study the H-bonds in these model polyurethanes. The model polyurethane hard segment prepared from HDI and 1,4-butanodiol (BDO) was used for comparison. It was found that the incorporation of the pendent carboxyl through DMBA into the model hard segments weakens the original NH…O = C H-bond but gives more H-bond patterns based on the two H-bond donors, urethane NH and carboxylic OH. The carboxylic dimer is one of the main H-bond types and is stronger than another main H-bond type NH…O=C. In addition, the H-bond in aromatic model hard segments is stronger than that of aliphatic hard segments. The appearance of the free C=O and the fact that almost all N—H is H-bonded suggest that there possibly exist either the third H-bond acceptor or the H-bond formed by one acceptor with two donors.
基金Supported by the National Key Basic Research and Development Program(No.2006CB701303, No. 2004CB318206)
文摘The texture analysis is often discussed in image processing domain, but most methods are limited within gray-level image or color image, and the present conception of texture is defined mainly based on gray-level image of single band. One of the essential characters of remote sensing image is multidimensional or even high-dimensional, and the traditional texture conception cannot contain enough information for these. Therefore, it is necessary to pursuit a proper texture definition based on remote sensing images, which is the first discussion in this paper. This paper describes the mapping model of spectral vector in two-dimensional image space using Markov random field (MRF), establishes a texture model of multiband remote sensing image based on MRF, and analyzes the calculations of Gibbs potential energy and Gibbs parameters. Further, this paper also analyzes the limitations of the traditional Gibbs model, prefers a new Gibbs model avoiding estimation of parameters, and presents a new texture segmentation algorithm for hy-perspectral remote sensing image later.
基金the National Institute of Child Health and Human Development (No.NICHD,No.R15HD059080,and No.R15HD059080-01A1S1)
文摘Background:The greater trochanter marker is commonly used in 3-dimensional(3D) models;however,its influence on hip and knee kinematics during gait is unclear.Understanding the influence of the greater trochanter marker is important when quantifying frontal and transverse plane hip and knee kinematics,parameters which are particularly relevant to investigate in individuals with conditions such as patellofemoral pain,knee osteoarthritis,anterior cruciate ligament(ACL) injury,and hip pain.The aim of this study was to evaluate the effect of including the greater trochanter in the construction of the thigh segment on hip and knee kinematics during gait.Methods:3D kinematics were collected in 19 healthy subjects during walking using a surface marker system.Hip and knee angles were compared across two thigh segment definitions(with and without greater trochanter) at two time points during stance:peak knee flexion(PKF) and minimum knee flexion(Min KF).Results:Hip and knee angles differed in magnitude and direction in the transverse plane at both time points.In the thigh model with the greater trochanter the hip was more externally rotated than in the thigh model without the greater trochanter(PKF:-9.34°± 5.21° vs.1.40°± 5.22°,Min KF:-5.68°± 4.24° vs.5.01°± 4.86°;p < 0.001).In the thigh model with the greater trochanter,the knee angle was more internally rotated compared to the knee angle calculated using the thigh definition without the greater trochanter(PKF:14.67°± 6.78° vs.4.33°± 4.18°,Min KF:10.54°± 6.71° vs.-0.01°± 2.69°;p < 0.001).Small but significant differences were detected in the sagittal and frontal plane angles at both time points(p < 0.001).Conclusion:Hip and knee kinematics differed across different segment definitions including or excluding the greater trochanter marker,especially in the transverse plane.Therefore when considering whether to include the greater trochanter in the thigh segment model when using a surface markers to calculate 3D kinematics for movement assessment,it is important to have a clear understanding of the effect of different marker sets and segment models in use.
基金supported by Sichuan Province Outstanding Young Scientist Fund(Grant No.2025NSFJQ0009)Sichuan Regional Innovation Cooperation Fund(Grant No.2025YFHZ0270)。
文摘The quantitative analysis of dispersed phases(bubbles,droplets,and particles)in multiphase flow systems represents a persistent technological challenge in petroleum engineering applications,including CO2-enhanced oil recovery,foam flooding,and unconventional reservoir development.Current characterization methods remain constrained by labor-intensive manual workflows and limited dynamic analysis capabilities,particularly for processing large-scale microscopy data and video sequences that capture critical transient behavior like gas cluster migration and droplet coalescence.These limitations hinder the establishment of robust correlations between pore-scale flow patterns and reservoir-scale production performance.This study introduces a novel computer vision framework that integrates foundation models with lightweight neural networks to address these industry challenges.Leveraging the segment anything model's zero-shot learning capability,we developed an automated workflow that achieves an efficiency improvement of approximately 29 times in bubble labeling compared to manual methods while maintaining less than 2%deviation from expert annotations.Engineering-oriented optimization ensures lightweight deployment with 94%segmentation accuracy,while the integrated quantification system precisely resolves gas saturation,shape factors,and interfacial dynamics,parameters critical for optimizing gas injection strategies and predicting phase redistribution patterns.Validated through microfluidic gas-liquid displacement experiments for discontinuous phase segmentation accuracy,this methodology enables precise bubble morphology quantification with broad application potential in multiphase systems,including emulsion droplet dynamics characterization and particle transport behavior analysis.This work bridges the critical gap between pore-scale dynamics characterization and reservoir-scale simulation requirements,providing a foundational framework for intelligent flow diagnostics and predictive modeling in next-generation digital oilfield systems.
基金supported in part by the Key National Research Project of China under Grant 2023YFC3707900in part by the National Natural Science Foundation of China under Grant 42204122 and Grant 42072323。
文摘CO_(2) storage capacity is significantly influenced by the saturation levels of reservoir rocks,with underground fluid saturation typically evaluated using resistivity data.The conductive pathways of fluids in various states within rock pores differ,alongside variations in conductive mechanisms.To clarify the conductivity of water in rocks across different states,this study employed a three-pore segment saturation model,which corrected for the additional conductivity of clay by categorizing water into large-pore segment,medium-pore segment,and small-pore segment types.Addressing the heterogeneity of tight sandstone reservoirs,we classified distinct pore structures and inverted Archie equation parameters from NMR logging data using a segmented characterization approach,yielding dynamic Archie parameters that vary with depth.Ultimately,we established an improved saturation parameter method based on joint inversion of NMR and resistivity data,which was validated through laboratory experiments and practical downhole applications.The results indicate that this saturation parameter inversion method has been effectively applied in both settings.Furthermore,we discussed the varying conductive behaviors of fluids in large and medium pore segment under saturated and drained states.Lastly,we proposed a workflow for inverting saturation based on downhole data,providing a robust foundation for CO_(2) storage and predicting underground fluid saturation.
文摘Large-scale unsupervised semantic segmentation(LUSS)is a sophisticated process that aims to segment similar areas within an image without relying on labeled training data.While existing methodologies have made substantial progress in this area,there is ample scope for enhancement.We thus introduce the PASS-SAM model,a comprehensive solution that amalgamates the benefits of various models to improve segmentation performance.
文摘Although the genetic algorithm (GA) has very powerful robustness and fitness, it needs a large size of population and a large number of iterations to reach the optimum result. Especially when GA is used in complex structural optimization problems, if the structural reanalysis technique is not adopted, the more the number of finite element analysis (FEA) is, the more the consuming time is. In the conventional structural optimization the number of FEA can be reduced by the structural reanalysis technique based on the approximation techniques and sensitivity analysis. With these techniques, this paper provides a new approximation model-segment approximation model, adopted for the GA application. This segment approximation model can decrease the number of FEA and increase the convergence rate of GA. So it can apparently decrease the computation time of GA. Two examples demonstrate the availability of the new segment approximation model.
基金Supported by Shenzhen Science and Technology Program(No.JCYJ20220530153604010).
文摘AIM:To develop a deep learning-based model for automatic retinal vascular segmentation,analyzing and comparing parameters under diverse glucose metabolic status(normal,prediabetes,diabetes)and to assess the potential of artificial intelligence(AI)in image segmentation and retinal vascular parameters for predicting prediabetes and diabetes.METHODS:Retinal fundus photos from 200 normal individuals,200 prediabetic patients,and 200 diabetic patients(600 eyes in total)were used.The U-Net network served as the foundational architecture for retinal arteryvein segmentation.An automatic segmentation and evaluation system for retinal vascular parameters was trained,encompassing 26 parameters.RESULTS:Significant differences were found in retinal vascular parameters across normal,prediabetes,and diabetes groups,including artery diameter(P=0.008),fractal dimension(P=0.000),vein curvature(P=0.003),C-zone artery branching vessel count(P=0.049),C-zone vein branching vessel count(P=0.041),artery branching angle(P=0.005),vein branching angle(P=0.001),artery angle asymmetry degree(P=0.003),vessel length density(P=0.000),and vessel area density(P=0.000),totaling 10 parameters.CONCLUSION:The deep learning-based model facilitates retinal vascular parameter identification and quantification,revealing significant differences.These parameters exhibit potential as biomarkers for prediabetes and diabetes.
基金supported by European Union’s Horizon Europe research and innovation programme,project AGILEHAND(Smart Grading,Handling and Packaging Solutions for Soft and Deformable Products in Agile and Reconfigurable Lines)(101092043).
文摘Dear Editor,This letter presents techniques to simplify dataset generation for instance segmentation of raw meat products,a critical step toward automating food production lines.Accurate segmentation is essential for addressing challenges such as occlusions,indistinct edges,and stacked configurations,which demand large,diverse datasets.To meet these demands,we propose two complementary approaches:a semi-automatic annotation interface using tools like the segment anything model(SAM)and GrabCut and a synthetic data generation pipeline leveraging 3D-scanned models.These methods reduce reliance on real meat,mitigate food waste,and improve scalability.Experimental results demonstrate that incorporating synthetic data enhances segmentation model performance and,when combined with real data,further boosts accuracy,paving the way for more efficient automation in the food industry.
