This study proposes an image-based three-dimensional(3D)vector reconstruction of industrial parts that can gener-ate non-uniform rational B-splines(NURBS)surfaces with high fidelity and flexibility.The contributions o...This study proposes an image-based three-dimensional(3D)vector reconstruction of industrial parts that can gener-ate non-uniform rational B-splines(NURBS)surfaces with high fidelity and flexibility.The contributions of this study include three parts:first,a dataset of two-dimensional images is constructed for typical industrial parts,including hex-agonal head bolts,cylindrical gears,shoulder rings,hexagonal nuts,and cylindrical roller bearings;second,a deep learning algorithm is developed for parameter extraction of 3D industrial parts,which can determine the final 3D parameters and pose information of the reconstructed model using two new nets,CAD-ClassNet and CAD-ReconNet;and finally,a 3D vector shape reconstruction of mechanical parts is presented to generate NURBS from the obtained shape parameters.The final reconstructed models show that the proposed approach is highly accurate,efficient,and practical.展开更多
Image super-resolution reconstruction technology is currently widely used in medical imaging,video surveillance,and industrial quality inspection.It not only enhances image quality but also improves details and visual...Image super-resolution reconstruction technology is currently widely used in medical imaging,video surveillance,and industrial quality inspection.It not only enhances image quality but also improves details and visual perception,significantly increasing the utility of low-resolution images.In this study,an improved image superresolution reconstruction model based on Generative Adversarial Networks(SRGAN)was proposed.This model introduced a channel and spatial attention mechanism(CSAB)in the generator,allowing it to effectively leverage the information from the input image to enhance feature representations and capture important details.The discriminator was designed with an improved PatchGAN architecture,which more accurately captured local details and texture information of the image.With these enhanced generator and discriminator architectures and an optimized loss function design,this method demonstrated superior performance in image quality assessment metrics.Experimental results showed that this model outperforms traditional methods,presenting more detailed and realistic image details in the visual effects.展开更多
As a form of discrete representation learning,Vector Quantized Variational Autoencoders(VQ-VAE)have increasingly been applied to generative and multimodal tasks due to their ease of embedding and representative capaci...As a form of discrete representation learning,Vector Quantized Variational Autoencoders(VQ-VAE)have increasingly been applied to generative and multimodal tasks due to their ease of embedding and representative capacity.However,existing VQ-VAEs often perform quantization in the spatial domain,ignoring global structural information and potentially suffering from codebook collapse and information coupling issues.This paper proposes a frequency quantized variational autoencoder(FQ-VAE)to address these issues.The proposed method transforms image features into linear combinations in the frequency domain using a 2D fast Fourier transform(2D-FFT)and performs adaptive quantization on these frequency components to preserve image’s global relationships.The codebook is dynamically optimized to avoid collapse and information coupling issue by considering the usage frequency and dependency of code vectors.Furthermore,we introduce a post-processing module based on graph convolutional networks to further improve reconstruction quality.Experimental results on four public datasets demonstrate that the proposed method outperforms state-of-the-art approaches in terms of Structural Similarity Index(SSIM),Learned Perceptual Image Patch Similarity(LPIPS),and Reconstruction Fréchet Inception Distance(rFID).In the experiments on the CIFAR-10 dataset,compared to the baselinemethod VQ-VAE,the proposedmethod improves the abovemetrics by 4.9%,36.4%,and 52.8%,respectively.展开更多
BACKGROUND Inguinal hernias are common after surgery.Tension-free repair is widely accepted as the main method for managing inguinal hernias.Adequate exposure,coverage,and repair of the myopectineal orifice(MPO)are ne...BACKGROUND Inguinal hernias are common after surgery.Tension-free repair is widely accepted as the main method for managing inguinal hernias.Adequate exposure,coverage,and repair of the myopectineal orifice(MPO)are necessary.However,due to differences in race and sex,people’s body shapes vary.According to European guidelines,the patch should measure 10 cm×15 cm.If any part of the MPO is dissected,injury to the nerves,vascular network,or organs may occur during surgery,thereby leading to inguinal discomfort,pain,and seroma formation after surgery.Therefore,accurate localization and measurement of the boundary of the MPO are crucial for selecting the optimal patch for inguinal hernia repair.AIM To compare the size of the MPO measured on three-dimensional multislice spiral computed tomography(CT)with that measured via laparoscopy and explore the relevant factors influencing the size of the MPO.METHODS Clinical data from 74 patients who underwent laparoscopic tension-free inguinal hernia repair at the General Surgery Department of the First Affiliated Hospital of Anhui University of Science and Technology between September 2022 and July 2024 were collected and analyzed retrospectively.Transabdominal preperitoneal was performed.Sixty-four males and 10 females,with an average age of 58.30±12.32 years,were included.The clinical data of the patients were collected.The boundary of the MPO was measured on three-dimensional CT images before surgery and then again during transabdominal preperitoneal.All the preoperative and intraoperative data were analyzed via paired t-tests.A t-test was used for comparisons of age,body mass index,and sex between the groups.In the comparative analysis,a P value less than 0.05 indicated a significant difference.RESULTS The boundaries of the MPO on 3-dimensional CT images measured 7.05±0.47 cm and 6.27±0.61 cm,and the area of the MPO was 19.54±3.33 cm^(2).The boundaries of the MPO during surgery were 7.18±0.51 cm and 6.17±0.40 cm.The errors were not statistically significant.However,the intraoperative BD(the width of the MPO,P=0.024,P<0.05)and preoperative AC(the length of the MPO,P=0.045,P<0.05)significantly differed according to sex.The AC and BD measurements before and during surgery were not significantly different according to age,body mass index,hernia side or hernia type(P>0.05).CONCLUSION The application of this technology can aid in determining the most appropriate dissection range and patch size.展开更多
Among hyperspectral imaging technologies, interferometric spectral imaging is widely used in remote sening due to advantages of large luminous flux and high resolution. However, with complicated mechanism, interferome...Among hyperspectral imaging technologies, interferometric spectral imaging is widely used in remote sening due to advantages of large luminous flux and high resolution. However, with complicated mechanism, interferometric imaging faces the impact of multi-stage degradation. Most exsiting interferometric spectrum reconstruction methods are based on tradition model-based framework with multiple steps, showing poor efficiency and restricted performance. Thus, we propose an interferometric spectrum reconstruction method based on degradation synthesis and deep learning.Firstly, based on imaging mechanism, we proposed an mathematical model of interferometric imaging to analyse the degradation components as noises and trends during imaging. The model consists of three stages, namely instrument degradation, sensing degradation, and signal-independent degradation process. Then, we designed calibration-based method to estimate parameters in the model, of which the results are used for synthesizing realistic dataset for learning-based algorithms.In addition, we proposed a dual-stage interferogram spectrum reconstruction framework, which supports pre-training and integration of denoising DNNs. Experiments exhibits the reliability of our degradation model and synthesized data, and the effectiveness of the proposed reconstruction method.展开更多
Deep learning(DL)-based image reconstruction methods have garnered increasing interest in the last few years.Numerous studies demonstrate that DL-based reconstruction methods function admirably in optical tomographic ...Deep learning(DL)-based image reconstruction methods have garnered increasing interest in the last few years.Numerous studies demonstrate that DL-based reconstruction methods function admirably in optical tomographic imaging techniques,such as bioluminescence tomography(BLT).Nevertheless,nearly every existing DL-based method utilizes an explicit neural representation for the reconstruction problem,which either consumes much memory space or requires various complicated computations.In this paper,we present a neural field(NF)-based image reconstruction scheme for BLT that uses an implicit neural representation.The proposed NFbased method establishes a transformation between the coordinate of an arbitrary spatial point and the source value of the point with a relatively light-weight multilayer perceptron,which has remarkable computational efficiency.Another simple neural network composed of two fully connected layers and a 1D convolutional layer is used to generate the neural features.Results of simulations and experiments show that the proposed NF-based method has similar performance to the photon density complement network and the two-stage network,while consuming fewer floating point operations with fewer model parameters.展开更多
In Hyperspectral Imaging(HSI),the detrimental influence of noise and distortions on data quality is profound,which has severely affected the following-on analytics and decisionmaking such as land mapping.This study pr...In Hyperspectral Imaging(HSI),the detrimental influence of noise and distortions on data quality is profound,which has severely affected the following-on analytics and decisionmaking such as land mapping.This study presents an innovative framework for assessing HSI band quality and reconstructing the low-quality bands,based on the Prophet model.By introducing a comprehensive quality metric to start,the authors approach factors in both spatial and spectral characteristics across local and global scales.This metric effectively captures the intricate noise and distortions inherent in the HSI data.Subsequently,the authors employ the Prophet model to forecast information within the low-quality bands,leveraging insights from neighbouring high-quality bands.To validate the effectiveness of the authors’proposed model,extensive experiments on three publicly available uncorrected datasets are conducted.In a head-to-head comparison,the framework against six state-ofthe-art band reconstruction algorithms including three spectral methods,two spatialspectral methods and one deep learning method is benchmarked.The authors’experiments also delve into strategies for band selection based on quality metrics and the quality evaluation of the reconstructed bands.In addition,the authors assess the classification accuracy utilising these reconstructed bands.In various experiments,the results consistently affirm the efficacy of the authors’method in HSI quality assessment and band reconstruction.Notably,the authors’approach obviates the need for manually prefiltering of noisy bands.This comprehensive framework holds promise in addressing HSI data quality concerns whilst enhancing the overall utility of HSI.展开更多
In order to obtain a better sandstone three-dimensional （3D） reconstruction result which is more similar to the original sample, an algorithm based on stationarity for a two-dimensional （2D） training image is prop...In order to obtain a better sandstone three-dimensional （3D） reconstruction result which is more similar to the original sample, an algorithm based on stationarity for a two-dimensional （2D） training image is proposed. The second-order statistics based on texture features are analyzed to evaluate the scale stationarity of the training image. The multiple-point statistics of the training image are applied to obtain the multiple-point statistics stationarity estimation by the multi-point density function. The results show that the reconstructed 3D structures are closer to reality when the training image has better scale stationarity and multiple-point statistics stationarity by the indications of local percolation probability and two-point probability. Moreover, training images with higher multiple-point statistics stationarity and lower scale stationarity are likely to obtain closer results to the real 3D structure, and vice versa. Thus, stationarity analysis of the training image has far-reaching significance in choosing a better 2D thin section image for the 3D reconstruction of porous media. Especially, high-order statistics perform better than low-order statistics.展开更多
A method and procedure is presented to reconstruct three-dimensional(3D) positions of scattering centers from multiple synthetic aperture radar(SAR) images. Firstly, two-dimensional(2D) attribute scattering centers of...A method and procedure is presented to reconstruct three-dimensional(3D) positions of scattering centers from multiple synthetic aperture radar(SAR) images. Firstly, two-dimensional(2D) attribute scattering centers of targets are extracted from 2D SAR images. Secondly, similarity measure is developed based on 2D attributed scatter centers' location, type, and radargrammetry principle between multiple SAR images. By this similarity, we can associate 2D scatter centers and then obtain candidate 3D scattering centers. Thirdly, these candidate scattering centers are clustered in 3D space to reconstruct final 3D positions. Compared with presented methods, the proposed method has a capability of describing distributed scattering center, reduces false and missing 3D scattering centers, and has fewer restrictionson modeling data. Finally, results of experiments have demonstrated the effectiveness of the proposed method.展开更多
We establish an improved GP iterative algorithm for the extrapolation of band-limited function to fully 3-dimensional image reconstruction by the convolution-backprojection algorithm. Numerical experiments demonstrate...We establish an improved GP iterative algorithm for the extrapolation of band-limited function to fully 3-dimensional image reconstruction by the convolution-backprojection algorithm. Numerical experiments demonstrate that the image resolving power of IGP algorithm is better than that of the original GP algorithm for noisy data.展开更多
A novel technique of three-dimensional (3D) reconstruction, segmentation, display and analysis of series slices of images including microscopic wide field optical sectioning by deconvolution method, cryo-electron micr...A novel technique of three-dimensional (3D) reconstruction, segmentation, display and analysis of series slices of images including microscopic wide field optical sectioning by deconvolution method, cryo-electron microscope slices by Fou-rier-Bessel synthesis and electron tomography (ET), and a series of computed tomography (CT) was developed to perform si-multaneous measurement on the structure and function of biomedical samples. The paper presents the 3D reconstruction seg-mentation display and analysis results of pollen spore, chaperonin, virus, head, cervical bone, tibia and carpus. At the same time, it also puts forward some potential applications of the new technique in the biomedical realm.展开更多
The three-dimensional(3D)model is of great significance to analyze the performance of nonwovens.However,the existing modelling methods could not reconstruct the 3D structure of nonwovens at low cost.A new method based...The three-dimensional(3D)model is of great significance to analyze the performance of nonwovens.However,the existing modelling methods could not reconstruct the 3D structure of nonwovens at low cost.A new method based on deep learning was proposed to reconstruct 3D models of nonwovens from multi-focus images.A convolutional neural network was trained to extract clear fibers from sequence images.Image processing algorithms were used to obtain the radius,the central axis,and depth information of fibers from the extraction results.Based on this information,3D models were built in 3D space.Furthermore,self-developed algorithms optimized the central axis and depth of fibers,which made fibers more realistic and continuous.The method with lower cost could reconstruct 3D models of nonwovens conveniently.展开更多
BACKGROUND Sigmoid colon cancer faces challenges due to anatomical diversity,including variable inferior mesenteric artery(IMA)branching and tumor localization complexities,which increase intraoperative risks.AIM To c...BACKGROUND Sigmoid colon cancer faces challenges due to anatomical diversity,including variable inferior mesenteric artery(IMA)branching and tumor localization complexities,which increase intraoperative risks.AIM To comprehensively evaluate the impact of three-dimensional(3D)visualization technology on enhancing surgical precision and safety,as well as optimizing perioperative outcomes in laparoscopic sigmoid cancer resection.METHODS A prospective cohort of 106 patients(January 2023 to December 2024)undergoing laparoscopic sigmoid cancer resection was divided into the 3D(n=55)group and the control(n=51)group.The 3D group underwent preoperative enhanced computed tomography reconstruction(3D Slicer 5.2.2&Mimics 19.0).3D reconstruction visualization navigation intraoperatively guided the following key steps:Tumor location,Toldt’s space dissection,IMA ligation level selection,regional lymph node dissection,and marginal artery preservation.Outcomes included operative parameters,lymph node yield,and recovery metrics.RESULTS The 3D group demonstrated a significantly shorter operative time(172.91±20.69 minutes vs 190.29±32.29 minutes;P=0.002),reduced blood loss(31.5±11.8 mL vs 44.1±23.4 mL,P=0.001),earlier postoperative flatus(2.23±0.54 days vs 2.53±0.61 days;P=0.013),shorter hospital length of stay(13.47±1.74 days vs 16.20±7.71 days;P=0.013),shorter postoperative length of stay(8.6±2.6 days vs 10.5±4.9 days;P=0.014),and earlier postoperative exhaust time(2.23±0.54 days vs 2.53±0.61 days;P=0.013).Furthermore,the 3D group exhibited a higher mean number of lymph nodes harvested(16.91±5.74 vs 14.45±5.66;P=0.030).CONCLUSION The 3D visualization technology effectively addresses sigmoid colon anatomical complexity through surgical navigation,improving procedural safety and efficiency.展开更多
BACKGROUND Deep learning-based super-resolution(SR)reconstruction can obtain high-quality images with more detailed information.AIM To compare multiparametric normal-resolution(NR)and SR magnetic resonance imaging(MRI...BACKGROUND Deep learning-based super-resolution(SR)reconstruction can obtain high-quality images with more detailed information.AIM To compare multiparametric normal-resolution(NR)and SR magnetic resonance imaging(MRI)in predicting the histopathologic grade in hepatocellular carcinoma.METHODS We retrospectively analyzed a total of 826 patients from two medical centers(training 459;validation 196;test 171).T2-weighted imaging,diffusion-weighted imaging,and portal venous phases were collected.Tumor segmentations were conducted automatically by 3D U-Net.Based on generative adversarial network,we utilized 3D SR reconstruction to produce SR MRI.Radiomics models were developed and validated by XGBoost and Catboost.The predictive efficiency was demonstrated by calibration curves,decision curve analysis,area under the curve(AUC)and net reclassification index(NRI).RESULTS We extracted 3045 radiomic features from both NR and SR MRI,retaining 29 and 28 features,respectively.For XGBoost models,SR MRI yielded higher AUC value than NR MRI in the validation and test cohorts(0.83 vs 0.79;0.80 vs 0.78),respectively.Consistent trends were seen in CatBoost models:SR MRI achieved AUCs of 0.89 and 0.80 compared to NR MRI’s 0.81 and 0.76.NRI indicated that the SR MRI models could improve the prediction accuracy by-1.6%to 20.9%compared to the NR MRI models.CONCLUSION Deep learning-based SR MRI could improve the predictive performance of histopathologic grade in HCC.It may be a powerful tool for better stratification management for patients with operable HCC.展开更多
BACKGROUND Hepatobiliary surgery is complex and requires a thorough understanding of the liver’s anatomy,biliary system,and vasculature.Traditional imaging methods such as computed tomography(CT)and magnetic resonanc...BACKGROUND Hepatobiliary surgery is complex and requires a thorough understanding of the liver’s anatomy,biliary system,and vasculature.Traditional imaging methods such as computed tomography(CT)and magnetic resonance imaging(MRI),although helpful,fail to provide three-dimensional(3D)relationships of these structures,which are critical for planning and executing complicated surgeries.AIM To explore the use of 3D imaging and virtual surgical planning(VSP)technologies to improve surgical accuracy,reduce complications,and enhance patient recovery in hepatobiliary surgeries.METHODS A comprehensive review of studies published between 2017 and 2024 was conducted through PubMed,Scopus,Google Scholar,and Web of Science.Studies selected focused on 3D imaging and VSP applications in hepatobiliary surgery,assessing surgical precision,complications,and patient outcomes.Thirty studies,including randomized controlled trials,cohort studies,and case reports,were included in the final analysis.RESULTS Various 3D imaging modalities,including multidetector CT,MRI,and 3D rotational angiography,provide high-resolution views of the liver’s vascular and biliary anatomy.VSP allows surgeons to simulate complex surgeries,improving preoperative planning and reducing complications like bleeding and bile leaks.Several studies have demonstrated improved surgical precision,reduced complications,and faster recovery times when 3D imaging and VSP were used in complex surgeries.CONCLUSION 3D imaging and VSP technologies significantly enhance the accuracy and outcomes of hepatobiliary surgeries by providing individualized preoperative planning.While promising,further research,particularly randomized controlled trials,is needed to standardize protocols and evaluate long-term efficacy.展开更多
In order to improve the reconstruction accuracy of magnetic resonance imaging(MRI),an accurate natural image compressed sensing(CS)reconstruction network is proposed,which combines the advantages of model-based and de...In order to improve the reconstruction accuracy of magnetic resonance imaging(MRI),an accurate natural image compressed sensing(CS)reconstruction network is proposed,which combines the advantages of model-based and deep learning-based CS-MRI methods.In theory,enhancing geometric texture details in linear reconstruction is possible.First,the optimization problem is decomposed into two problems:linear approximation and geometric compensation.Aimed at the problem of image linear approximation,the data consistency module is used to deal with it.Since the processing process will lose texture details,a neural network layer that explicitly combines image and frequency feature representation is proposed,which is named butterfly dilated geometric distillation network.The network introduces the idea of butterfly operation,skillfully integrates the features of image domain and frequency domain,and avoids the loss of texture details when extracting features in a single domain.Finally,a channel feature fusion module is designed by combining channel attention mechanism and dilated convolution.The attention of the channel makes the final output feature map focus on the more important part,thus improving the feature representation ability.The dilated convolution enlarges the receptive field,thereby obtaining more dense image feature data.The experimental results show that the peak signal-to-noise ratio of the network is 5.43 dB,5.24 dB and 3.89 dB higher than that of ISTA-Net+,FISTA and DGDN networks on the brain data set with a Cartesian sampling mask CS ratio of 10%.展开更多
Stochastic optical reconstruction microscopy(STORM),as a typical technique of single-molecule localization microscopy(SMLM),has overcome the diffraction limit by randomly switching fluorophores between fluorescent and...Stochastic optical reconstruction microscopy(STORM),as a typical technique of single-molecule localization microscopy(SMLM),has overcome the diffraction limit by randomly switching fluorophores between fluorescent and dark states,allowing for the precise localization of isolated emission patterns and the super-resolution reconstruction from millions of localized positions of single fluorophores.A critical factor influencing localization precision is the photo-switching behavior of fluorophores,which is affected by the imaging buffer.The imaging buffer typically comprises oxygen scavengers,photo-switching reagents,and refractive index regulators.Oxygen scavengers help prevent photobleaching,photo-switching reagents assist in facilitating the conversion of fluorophores,and refractive index regulators are used to adjust the refractive index of the solution.The synergistic interaction of these components promotes stable blinking of fluorophores,reduces irreversible photobleaching,and thereby ensures high-quality super-resolution imaging.This review provides a comprehensive overview of the essential compositions and functionalities of imaging buffers used in STORM,serving as a valuable resource for researchers seeking to select appropriate imaging buffers for their experiments.展开更多
A new medical image fusion technique is presented.The method is based on three-dimensional reconstruction.After reconstruction,the three-dimensional volume data is normalized by three-dimensional coordinate conversion...A new medical image fusion technique is presented.The method is based on three-dimensional reconstruction.After reconstruction,the three-dimensional volume data is normalized by three-dimensional coordinate conversion in the same way and intercepted through setting up cutting plane including anatomical structure,as a result two images in entire registration on space and geometry are obtained and the images are fused at last.Compared with traditional two-dimensional fusion technique,three-dimensional fusion technique can not only resolve the different problems existed in the two kinds of images,but also avoid the registration error of the two kinds of images when they have different scan and imaging parameter.The research proves this fusion technique is more exact and has no registration,so it is more adapt to arbitrary medical image fusion with different equipments.展开更多
Neutron capture event imaging is a novel technique that has the potential to substantially enhance the resolution of existing imaging systems.This study provides a measurement method for neutron capture event distribu...Neutron capture event imaging is a novel technique that has the potential to substantially enhance the resolution of existing imaging systems.This study provides a measurement method for neutron capture event distribution along with multiple reconstruction methods for super-resolution imaging.The proposed technology reduces the point-spread function of an imag-ing system through single-neutron detection and event reconstruction,thereby significantly improving imaging resolution.A single-neutron detection experiment was conducted using a highly practical and efficient^(6)LiF-ZnS scintillation screen of a cold neutron imaging device in the research reactor.In milliseconds of exposure time,a large number of weak light clusters and their distribution in the scintillation screen were recorded frame by frame,to complete single-neutron detection.Several reconstruction algorithms were proposed for the calculations.The location of neutron capture was calculated using several processing methods such as noise removal,filtering,spot segmentation,contour analysis,and local positioning.The proposed algorithm achieved a higher imaging resolution and faster reconstruction speed,and single-neutron super-resolution imaging was realized by combining single-neutron detection experiments and reconstruction calculations.The results show that the resolution of the 100μm thick^(6)LiF-ZnS scintillation screen can be improved from 125 to 40 microns.This indicates that the proposed single-neutron detection and calculation method is effective and can significantly improve imaging resolution.展开更多
Accurate segmentation of camouflage objects in aerial imagery is vital for improving the efficiency of UAV-based reconnaissance and rescue missions.However,camouflage object segmentation is increasingly challenging du...Accurate segmentation of camouflage objects in aerial imagery is vital for improving the efficiency of UAV-based reconnaissance and rescue missions.However,camouflage object segmentation is increasingly challenging due to advances in both camouflage materials and biological mimicry.Although multispectral-RGB based technology shows promise,conventional dual-aperture multispectral-RGB imaging systems are constrained by imprecise and time-consuming registration and fusion across different modalities,limiting their performance.Here,we propose the Reconstructed Multispectral-RGB Fusion Network(RMRF-Net),which reconstructs RGB images into multispectral ones,enabling efficient multimodal segmentation using only an RGB camera.Specifically,RMRF-Net employs a divergentsimilarity feature correction strategy to minimize reconstruction errors and includes an efficient boundary-aware decoder to enhance object contours.Notably,we establish the first real-world aerial multispectral-RGB semantic segmentation of camouflage objects dataset,including 11 object categories.Experimental results demonstrate that RMRF-Net outperforms existing methods,achieving 17.38 FPS on the NVIDIA Jetson AGX Orin,with only a 0.96%drop in mIoU compared to the RTX 3090,showing its practical applicability in multimodal remote sensing.展开更多
基金supported by the Aeronautical Science Foundation of China,No.2023Z0680510022021 Special Scientific Research on Civil Aircraft Project+1 种基金the Natural Science Foundation of China,Nos.61572056 and 61872347the Special Plan for the Development of Distinguished Young Scientists of ISCAS,No.Y8RC535018.
文摘This study proposes an image-based three-dimensional(3D)vector reconstruction of industrial parts that can gener-ate non-uniform rational B-splines(NURBS)surfaces with high fidelity and flexibility.The contributions of this study include three parts:first,a dataset of two-dimensional images is constructed for typical industrial parts,including hex-agonal head bolts,cylindrical gears,shoulder rings,hexagonal nuts,and cylindrical roller bearings;second,a deep learning algorithm is developed for parameter extraction of 3D industrial parts,which can determine the final 3D parameters and pose information of the reconstructed model using two new nets,CAD-ClassNet and CAD-ReconNet;and finally,a 3D vector shape reconstruction of mechanical parts is presented to generate NURBS from the obtained shape parameters.The final reconstructed models show that the proposed approach is highly accurate,efficient,and practical.
文摘Image super-resolution reconstruction technology is currently widely used in medical imaging,video surveillance,and industrial quality inspection.It not only enhances image quality but also improves details and visual perception,significantly increasing the utility of low-resolution images.In this study,an improved image superresolution reconstruction model based on Generative Adversarial Networks(SRGAN)was proposed.This model introduced a channel and spatial attention mechanism(CSAB)in the generator,allowing it to effectively leverage the information from the input image to enhance feature representations and capture important details.The discriminator was designed with an improved PatchGAN architecture,which more accurately captured local details and texture information of the image.With these enhanced generator and discriminator architectures and an optimized loss function design,this method demonstrated superior performance in image quality assessment metrics.Experimental results showed that this model outperforms traditional methods,presenting more detailed and realistic image details in the visual effects.
基金supported by the Interdisciplinary project of Dalian University DLUXK-2023-ZD-001.
文摘As a form of discrete representation learning,Vector Quantized Variational Autoencoders(VQ-VAE)have increasingly been applied to generative and multimodal tasks due to their ease of embedding and representative capacity.However,existing VQ-VAEs often perform quantization in the spatial domain,ignoring global structural information and potentially suffering from codebook collapse and information coupling issues.This paper proposes a frequency quantized variational autoencoder(FQ-VAE)to address these issues.The proposed method transforms image features into linear combinations in the frequency domain using a 2D fast Fourier transform(2D-FFT)and performs adaptive quantization on these frequency components to preserve image’s global relationships.The codebook is dynamically optimized to avoid collapse and information coupling issue by considering the usage frequency and dependency of code vectors.Furthermore,we introduce a post-processing module based on graph convolutional networks to further improve reconstruction quality.Experimental results on four public datasets demonstrate that the proposed method outperforms state-of-the-art approaches in terms of Structural Similarity Index(SSIM),Learned Perceptual Image Patch Similarity(LPIPS),and Reconstruction Fréchet Inception Distance(rFID).In the experiments on the CIFAR-10 dataset,compared to the baselinemethod VQ-VAE,the proposedmethod improves the abovemetrics by 4.9%,36.4%,and 52.8%,respectively.
基金Supported by the 2022 Provincial Quality Engineering Project for Higher Education Institutions,No.2022sx031the 2023 Provincial Quality Engineering Project for Higher Education Institutions,No.2023jyxm1071.
文摘BACKGROUND Inguinal hernias are common after surgery.Tension-free repair is widely accepted as the main method for managing inguinal hernias.Adequate exposure,coverage,and repair of the myopectineal orifice(MPO)are necessary.However,due to differences in race and sex,people’s body shapes vary.According to European guidelines,the patch should measure 10 cm×15 cm.If any part of the MPO is dissected,injury to the nerves,vascular network,or organs may occur during surgery,thereby leading to inguinal discomfort,pain,and seroma formation after surgery.Therefore,accurate localization and measurement of the boundary of the MPO are crucial for selecting the optimal patch for inguinal hernia repair.AIM To compare the size of the MPO measured on three-dimensional multislice spiral computed tomography(CT)with that measured via laparoscopy and explore the relevant factors influencing the size of the MPO.METHODS Clinical data from 74 patients who underwent laparoscopic tension-free inguinal hernia repair at the General Surgery Department of the First Affiliated Hospital of Anhui University of Science and Technology between September 2022 and July 2024 were collected and analyzed retrospectively.Transabdominal preperitoneal was performed.Sixty-four males and 10 females,with an average age of 58.30±12.32 years,were included.The clinical data of the patients were collected.The boundary of the MPO was measured on three-dimensional CT images before surgery and then again during transabdominal preperitoneal.All the preoperative and intraoperative data were analyzed via paired t-tests.A t-test was used for comparisons of age,body mass index,and sex between the groups.In the comparative analysis,a P value less than 0.05 indicated a significant difference.RESULTS The boundaries of the MPO on 3-dimensional CT images measured 7.05±0.47 cm and 6.27±0.61 cm,and the area of the MPO was 19.54±3.33 cm^(2).The boundaries of the MPO during surgery were 7.18±0.51 cm and 6.17±0.40 cm.The errors were not statistically significant.However,the intraoperative BD(the width of the MPO,P=0.024,P<0.05)and preoperative AC(the length of the MPO,P=0.045,P<0.05)significantly differed according to sex.The AC and BD measurements before and during surgery were not significantly different according to age,body mass index,hernia side or hernia type(P>0.05).CONCLUSION The application of this technology can aid in determining the most appropriate dissection range and patch size.
文摘Among hyperspectral imaging technologies, interferometric spectral imaging is widely used in remote sening due to advantages of large luminous flux and high resolution. However, with complicated mechanism, interferometric imaging faces the impact of multi-stage degradation. Most exsiting interferometric spectrum reconstruction methods are based on tradition model-based framework with multiple steps, showing poor efficiency and restricted performance. Thus, we propose an interferometric spectrum reconstruction method based on degradation synthesis and deep learning.Firstly, based on imaging mechanism, we proposed an mathematical model of interferometric imaging to analyse the degradation components as noises and trends during imaging. The model consists of three stages, namely instrument degradation, sensing degradation, and signal-independent degradation process. Then, we designed calibration-based method to estimate parameters in the model, of which the results are used for synthesizing realistic dataset for learning-based algorithms.In addition, we proposed a dual-stage interferogram spectrum reconstruction framework, which supports pre-training and integration of denoising DNNs. Experiments exhibits the reliability of our degradation model and synthesized data, and the effectiveness of the proposed reconstruction method.
基金supported in part by the National Natural Science Foundation of China(62101278,62001379,62271023)Beijing Natural Science Foundation(7242269).
文摘Deep learning(DL)-based image reconstruction methods have garnered increasing interest in the last few years.Numerous studies demonstrate that DL-based reconstruction methods function admirably in optical tomographic imaging techniques,such as bioluminescence tomography(BLT).Nevertheless,nearly every existing DL-based method utilizes an explicit neural representation for the reconstruction problem,which either consumes much memory space or requires various complicated computations.In this paper,we present a neural field(NF)-based image reconstruction scheme for BLT that uses an implicit neural representation.The proposed NFbased method establishes a transformation between the coordinate of an arbitrary spatial point and the source value of the point with a relatively light-weight multilayer perceptron,which has remarkable computational efficiency.Another simple neural network composed of two fully connected layers and a 1D convolutional layer is used to generate the neural features.Results of simulations and experiments show that the proposed NF-based method has similar performance to the photon density complement network and the two-stage network,while consuming fewer floating point operations with fewer model parameters.
基金National Natural Science Foundation Major Project of China,Grant/Award Number:42192580Guangdong Province Key Construction Discipline Scientific Research Ability Promotion Project,Grant/Award Number:2022ZDJS015。
文摘In Hyperspectral Imaging(HSI),the detrimental influence of noise and distortions on data quality is profound,which has severely affected the following-on analytics and decisionmaking such as land mapping.This study presents an innovative framework for assessing HSI band quality and reconstructing the low-quality bands,based on the Prophet model.By introducing a comprehensive quality metric to start,the authors approach factors in both spatial and spectral characteristics across local and global scales.This metric effectively captures the intricate noise and distortions inherent in the HSI data.Subsequently,the authors employ the Prophet model to forecast information within the low-quality bands,leveraging insights from neighbouring high-quality bands.To validate the effectiveness of the authors’proposed model,extensive experiments on three publicly available uncorrected datasets are conducted.In a head-to-head comparison,the framework against six state-ofthe-art band reconstruction algorithms including three spectral methods,two spatialspectral methods and one deep learning method is benchmarked.The authors’experiments also delve into strategies for band selection based on quality metrics and the quality evaluation of the reconstructed bands.In addition,the authors assess the classification accuracy utilising these reconstructed bands.In various experiments,the results consistently affirm the efficacy of the authors’method in HSI quality assessment and band reconstruction.Notably,the authors’approach obviates the need for manually prefiltering of noisy bands.This comprehensive framework holds promise in addressing HSI data quality concerns whilst enhancing the overall utility of HSI.
基金The National Natural Science Foundation of China(No.60972130)
文摘In order to obtain a better sandstone three-dimensional （3D） reconstruction result which is more similar to the original sample, an algorithm based on stationarity for a two-dimensional （2D） training image is proposed. The second-order statistics based on texture features are analyzed to evaluate the scale stationarity of the training image. The multiple-point statistics of the training image are applied to obtain the multiple-point statistics stationarity estimation by the multi-point density function. The results show that the reconstructed 3D structures are closer to reality when the training image has better scale stationarity and multiple-point statistics stationarity by the indications of local percolation probability and two-point probability. Moreover, training images with higher multiple-point statistics stationarity and lower scale stationarity are likely to obtain closer results to the real 3D structure, and vice versa. Thus, stationarity analysis of the training image has far-reaching significance in choosing a better 2D thin section image for the 3D reconstruction of porous media. Especially, high-order statistics perform better than low-order statistics.
文摘A method and procedure is presented to reconstruct three-dimensional(3D) positions of scattering centers from multiple synthetic aperture radar(SAR) images. Firstly, two-dimensional(2D) attribute scattering centers of targets are extracted from 2D SAR images. Secondly, similarity measure is developed based on 2D attributed scatter centers' location, type, and radargrammetry principle between multiple SAR images. By this similarity, we can associate 2D scatter centers and then obtain candidate 3D scattering centers. Thirdly, these candidate scattering centers are clustered in 3D space to reconstruct final 3D positions. Compared with presented methods, the proposed method has a capability of describing distributed scattering center, reduces false and missing 3D scattering centers, and has fewer restrictionson modeling data. Finally, results of experiments have demonstrated the effectiveness of the proposed method.
基金G.R. Qu is partially supported by the National Natural Science Foundation of China (No. 60372015, 60772041) the Science Foundation of Beijing Jiaotong University (2002SM54)+2 种基金M. Jiang is partially supported by the National Key Basic Research Special Foundation of China (2003CB716101) National Science Foundation of China (60325101, 60272018, 60628102) Ministry of Education (306017), Engineering Research Institute of Peking University, and Microsoft Research Asia.
文摘We establish an improved GP iterative algorithm for the extrapolation of band-limited function to fully 3-dimensional image reconstruction by the convolution-backprojection algorithm. Numerical experiments demonstrate that the image resolving power of IGP algorithm is better than that of the original GP algorithm for noisy data.
文摘A novel technique of three-dimensional (3D) reconstruction, segmentation, display and analysis of series slices of images including microscopic wide field optical sectioning by deconvolution method, cryo-electron microscope slices by Fou-rier-Bessel synthesis and electron tomography (ET), and a series of computed tomography (CT) was developed to perform si-multaneous measurement on the structure and function of biomedical samples. The paper presents the 3D reconstruction seg-mentation display and analysis results of pollen spore, chaperonin, virus, head, cervical bone, tibia and carpus. At the same time, it also puts forward some potential applications of the new technique in the biomedical realm.
基金National Natural Science Foundation of China(No.61771123)。
文摘The three-dimensional(3D)model is of great significance to analyze the performance of nonwovens.However,the existing modelling methods could not reconstruct the 3D structure of nonwovens at low cost.A new method based on deep learning was proposed to reconstruct 3D models of nonwovens from multi-focus images.A convolutional neural network was trained to extract clear fibers from sequence images.Image processing algorithms were used to obtain the radius,the central axis,and depth information of fibers from the extraction results.Based on this information,3D models were built in 3D space.Furthermore,self-developed algorithms optimized the central axis and depth of fibers,which made fibers more realistic and continuous.The method with lower cost could reconstruct 3D models of nonwovens conveniently.
基金Supported by the Health Commission of Fuyang City,Anhui,China,No.FY2023-45Fuyang Municipal Science and Technology Bureau,Anhui,China,No.FK20245505+1 种基金Anhui Provincial Health Commission,No.AHWJ2023Baa20164Bengbu Medical University,No.2023byzd215.
文摘BACKGROUND Sigmoid colon cancer faces challenges due to anatomical diversity,including variable inferior mesenteric artery(IMA)branching and tumor localization complexities,which increase intraoperative risks.AIM To comprehensively evaluate the impact of three-dimensional(3D)visualization technology on enhancing surgical precision and safety,as well as optimizing perioperative outcomes in laparoscopic sigmoid cancer resection.METHODS A prospective cohort of 106 patients(January 2023 to December 2024)undergoing laparoscopic sigmoid cancer resection was divided into the 3D(n=55)group and the control(n=51)group.The 3D group underwent preoperative enhanced computed tomography reconstruction(3D Slicer 5.2.2&Mimics 19.0).3D reconstruction visualization navigation intraoperatively guided the following key steps:Tumor location,Toldt’s space dissection,IMA ligation level selection,regional lymph node dissection,and marginal artery preservation.Outcomes included operative parameters,lymph node yield,and recovery metrics.RESULTS The 3D group demonstrated a significantly shorter operative time(172.91±20.69 minutes vs 190.29±32.29 minutes;P=0.002),reduced blood loss(31.5±11.8 mL vs 44.1±23.4 mL,P=0.001),earlier postoperative flatus(2.23±0.54 days vs 2.53±0.61 days;P=0.013),shorter hospital length of stay(13.47±1.74 days vs 16.20±7.71 days;P=0.013),shorter postoperative length of stay(8.6±2.6 days vs 10.5±4.9 days;P=0.014),and earlier postoperative exhaust time(2.23±0.54 days vs 2.53±0.61 days;P=0.013).Furthermore,the 3D group exhibited a higher mean number of lymph nodes harvested(16.91±5.74 vs 14.45±5.66;P=0.030).CONCLUSION The 3D visualization technology effectively addresses sigmoid colon anatomical complexity through surgical navigation,improving procedural safety and efficiency.
基金Supported by AI+Health Collaborative Innovation Cultivation Project of Beijing City,No.Z221100003522005.
文摘BACKGROUND Deep learning-based super-resolution(SR)reconstruction can obtain high-quality images with more detailed information.AIM To compare multiparametric normal-resolution(NR)and SR magnetic resonance imaging(MRI)in predicting the histopathologic grade in hepatocellular carcinoma.METHODS We retrospectively analyzed a total of 826 patients from two medical centers(training 459;validation 196;test 171).T2-weighted imaging,diffusion-weighted imaging,and portal venous phases were collected.Tumor segmentations were conducted automatically by 3D U-Net.Based on generative adversarial network,we utilized 3D SR reconstruction to produce SR MRI.Radiomics models were developed and validated by XGBoost and Catboost.The predictive efficiency was demonstrated by calibration curves,decision curve analysis,area under the curve(AUC)and net reclassification index(NRI).RESULTS We extracted 3045 radiomic features from both NR and SR MRI,retaining 29 and 28 features,respectively.For XGBoost models,SR MRI yielded higher AUC value than NR MRI in the validation and test cohorts(0.83 vs 0.79;0.80 vs 0.78),respectively.Consistent trends were seen in CatBoost models:SR MRI achieved AUCs of 0.89 and 0.80 compared to NR MRI’s 0.81 and 0.76.NRI indicated that the SR MRI models could improve the prediction accuracy by-1.6%to 20.9%compared to the NR MRI models.CONCLUSION Deep learning-based SR MRI could improve the predictive performance of histopathologic grade in HCC.It may be a powerful tool for better stratification management for patients with operable HCC.
文摘BACKGROUND Hepatobiliary surgery is complex and requires a thorough understanding of the liver’s anatomy,biliary system,and vasculature.Traditional imaging methods such as computed tomography(CT)and magnetic resonance imaging(MRI),although helpful,fail to provide three-dimensional(3D)relationships of these structures,which are critical for planning and executing complicated surgeries.AIM To explore the use of 3D imaging and virtual surgical planning(VSP)technologies to improve surgical accuracy,reduce complications,and enhance patient recovery in hepatobiliary surgeries.METHODS A comprehensive review of studies published between 2017 and 2024 was conducted through PubMed,Scopus,Google Scholar,and Web of Science.Studies selected focused on 3D imaging and VSP applications in hepatobiliary surgery,assessing surgical precision,complications,and patient outcomes.Thirty studies,including randomized controlled trials,cohort studies,and case reports,were included in the final analysis.RESULTS Various 3D imaging modalities,including multidetector CT,MRI,and 3D rotational angiography,provide high-resolution views of the liver’s vascular and biliary anatomy.VSP allows surgeons to simulate complex surgeries,improving preoperative planning and reducing complications like bleeding and bile leaks.Several studies have demonstrated improved surgical precision,reduced complications,and faster recovery times when 3D imaging and VSP were used in complex surgeries.CONCLUSION 3D imaging and VSP technologies significantly enhance the accuracy and outcomes of hepatobiliary surgeries by providing individualized preoperative planning.While promising,further research,particularly randomized controlled trials,is needed to standardize protocols and evaluate long-term efficacy.
基金the National Natural Science Foundation of China(No.61962032)。
文摘In order to improve the reconstruction accuracy of magnetic resonance imaging(MRI),an accurate natural image compressed sensing(CS)reconstruction network is proposed,which combines the advantages of model-based and deep learning-based CS-MRI methods.In theory,enhancing geometric texture details in linear reconstruction is possible.First,the optimization problem is decomposed into two problems:linear approximation and geometric compensation.Aimed at the problem of image linear approximation,the data consistency module is used to deal with it.Since the processing process will lose texture details,a neural network layer that explicitly combines image and frequency feature representation is proposed,which is named butterfly dilated geometric distillation network.The network introduces the idea of butterfly operation,skillfully integrates the features of image domain and frequency domain,and avoids the loss of texture details when extracting features in a single domain.Finally,a channel feature fusion module is designed by combining channel attention mechanism and dilated convolution.The attention of the channel makes the final output feature map focus on the more important part,thus improving the feature representation ability.The dilated convolution enlarges the receptive field,thereby obtaining more dense image feature data.The experimental results show that the peak signal-to-noise ratio of the network is 5.43 dB,5.24 dB and 3.89 dB higher than that of ISTA-Net+,FISTA and DGDN networks on the brain data set with a Cartesian sampling mask CS ratio of 10%.
基金funded by the National Natural Science Foundation of China(No.62305041)the Natural Science Foundation of Liaoning Province(No.2023-MS-103)。
文摘Stochastic optical reconstruction microscopy(STORM),as a typical technique of single-molecule localization microscopy(SMLM),has overcome the diffraction limit by randomly switching fluorophores between fluorescent and dark states,allowing for the precise localization of isolated emission patterns and the super-resolution reconstruction from millions of localized positions of single fluorophores.A critical factor influencing localization precision is the photo-switching behavior of fluorophores,which is affected by the imaging buffer.The imaging buffer typically comprises oxygen scavengers,photo-switching reagents,and refractive index regulators.Oxygen scavengers help prevent photobleaching,photo-switching reagents assist in facilitating the conversion of fluorophores,and refractive index regulators are used to adjust the refractive index of the solution.The synergistic interaction of these components promotes stable blinking of fluorophores,reduces irreversible photobleaching,and thereby ensures high-quality super-resolution imaging.This review provides a comprehensive overview of the essential compositions and functionalities of imaging buffers used in STORM,serving as a valuable resource for researchers seeking to select appropriate imaging buffers for their experiments.
文摘A new medical image fusion technique is presented.The method is based on three-dimensional reconstruction.After reconstruction,the three-dimensional volume data is normalized by three-dimensional coordinate conversion in the same way and intercepted through setting up cutting plane including anatomical structure,as a result two images in entire registration on space and geometry are obtained and the images are fused at last.Compared with traditional two-dimensional fusion technique,three-dimensional fusion technique can not only resolve the different problems existed in the two kinds of images,but also avoid the registration error of the two kinds of images when they have different scan and imaging parameter.The research proves this fusion technique is more exact and has no registration,so it is more adapt to arbitrary medical image fusion with different equipments.
基金supported by the National Natural Science Foundation of China(Nos.12205271,12075217,U20B2011,and 51978218)Sichuan Science and Technology Program(No.2019ZDZX0010)the National Key R&D Program of China(No.2022YFA1604002).
文摘Neutron capture event imaging is a novel technique that has the potential to substantially enhance the resolution of existing imaging systems.This study provides a measurement method for neutron capture event distribution along with multiple reconstruction methods for super-resolution imaging.The proposed technology reduces the point-spread function of an imag-ing system through single-neutron detection and event reconstruction,thereby significantly improving imaging resolution.A single-neutron detection experiment was conducted using a highly practical and efficient^(6)LiF-ZnS scintillation screen of a cold neutron imaging device in the research reactor.In milliseconds of exposure time,a large number of weak light clusters and their distribution in the scintillation screen were recorded frame by frame,to complete single-neutron detection.Several reconstruction algorithms were proposed for the calculations.The location of neutron capture was calculated using several processing methods such as noise removal,filtering,spot segmentation,contour analysis,and local positioning.The proposed algorithm achieved a higher imaging resolution and faster reconstruction speed,and single-neutron super-resolution imaging was realized by combining single-neutron detection experiments and reconstruction calculations.The results show that the resolution of the 100μm thick^(6)LiF-ZnS scintillation screen can be improved from 125 to 40 microns.This indicates that the proposed single-neutron detection and calculation method is effective and can significantly improve imaging resolution.
基金National Natural Science Foundation of China(Grant Nos.62005049 and 62072110)Natural Science Foundation of Fujian Province(Grant No.2020J01451).
文摘Accurate segmentation of camouflage objects in aerial imagery is vital for improving the efficiency of UAV-based reconnaissance and rescue missions.However,camouflage object segmentation is increasingly challenging due to advances in both camouflage materials and biological mimicry.Although multispectral-RGB based technology shows promise,conventional dual-aperture multispectral-RGB imaging systems are constrained by imprecise and time-consuming registration and fusion across different modalities,limiting their performance.Here,we propose the Reconstructed Multispectral-RGB Fusion Network(RMRF-Net),which reconstructs RGB images into multispectral ones,enabling efficient multimodal segmentation using only an RGB camera.Specifically,RMRF-Net employs a divergentsimilarity feature correction strategy to minimize reconstruction errors and includes an efficient boundary-aware decoder to enhance object contours.Notably,we establish the first real-world aerial multispectral-RGB semantic segmentation of camouflage objects dataset,including 11 object categories.Experimental results demonstrate that RMRF-Net outperforms existing methods,achieving 17.38 FPS on the NVIDIA Jetson AGX Orin,with only a 0.96%drop in mIoU compared to the RTX 3090,showing its practical applicability in multimodal remote sensing.
