The impact of heavy reduction on dendritic morphology was explored by combining experimental research and numerical simulation in metallurgy,including a detailed three-dimensional(3D)analysis and reconstruction of den...The impact of heavy reduction on dendritic morphology was explored by combining experimental research and numerical simulation in metallurgy,including a detailed three-dimensional(3D)analysis and reconstruction of dendritic solidification structures.Combining scanning electron microscopy and energy-dispersive scanning analysis and ANSYS simulation,the high-precision image processing software Mimics Research was utilized to conduct the extraction of dendritic morphologies.Reverse engineering software NX Imageware was employed for the 3D reconstruction of two-dimensional dendritic morphologies,restoring the dendritic characteristics in three-dimensional space.The results demonstrate that in a two-dimensional plane,dendrites connect with each other to form irregularly shaped“ring-like”structures.These dendrites have a thickness greater than 0.1 mm along the Z-axis direction,leading to the envelopment of molten steel by dendrites in a 3D space of at least 0.1 mm.This results in obstructed flow,confirming the“bridging”of dendrites in three-dimensional space,resulting in a tendency for central segregation.Dense and dispersed tiny dendrites,under the influence of heat flow direction,interconnect and continuously grow,gradually forming primary and secondary dendrites in three-dimensional space.After the completion of dendritic solidification and growth,these microdendrites appear dense and dispersed on the two-dimensional plane,providing the nuclei for the formation of new dendrites.When reduction occurs at a solid fraction of 0.46,there is a noticeable decrease in dendritic spacing,resulting in improved central segregation.展开更多
This article proposes a three-dimensional light field reconstruction method based on neural radiation field(NeRF)called Infrared NeRF for low resolution thermal infrared scenes.Based on the characteristics of the low ...This article proposes a three-dimensional light field reconstruction method based on neural radiation field(NeRF)called Infrared NeRF for low resolution thermal infrared scenes.Based on the characteristics of the low resolution thermal infrared imaging,various optimizations have been carried out to improve the speed and accuracy of thermal infrared 3D reconstruction.Firstly,inspired by Boltzmann's law of thermal radiation,distance is incorporated into the NeRF model for the first time,resulting in a nonlinear propagation of a single ray and a more accurate description of the physical property that infrared radiation intensity decreases with increasing distance.Secondly,in terms of improving inference speed,based on the phenomenon of high and low frequency distribution of foreground and background in infrared images,a multi ray non-uniform light synthesis strategy is proposed to make the model pay more attention to foreground objects in the scene,reduce the distribution of light in the background,and significantly reduce training time without reducing accuracy.In addition,compared to visible light scenes,infrared images only have a single channel,so fewer network parameters are required.Experiments using the same training data and data filtering method showed that,compared to the original NeRF,the improved network achieved an average improvement of 13.8%and 4.62%in PSNR and SSIM,respectively,while an average decreases of 46%in LPIPS.And thanks to the optimization of network layers and data filtering methods,training only takes about 25%of the original method's time to achieve convergence.Finally,for scenes with weak backgrounds,this article improves the inference speed of the model by 4-6 times compared to the original NeRF by limiting the query interval of the model.展开更多
3D medical image reconstruction has significantly enhanced diagnostic accuracy,yet the reliance on densely sampled projection data remains a major limitation in clinical practice.Sparse-angle X-ray imaging,though safe...3D medical image reconstruction has significantly enhanced diagnostic accuracy,yet the reliance on densely sampled projection data remains a major limitation in clinical practice.Sparse-angle X-ray imaging,though safer and faster,poses challenges for accurate volumetric reconstruction due to limited spatial information.This study proposes a 3D reconstruction neural network based on adaptive weight fusion(AdapFusionNet)to achieve high-quality 3D medical image reconstruction from sparse-angle X-ray images.To address the issue of spatial inconsistency in multi-angle image reconstruction,an innovative adaptive fusion module was designed to score initial reconstruction results during the inference stage and perform weighted fusion,thereby improving the final reconstruction quality.The reconstruction network is built on an autoencoder(AE)framework and uses orthogonal-angle X-ray images(frontal and lateral projections)as inputs.The encoder extracts 2D features,which the decoder maps into 3D space.This study utilizes a lung CT dataset to obtain complete three-dimensional volumetric data,from which digitally reconstructed radiographs(DRR)are generated at various angles to simulate X-ray images.Since real-world clinical X-ray images rarely come with perfectly corresponding 3D“ground truth,”using CT scans as the three-dimensional reference effectively supports the training and evaluation of deep networks for sparse-angle X-ray 3D reconstruction.Experiments conducted on the LIDC-IDRI dataset with simulated X-ray images(DRR images)as training data demonstrate the superior performance of AdapFusionNet compared to other fusion methods.Quantitative results show that AdapFusionNet achieves SSIM,PSNR,and MAE values of 0.332,13.404,and 0.163,respectively,outperforming other methods(SingleViewNet:0.289,12.363,0.182;AvgFusionNet:0.306,13.384,0.159).Qualitative analysis further confirms that AdapFusionNet significantly enhances the reconstruction of lung and chest contours while effectively reducing noise during the reconstruction process.The findings demonstrate that AdapFusionNet offers significant advantages in 3D reconstruction of sparse-angle X-ray images.展开更多
Neural organoids and confocal microscopy have the potential to play an important role in microconnectome research to understand neural patterns.We present PLayer,a plug-and-play embedded neural system,which demonstrat...Neural organoids and confocal microscopy have the potential to play an important role in microconnectome research to understand neural patterns.We present PLayer,a plug-and-play embedded neural system,which demonstrates the utilization of sparse confocal microscopy layers to interpolate continuous axial resolution.With an embedded system focused on neural network pruning,image scaling,and post-processing,PLayer achieves high-performance metrics with an average structural similarity index of 0.9217 and a peak signal-to-noise ratio of 27.75 dB,all within 20 s.This represents a significant time saving of 85.71%with simplified image processing.By harnessing statistical map estimation in interpolation and incorporating the Vision Transformer–based Restorer,PLayer ensures 2D layer consistency while mitigating heavy computational dependence.As such,PLayer can reconstruct 3D neural organoid confocal data continuously under limited computational power for the wide acceptance of fundamental connectomics and pattern-related research with embedded devices.展开更多
BACKGROUND Gastric cancer(GC)remains a significant global health challenge,with high incidence and mortality rates.Neoadjuvant chemotherapy is increasingly used to improve surgical outcomes and long-term survival in a...BACKGROUND Gastric cancer(GC)remains a significant global health challenge,with high incidence and mortality rates.Neoadjuvant chemotherapy is increasingly used to improve surgical outcomes and long-term survival in advanced cases.However,individual responses to treatment vary widely,and current imaging methods often fall short in accurately predicting efficacy.Advanced imaging techniques,such as computed tomography(CT)3D reconstruction and texture analysis,offer potential for more precise assessment of therapeutic response.AIM To explore the application value of CT 3D reconstruction volume change rate,texture feature analysis,and visual features in assessing the efficacy of neoadjuvant chemotherapy for advanced GC.METHODS A retrospective analysis was conducted on the clinical and imaging data of 97 patients with advanced GC who received S-1 plus Oxaliplatin combined chemotherapy regimen neoadjuvant chemotherapy from January 2022 to March 2024.CT texture feature analysis was performed using MaZda software,and ITK-snap software was used to measure the tumor volume change rate before and after chemotherapy.CT visual features were also evaluated.Using postoperative pathological tumor regression grade(TRG)as the gold standard,the correlation between various indicators and chemotherapy efficacy was analyzed,and a predictive model was constructed and internally validated.RESULTS The minimum misclassification rate of texture features in venous phase CT images(7.85%)was lower than in the arterial phase(13.92%).The volume change rate in the effective chemotherapy group(75.20%)was significantly higher than in the ineffective group(41.75%).There was a strong correlation between volume change rate and TRG grade(r=-0.886,P<0.001).Multivariate analysis showed that gastric wall peristalsis(OR=0.286)and thickness change rate≥40%(OR=0.265)were independent predictive factors.Receiver operating characteristic curve analysis indicated that the volume change rate[area under the curve(AUC)=0.885]was superior to the CT visual feature model(AUC=0.795).When the cutoff value was 82.56%,the sensitivity and specificity were 85.62%and 96.45%,respectively.CONCLUSION The CT 3D reconstruction volume change rate can serve as a preferred quantitative indicator for evaluating the efficacy of neoadjuvant chemotherapy in GC.Combining it with a CT visual feature predictive model can further improve the accuracy of efficacy evaluation.展开更多
Large-scale indoor 3D reconstruction with multiple robots faces challenges in core enabling technologies.This work contributes to a framework addressing localization,coordination,and vision processing for multi-agent ...Large-scale indoor 3D reconstruction with multiple robots faces challenges in core enabling technologies.This work contributes to a framework addressing localization,coordination,and vision processing for multi-agent reconstruction.A system architecture fusing visible light positioning,multi-agent path finding via reinforcement learning,and 360°camera techniques for 3D reconstruction is proposed.Our visible light positioning algorithm leverages existing lighting for centimeter-level localization without additional infrastructure.Meanwhile,a decentralized reinforcement learning approach is developed to solve the multi-agent path finding problem,with communications among agents optimized.Our 3D reconstruction pipeline utilizes equirectangular projection from 360°cameras to facilitate depth-independent reconstruction from posed monocular images using neural networks.Experimental validation demonstrates centimeter-level indoor navigation and 3D scene reconstruction capabilities of our framework.The challenges and limitations stemming from the above enabling technologies are discussed at the end of each corresponding section.In summary,this research advances fundamental techniques for multi-robot indoor 3D modeling,contributing to automated,data-driven applications through coordinated robot navigation,perception,and modeling.展开更多
3D reconstruction plays an increasingly important role in modern photogrammetric systems.Conventional satellite or aerial-based remote sensing(RS)platforms can provide the necessary data sources for the 3D reconstruct...3D reconstruction plays an increasingly important role in modern photogrammetric systems.Conventional satellite or aerial-based remote sensing(RS)platforms can provide the necessary data sources for the 3D reconstruction of large-scale landforms and cities.Even with low-altitude Unmanned Aerial Vehicles(UAVs),3D reconstruction in complicated situations,such as urban canyons and indoor scenes,is challenging due to frequent tracking failures between camera frames and high data collection costs.Recently,spherical images have been extensively used due to the capability of recording surrounding environments from one image.In contrast to perspective images with limited Field of View(FOV),spherical images can cover the whole scene with full horizontal and vertical FOV and facilitate camera tracking and data acquisition in these complex scenes.With the rapid evolution and extensive use of professional and consumer-grade spherical cameras,spherical images show great potential for the 3D modeling of urban and indoor scenes.Classical 3D reconstruction pipelines,however,cannot be directly used for spherical images.Besides,there exist few software packages that are designed for the 3D reconstruction from spherical images.As a result,this research provides a thorough survey of the state-of-the-art for 3D reconstruction from spherical images in terms of data acquisition,feature detection and matching,image orientation,and dense matching as well as presenting promising applications and discussing potential prospects.We anticipate that this study offers insightful clues to direct future research.展开更多
Three-dimensional reconstruction technology plays an important role in indoor scenes by converting objects and structures in indoor environments into accurate 3D models using multi-view RGB images.It offers a wide ran...Three-dimensional reconstruction technology plays an important role in indoor scenes by converting objects and structures in indoor environments into accurate 3D models using multi-view RGB images.It offers a wide range of applications in fields such as virtual reality,augmented reality,indoor navigation,and game development.Existing methods based on multi-view RGB images have made significant progress in 3D reconstruction.These image-based reconstruction methods not only possess good expressive power and generalization performance,but also handle complex geometric shapes and textures effectively.Despite facing challenges such as lighting variations,occlusion,and texture loss in indoor scenes,these challenges can be effectively addressed through deep neural networks,neural implicit surface representations,and other techniques.The technology of indoor 3D reconstruction based on multi-view RGB images has a promising future.It not only provides immersive and interactive virtual experiences but also brings convenience and innovation to indoor navigation,interior design,and virtual tours.As the technology evolves,these image-based reconstruction methods will be further improved to provide higher quality and more accurate solutions to indoor scene reconstruction.展开更多
Three-dimensional(3D) synthetic aperture radar(SAR)extends the conventional 2D images into 3D features by several acquisitions in different aspects. Compared with 3D techniques via multiple observations in elevation, ...Three-dimensional(3D) synthetic aperture radar(SAR)extends the conventional 2D images into 3D features by several acquisitions in different aspects. Compared with 3D techniques via multiple observations in elevation, e.g. SAR interferometry(InSAR) and SAR tomography(TomoSAR), holographic SAR can retrieve 3D structure by observations in azimuth. This paper focuses on designing a novel type of orbit to achieve SAR regional all-azimuth observation(AAO) for embedded targets detection and holographic 3D reconstruction. The ground tracks of the AAO orbit separate the earth surface into grids. Target in these grids can be accessed with an azimuth angle span of360°, which is similar to the flight path of airborne circular SAR(CSAR). Inspired from the successive coverage orbits of optical sensors, several optimizations are made in the proposed method to ensure favorable grazing angles, the performance of 3D reconstruction, and long-term supervision for SAR sensors. Simulation experiments show the regional AAO can be completed within five hours. In addition, a second AAO of the same area can be duplicated in two days. Finally, an airborne SAR data process result is presented to illustrate the significance of AAO in 3D reconstruction.展开更多
The development of digital intelligent diagnostic and treatment technology has opened countless new opportunities for liver surgery from the era of digital anatomy to a new era of digital diagnostics,virtual surgery s...The development of digital intelligent diagnostic and treatment technology has opened countless new opportunities for liver surgery from the era of digital anatomy to a new era of digital diagnostics,virtual surgery simulation and using the created scenarios in real-time surgery using mixed reality.In this article,we described our experience on developing a dedicated 3 dimensional visualization and reconstruction software for surgeons to be used in advanced liver surgery and living donor liver transplantation.Furthermore,we shared the recent developments in the field by explaining the outreach of the software from virtual reality to augmented reality and mixed reality.展开更多
The biomechanical relationship between the articular cartilage defect and knee osteoarthritis (OA) has not been clearly defined. This study presents a 3D knee finite element model (FEM) to determine the effect of cart...The biomechanical relationship between the articular cartilage defect and knee osteoarthritis (OA) has not been clearly defined. This study presents a 3D knee finite element model (FEM) to determine the effect of cartilage defects on the stress distribution around the defect rim. The complete knee FEM, which includes bones, articular cartilages, menisci and ligaments, is developed from computed tomography and magnetic resonance images. This FEM then is validated and used to simulate femoral cartilage defects. Based on the obtained results, it is confirmed that the 3D knee FEM is reconstructed with high-fidelity level and can faithfully predict the knee contact behavior. Cartilage defects drastically affect the stress distribution on articular cartilages. When the defect size was smaller than 1.00cm2, the stress elevation and redistribution were found undistinguishable. However, significant stress elevation and redistribution were detected due to the large defect sizes ( 1.00cm2). This alteration of stress distribution has important implications relating to the progression of cartilage defect to OA in the human knee joint.展开更多
The 3D reconstruction using deep learning-based intelligent systems can provide great help for measuring an individual’s height and shape quickly and accurately through 2D motion-blurred images.Generally,during the a...The 3D reconstruction using deep learning-based intelligent systems can provide great help for measuring an individual’s height and shape quickly and accurately through 2D motion-blurred images.Generally,during the acquisition of images in real-time,motion blur,caused by camera shaking or human motion,appears.Deep learning-based intelligent control applied in vision can help us solve the problem.To this end,we propose a 3D reconstruction method for motion-blurred images using deep learning.First,we develop a BF-WGAN algorithm that combines the bilateral filtering(BF)denoising theory with a Wasserstein generative adversarial network(WGAN)to remove motion blur.The bilateral filter denoising algorithm is used to remove the noise and to retain the details of the blurred image.Then,the blurred image and the corresponding sharp image are input into the WGAN.This algorithm distinguishes the motion-blurred image from the corresponding sharp image according to the WGAN loss and perceptual loss functions.Next,we use the deblurred images generated by the BFWGAN algorithm for 3D reconstruction.We propose a threshold optimization random sample consensus(TO-RANSAC)algorithm that can remove the wrong relationship between two views in the 3D reconstructed model relatively accurately.Compared with the traditional RANSAC algorithm,the TO-RANSAC algorithm can adjust the threshold adaptively,which improves the accuracy of the 3D reconstruction results.The experimental results show that our BF-WGAN algorithm has a better deblurring effect and higher efficiency than do other representative algorithms.In addition,the TO-RANSAC algorithm yields a calculation accuracy considerably higher than that of the traditional RANSAC algorithm.展开更多
3D reconstruction of worn parts is the foundation for remanufacturing system based on robotic arc welding, because it can provide 3D geometric information for robot task plan. In this investigation, a novel 3D reconst...3D reconstruction of worn parts is the foundation for remanufacturing system based on robotic arc welding, because it can provide 3D geometric information for robot task plan. In this investigation, a novel 3D reconstruction system based on linear structured light vision sensing is developed. This system hardware consists of a MTC368-CB CCD camera, a MLH-645 laser projector and a DH-CG300 image grabbing card. This system software is developed to control the image data capture. In order to reconstruct the 3D geometric information from the captured image, a two steps rapid calibration algorithm is proposed. The 3D reconstruction experiment shows a satisfactory result.展开更多
BACKGROUND Hernia is a common condition requiring abdominal surgery.The current standard treatment for hernia is tension-free repair using meshes.Globally,more than 200 new types of meshes are licensed each year.Howev...BACKGROUND Hernia is a common condition requiring abdominal surgery.The current standard treatment for hernia is tension-free repair using meshes.Globally,more than 200 new types of meshes are licensed each year.However,their clinical applications are associated with a series of complications,such as recurrence(10%-24%)and infection(0.5%-9.0%).In contrast,3D-printed meshes have significantly reduced the postoperative complications in patients.They have also shortened operating time and minimized the loss of mesh materials.In this study,we used the myopectineal orifice(MPO)data obtained from preoperative computer tomography(CT)-based 3D reconstruction for the production of 3D-printed biologic meshes.AIM To investigate the application of multislice spiral CT-based 3D reconstruction technique in 3D-printed biologic mesh for hernia repair surgery.METHODS We retrospectively analyzed 60 patients who underwent laparoscopic tension-free repair for inguinal hernia in the Department of General Surgery of the First Hospital of Shanxi Medical University from September 2019 to December 2019.This study included 30 males and 30 females,with a mean age of 40±5.6 years.Data on the MPO were obtained from preoperative CT-based 3D reconstruction as well as from real-world intraoperative measurements for all patients.Anatomic points were set for the purpose of measurement based on the definition of MPO:A:The pubic tubercle;B:Intersection of the horizontal line extending from the summit of the inferior edge of the internal oblique and transversus abdominis and the outer edge of the rectus abdominis,C:Intersection of the horizontal line extending from the summit of the inferior edge of the internal oblique and transversus abdominis and the inguinal ligament,D:Intersection of the iliopsoas muscle and the inguinal ligament,and E:Intersection of the iliopsoas muscle and the superior pubic ramus.The distance between the points was measured.All preoperative and intraoperative data were analyzed using the t test.Differences with P<0.05 were considered significant in comparative analysis.RESULTS The distance between points AB,AC,BC,DE,and AE based on preoperative and intraoperative data was 7.576±0.212 cm vs 7.573±0.266 cm,7.627±0.212 cm vs 7.627±0.212 cm,7.677±0.229 cm vs 7.567±0.786 cm,7.589±0.204 cm vs 7.512±0.21 cm,and 7.617±0.231 cm vs 7.582±0.189 cm,respectively.All differences were not statistically significant(P>0.05).CONCLUSION The use of multislice spiral CT-based 3D reconstruction technique before hernia repair surgery allows accurate measurement of data and relationships of different anatomic sites in the MPO region.This technique can provide precise data for the production of 3D-printed biologic meshes.展开更多
With increasingly more smart cameras deployed in infrastructure and commercial buildings,3D reconstruction can quickly obtain cities’information and improve the efficiency of government services.Images collected in o...With increasingly more smart cameras deployed in infrastructure and commercial buildings,3D reconstruction can quickly obtain cities’information and improve the efficiency of government services.Images collected in outdoor hazy environments are prone to color distortion and low contrast;thus,the desired visual effect cannot be achieved and the difficulty of target detection is increased.Artificial intelligence(AI)solutions provide great help for dehazy images,which can automatically identify patterns or monitor the environment.Therefore,we propose a 3D reconstruction method of dehazed images for smart cities based on deep learning.First,we propose a fine transmission image deep convolutional regression network(FT-DCRN)dehazing algorithm that uses fine transmission image and atmospheric light value to compute dehazed image.The DCRN is used to obtain the coarse transmission image,which can not only expand the receptive field of the network but also retain the features to maintain the nonlinearity of the overall network.The fine transmission image is obtained by refining the coarse transmission image using a guided filter.The atmospheric light value is estimated according to the position and brightness of the pixels in the original hazy image.Second,we use the dehazed images generated by the FT-DCRN dehazing algorithm for 3D reconstruction.An advanced relaxed iterative fine matching based on the structure from motion(ARI-SFM)algorithm is proposed.The ARISFM algorithm,which obtains the fine matching corner pairs and reduces the number of iterations,establishes an accurate one-to-one matching corner relationship.The experimental results show that our FT-DCRN dehazing algorithm improves the accuracy compared to other representative algorithms.In addition,the ARI-SFM algorithm guarantees the precision and improves the efficiency.展开更多
Fabric pilling evaluation has been considered as an essential element for textile quality inspection. Traditional manual method is still based on human eyes and brain, which is subjective with low efficiency. This pap...Fabric pilling evaluation has been considered as an essential element for textile quality inspection. Traditional manual method is still based on human eyes and brain, which is subjective with low efficiency. This paper proposes an objective evaluation method based on semi-calibrated near-light Photometric Stereo(PS). Fabric images are digitalized by self-developed image acquisition system. The 3D depth information of each point could be obtained by PS algorithm and then mapped to 2D grayscale image. After that, the non-textured image could be filtered by using the Gaussian low-pass filter. The pilling segmentation is conducted by using global iterative threshold segmentation method,and then K-Nearest Neighbor(KNN) is finally selected as a tool for the grade classification of fabric pilling. Our experimental results show that the proposed evaluation system could achieve excellent judging performance for the objective pilling evaluation.展开更多
The three-dimensional(3D)reconstruction technology based on structured light has been widely used in the field of industrial measurement due to its many advantages.Aiming at the problems of high mismatch rate and poor...The three-dimensional(3D)reconstruction technology based on structured light has been widely used in the field of industrial measurement due to its many advantages.Aiming at the problems of high mismatch rate and poor real-time performance caused by factors such as system jitter and noise,a lightweight stripe image feature extraction algorithm based on You Only Look Once v4(YOLOv4)network is proposed.First,Mobilenetv3 is used as the backbone network to effectively extract features,and then the Mish activation function and Complete Intersection over Union(CIoU)loss function are used to calculate the improved target frame regression loss,which effectively improves the accuracy and real-time performance of feature detection.Simulation experiment results show that the model size after the improved algorithm is only 52 MB,the mean average accuracy(mAP)of fringe image data reconstruction reaches 82.11%,and the 3D point cloud restoration rate reaches 90.1%.Compared with the existing model,it has obvious advantages and can satisfy the accuracy and real-time requirements of reconstruction tasks in resource-constrained equipment.展开更多
BACKGROUND For treatment of hilar cholangiocarcinoma(HCCA),the rate of radical resection is low and prognosis is poor,and preoperative evaluation is not sufficiently accurate.3D visualization has the advantage of givi...BACKGROUND For treatment of hilar cholangiocarcinoma(HCCA),the rate of radical resection is low and prognosis is poor,and preoperative evaluation is not sufficiently accurate.3D visualization has the advantage of giving a stereoscopic view,which makes accurate resection of HCCA possible.AIM To establish precise resection of HCCA based on eOrganmap 3D reconstruction and full quantification technology.METHODS We retrospectively analyzed the clinical data of 73 patients who underwent HCCA surgery.All patients were assigned to two groups.The traditional group received traditional 2D imaging planning before surgery(n=35).The eOrganmap group underwent 3D reconstruction and full quantitative technical planning before surgery(n=38).The preoperative evaluation,anatomical classification of hilar hepatic vessels,indicators associated with surgery,postoperative complications,liver function,and stress response indexes were compared between the groups.RESULTS Compared with the traditional group,the amount of intraoperative blood loss in the eOrganmap group was lower,the operating time and postoperative intestinal ventilation time were shorter,and R0 resection rate and lymph node dissection number were higher(P<0.05).The total complication rate in the eOrganmap group was 21.05%compared with 25.71%in the traditional group(P>0.05).The levels of total bilirubin,Albumin(ALB),aspartate transaminase,and alanine transaminase in the eOrganmap group were significantly different from those in the traditional group(intergroup effect:F=450.400,79.120,95.730,and 13.240,respectively;all P<0.001).Total bilirubin,aspartate transaminase,and alanine transaminase in both groups showed a decreasing trend with time(time effect:F=30.270,17.340,and 13.380,respectively;all P<0.001).There was an interaction between patient group and time(interaction effect:F=3.072,2.965,and 2.703,respectively;P=0.0282,0.032,and 0.046,respectively);ALB levels in both groups tended to increase with time(time effect:F=22.490,P<0.001),and there was an interaction effect between groups and time(interaction effect:F=4.607,P=0.004).In the eOrganmap group,there was a high correlation between the actual volume of intraoperative liver specimen resection and the volume of preoperative virtual liver resection(t=0.916,P<0.001).CONCLUSION The establishment of accurate laparoscopic resection of hilar cholangiocarcinoma based on preoperative eOrganmap 3D reconstruction and full quantization technology can make laparoscopic resection of hilar cholangiocarcinoma more accurate and safe.展开更多
A fast 3D reconstruction method based on structured light to measure various parameters of the raceway groove is presented. Digital parallel grating stripes distributed with sine density are projected onto the raceway...A fast 3D reconstruction method based on structured light to measure various parameters of the raceway groove is presented. Digital parallel grating stripes distributed with sine density are projected onto the raceway groove by a DLP projector, and distorting of stripes is happened on the raceway. Simultaneously, aided by three-step phase-shifting approach, three images covered by different stripes are obtained by a high-resolution CCD camera at the same location, thus a more accuracy local topography can be obtained. And then the bearing is rotated on a high precision computer controlled rotational stage. Three images are also obtained as the former step at next planned location triggered by the motor. After one cycle, all images information is combined through the mosaics. As a result, the 3D information of raceway groove can be gained. Not only geometric properties but also surface flaws can be extracted by software. A preliminary hardware system has been built, with which some geometric parameters have been extracted from reconstructed local topography.展开更多
While Kinect was originally designed as a motion sensing input device of the gaming console Microsoft Xbox 360 for gaming purposes, it＇s easy-to-use, low-cost, reliability, speed of the depth measurement and relative...While Kinect was originally designed as a motion sensing input device of the gaming console Microsoft Xbox 360 for gaming purposes, it＇s easy-to-use, low-cost, reliability, speed of the depth measurement and relatively high quality of depth measurement make it can be used for 3D reconstruction. It could make 3D scanning technology more accessible to everyday users and turn 3D reconstruction models into much widely used asset for many applications. In this paper, we focus on Kinect 3D reconstruction.展开更多
基金supported by Open Foundation of the State Key Laboratory of Refractories and Metallurgy(No.G201711)the National Natural Science Foundation of China(Nos.52104317 and 51874001).
文摘The impact of heavy reduction on dendritic morphology was explored by combining experimental research and numerical simulation in metallurgy,including a detailed three-dimensional(3D)analysis and reconstruction of dendritic solidification structures.Combining scanning electron microscopy and energy-dispersive scanning analysis and ANSYS simulation,the high-precision image processing software Mimics Research was utilized to conduct the extraction of dendritic morphologies.Reverse engineering software NX Imageware was employed for the 3D reconstruction of two-dimensional dendritic morphologies,restoring the dendritic characteristics in three-dimensional space.The results demonstrate that in a two-dimensional plane,dendrites connect with each other to form irregularly shaped“ring-like”structures.These dendrites have a thickness greater than 0.1 mm along the Z-axis direction,leading to the envelopment of molten steel by dendrites in a 3D space of at least 0.1 mm.This results in obstructed flow,confirming the“bridging”of dendrites in three-dimensional space,resulting in a tendency for central segregation.Dense and dispersed tiny dendrites,under the influence of heat flow direction,interconnect and continuously grow,gradually forming primary and secondary dendrites in three-dimensional space.After the completion of dendritic solidification and growth,these microdendrites appear dense and dispersed on the two-dimensional plane,providing the nuclei for the formation of new dendrites.When reduction occurs at a solid fraction of 0.46,there is a noticeable decrease in dendritic spacing,resulting in improved central segregation.
基金Support by the Fundamental Research Funds for the Central Universities(2024300443)the National Natural Science Foundation of China(NSFC)Young Scientists Fund(62405131)。
文摘This article proposes a three-dimensional light field reconstruction method based on neural radiation field(NeRF)called Infrared NeRF for low resolution thermal infrared scenes.Based on the characteristics of the low resolution thermal infrared imaging,various optimizations have been carried out to improve the speed and accuracy of thermal infrared 3D reconstruction.Firstly,inspired by Boltzmann's law of thermal radiation,distance is incorporated into the NeRF model for the first time,resulting in a nonlinear propagation of a single ray and a more accurate description of the physical property that infrared radiation intensity decreases with increasing distance.Secondly,in terms of improving inference speed,based on the phenomenon of high and low frequency distribution of foreground and background in infrared images,a multi ray non-uniform light synthesis strategy is proposed to make the model pay more attention to foreground objects in the scene,reduce the distribution of light in the background,and significantly reduce training time without reducing accuracy.In addition,compared to visible light scenes,infrared images only have a single channel,so fewer network parameters are required.Experiments using the same training data and data filtering method showed that,compared to the original NeRF,the improved network achieved an average improvement of 13.8%and 4.62%in PSNR and SSIM,respectively,while an average decreases of 46%in LPIPS.And thanks to the optimization of network layers and data filtering methods,training only takes about 25%of the original method's time to achieve convergence.Finally,for scenes with weak backgrounds,this article improves the inference speed of the model by 4-6 times compared to the original NeRF by limiting the query interval of the model.
基金Supported by Sichuan Science and Technology Program(2023YFSY0026,2023YFH0004).
文摘3D medical image reconstruction has significantly enhanced diagnostic accuracy,yet the reliance on densely sampled projection data remains a major limitation in clinical practice.Sparse-angle X-ray imaging,though safer and faster,poses challenges for accurate volumetric reconstruction due to limited spatial information.This study proposes a 3D reconstruction neural network based on adaptive weight fusion(AdapFusionNet)to achieve high-quality 3D medical image reconstruction from sparse-angle X-ray images.To address the issue of spatial inconsistency in multi-angle image reconstruction,an innovative adaptive fusion module was designed to score initial reconstruction results during the inference stage and perform weighted fusion,thereby improving the final reconstruction quality.The reconstruction network is built on an autoencoder(AE)framework and uses orthogonal-angle X-ray images(frontal and lateral projections)as inputs.The encoder extracts 2D features,which the decoder maps into 3D space.This study utilizes a lung CT dataset to obtain complete three-dimensional volumetric data,from which digitally reconstructed radiographs(DRR)are generated at various angles to simulate X-ray images.Since real-world clinical X-ray images rarely come with perfectly corresponding 3D“ground truth,”using CT scans as the three-dimensional reference effectively supports the training and evaluation of deep networks for sparse-angle X-ray 3D reconstruction.Experiments conducted on the LIDC-IDRI dataset with simulated X-ray images(DRR images)as training data demonstrate the superior performance of AdapFusionNet compared to other fusion methods.Quantitative results show that AdapFusionNet achieves SSIM,PSNR,and MAE values of 0.332,13.404,and 0.163,respectively,outperforming other methods(SingleViewNet:0.289,12.363,0.182;AvgFusionNet:0.306,13.384,0.159).Qualitative analysis further confirms that AdapFusionNet significantly enhances the reconstruction of lung and chest contours while effectively reducing noise during the reconstruction process.The findings demonstrate that AdapFusionNet offers significant advantages in 3D reconstruction of sparse-angle X-ray images.
基金supported by the National Key R&D Program of China(Grant No.2021YFA1001000)the National Natural Science Foundation of China(Grant Nos.82111530212,U23A20282,and 61971255)+2 种基金the Natural Science Founda-tion of Guangdong Province(Grant No.2021B1515020092)the Shenzhen Bay Laboratory Fund(Grant No.SZBL2020090501014)the Shenzhen Science,Technology and Innovation Commission(Grant Nos.KJZD20231023094659002,JCYJ20220530142809022,and WDZC20220811170401001).
文摘Neural organoids and confocal microscopy have the potential to play an important role in microconnectome research to understand neural patterns.We present PLayer,a plug-and-play embedded neural system,which demonstrates the utilization of sparse confocal microscopy layers to interpolate continuous axial resolution.With an embedded system focused on neural network pruning,image scaling,and post-processing,PLayer achieves high-performance metrics with an average structural similarity index of 0.9217 and a peak signal-to-noise ratio of 27.75 dB,all within 20 s.This represents a significant time saving of 85.71%with simplified image processing.By harnessing statistical map estimation in interpolation and incorporating the Vision Transformer–based Restorer,PLayer ensures 2D layer consistency while mitigating heavy computational dependence.As such,PLayer can reconstruct 3D neural organoid confocal data continuously under limited computational power for the wide acceptance of fundamental connectomics and pattern-related research with embedded devices.
文摘BACKGROUND Gastric cancer(GC)remains a significant global health challenge,with high incidence and mortality rates.Neoadjuvant chemotherapy is increasingly used to improve surgical outcomes and long-term survival in advanced cases.However,individual responses to treatment vary widely,and current imaging methods often fall short in accurately predicting efficacy.Advanced imaging techniques,such as computed tomography(CT)3D reconstruction and texture analysis,offer potential for more precise assessment of therapeutic response.AIM To explore the application value of CT 3D reconstruction volume change rate,texture feature analysis,and visual features in assessing the efficacy of neoadjuvant chemotherapy for advanced GC.METHODS A retrospective analysis was conducted on the clinical and imaging data of 97 patients with advanced GC who received S-1 plus Oxaliplatin combined chemotherapy regimen neoadjuvant chemotherapy from January 2022 to March 2024.CT texture feature analysis was performed using MaZda software,and ITK-snap software was used to measure the tumor volume change rate before and after chemotherapy.CT visual features were also evaluated.Using postoperative pathological tumor regression grade(TRG)as the gold standard,the correlation between various indicators and chemotherapy efficacy was analyzed,and a predictive model was constructed and internally validated.RESULTS The minimum misclassification rate of texture features in venous phase CT images(7.85%)was lower than in the arterial phase(13.92%).The volume change rate in the effective chemotherapy group(75.20%)was significantly higher than in the ineffective group(41.75%).There was a strong correlation between volume change rate and TRG grade(r=-0.886,P<0.001).Multivariate analysis showed that gastric wall peristalsis(OR=0.286)and thickness change rate≥40%(OR=0.265)were independent predictive factors.Receiver operating characteristic curve analysis indicated that the volume change rate[area under the curve(AUC)=0.885]was superior to the CT visual feature model(AUC=0.795).When the cutoff value was 82.56%,the sensitivity and specificity were 85.62%and 96.45%,respectively.CONCLUSION The CT 3D reconstruction volume change rate can serve as a preferred quantitative indicator for evaluating the efficacy of neoadjuvant chemotherapy in GC.Combining it with a CT visual feature predictive model can further improve the accuracy of efficacy evaluation.
基金supported by Bright Dream Robotics and the HKUSTBDR Joint Research Institute Funding Scheme under Project HBJRI-FTP-005(Automated 3D Reconstruction using Robot-mounted 360-Degree Camera with Visible Light Positioning Technology for Building Information Modelling Applications,OKT22EG06).
文摘Large-scale indoor 3D reconstruction with multiple robots faces challenges in core enabling technologies.This work contributes to a framework addressing localization,coordination,and vision processing for multi-agent reconstruction.A system architecture fusing visible light positioning,multi-agent path finding via reinforcement learning,and 360°camera techniques for 3D reconstruction is proposed.Our visible light positioning algorithm leverages existing lighting for centimeter-level localization without additional infrastructure.Meanwhile,a decentralized reinforcement learning approach is developed to solve the multi-agent path finding problem,with communications among agents optimized.Our 3D reconstruction pipeline utilizes equirectangular projection from 360°cameras to facilitate depth-independent reconstruction from posed monocular images using neural networks.Experimental validation demonstrates centimeter-level indoor navigation and 3D scene reconstruction capabilities of our framework.The challenges and limitations stemming from the above enabling technologies are discussed at the end of each corresponding section.In summary,this research advances fundamental techniques for multi-robot indoor 3D modeling,contributing to automated,data-driven applications through coordinated robot navigation,perception,and modeling.
基金funded by the National Natural Science Foundation of China[Grant No.42371442]the Hubei Provincial Natural Science Foundation of China[Grant No.2023AFB568]+1 种基金the Hong Kong Scholars Program[Grant No.2021-114]the Open Research fund from the Hubei Luojia Laboratory[Grand No.230100013].
文摘3D reconstruction plays an increasingly important role in modern photogrammetric systems.Conventional satellite or aerial-based remote sensing(RS)platforms can provide the necessary data sources for the 3D reconstruction of large-scale landforms and cities.Even with low-altitude Unmanned Aerial Vehicles(UAVs),3D reconstruction in complicated situations,such as urban canyons and indoor scenes,is challenging due to frequent tracking failures between camera frames and high data collection costs.Recently,spherical images have been extensively used due to the capability of recording surrounding environments from one image.In contrast to perspective images with limited Field of View(FOV),spherical images can cover the whole scene with full horizontal and vertical FOV and facilitate camera tracking and data acquisition in these complex scenes.With the rapid evolution and extensive use of professional and consumer-grade spherical cameras,spherical images show great potential for the 3D modeling of urban and indoor scenes.Classical 3D reconstruction pipelines,however,cannot be directly used for spherical images.Besides,there exist few software packages that are designed for the 3D reconstruction from spherical images.As a result,this research provides a thorough survey of the state-of-the-art for 3D reconstruction from spherical images in terms of data acquisition,feature detection and matching,image orientation,and dense matching as well as presenting promising applications and discussing potential prospects.We anticipate that this study offers insightful clues to direct future research.
基金supported by ZTE Industry University Institute Cooperation Funds under Grant No.HCCN20221102002.
文摘Three-dimensional reconstruction technology plays an important role in indoor scenes by converting objects and structures in indoor environments into accurate 3D models using multi-view RGB images.It offers a wide range of applications in fields such as virtual reality,augmented reality,indoor navigation,and game development.Existing methods based on multi-view RGB images have made significant progress in 3D reconstruction.These image-based reconstruction methods not only possess good expressive power and generalization performance,but also handle complex geometric shapes and textures effectively.Despite facing challenges such as lighting variations,occlusion,and texture loss in indoor scenes,these challenges can be effectively addressed through deep neural networks,neural implicit surface representations,and other techniques.The technology of indoor 3D reconstruction based on multi-view RGB images has a promising future.It not only provides immersive and interactive virtual experiences but also brings convenience and innovation to indoor navigation,interior design,and virtual tours.As the technology evolves,these image-based reconstruction methods will be further improved to provide higher quality and more accurate solutions to indoor scene reconstruction.
基金supported by the National Natural Science Foundation of China (62001436)the Natural Science Foundation of Jiangsu Province under (BK 20190143,JSGG20190823094603691)。
文摘Three-dimensional(3D) synthetic aperture radar(SAR)extends the conventional 2D images into 3D features by several acquisitions in different aspects. Compared with 3D techniques via multiple observations in elevation, e.g. SAR interferometry(InSAR) and SAR tomography(TomoSAR), holographic SAR can retrieve 3D structure by observations in azimuth. This paper focuses on designing a novel type of orbit to achieve SAR regional all-azimuth observation(AAO) for embedded targets detection and holographic 3D reconstruction. The ground tracks of the AAO orbit separate the earth surface into grids. Target in these grids can be accessed with an azimuth angle span of360°, which is similar to the flight path of airborne circular SAR(CSAR). Inspired from the successive coverage orbits of optical sensors, several optimizations are made in the proposed method to ensure favorable grazing angles, the performance of 3D reconstruction, and long-term supervision for SAR sensors. Simulation experiments show the regional AAO can be completed within five hours. In addition, a second AAO of the same area can be duplicated in two days. Finally, an airborne SAR data process result is presented to illustrate the significance of AAO in 3D reconstruction.
文摘The development of digital intelligent diagnostic and treatment technology has opened countless new opportunities for liver surgery from the era of digital anatomy to a new era of digital diagnostics,virtual surgery simulation and using the created scenarios in real-time surgery using mixed reality.In this article,we described our experience on developing a dedicated 3 dimensional visualization and reconstruction software for surgeons to be used in advanced liver surgery and living donor liver transplantation.Furthermore,we shared the recent developments in the field by explaining the outreach of the software from virtual reality to augmented reality and mixed reality.
基金the National Natural Science Foundation of China (No. 81071235)the Medicine and Engineering Interdisciplinary Fund of Shanghai Jiaotong University (No. YG2010MS26)
文摘The biomechanical relationship between the articular cartilage defect and knee osteoarthritis (OA) has not been clearly defined. This study presents a 3D knee finite element model (FEM) to determine the effect of cartilage defects on the stress distribution around the defect rim. The complete knee FEM, which includes bones, articular cartilages, menisci and ligaments, is developed from computed tomography and magnetic resonance images. This FEM then is validated and used to simulate femoral cartilage defects. Based on the obtained results, it is confirmed that the 3D knee FEM is reconstructed with high-fidelity level and can faithfully predict the knee contact behavior. Cartilage defects drastically affect the stress distribution on articular cartilages. When the defect size was smaller than 1.00cm2, the stress elevation and redistribution were found undistinguishable. However, significant stress elevation and redistribution were detected due to the large defect sizes ( 1.00cm2). This alteration of stress distribution has important implications relating to the progression of cartilage defect to OA in the human knee joint.
基金the National Natural Science Foundation of China under Grant 61902311in part by the Japan Society for the Promotion of Science(JSPS)Grants-in-Aid for Scientific Research(KAKENHI)under Grant JP18K18044.
文摘The 3D reconstruction using deep learning-based intelligent systems can provide great help for measuring an individual’s height and shape quickly and accurately through 2D motion-blurred images.Generally,during the acquisition of images in real-time,motion blur,caused by camera shaking or human motion,appears.Deep learning-based intelligent control applied in vision can help us solve the problem.To this end,we propose a 3D reconstruction method for motion-blurred images using deep learning.First,we develop a BF-WGAN algorithm that combines the bilateral filtering(BF)denoising theory with a Wasserstein generative adversarial network(WGAN)to remove motion blur.The bilateral filter denoising algorithm is used to remove the noise and to retain the details of the blurred image.Then,the blurred image and the corresponding sharp image are input into the WGAN.This algorithm distinguishes the motion-blurred image from the corresponding sharp image according to the WGAN loss and perceptual loss functions.Next,we use the deblurred images generated by the BFWGAN algorithm for 3D reconstruction.We propose a threshold optimization random sample consensus(TO-RANSAC)algorithm that can remove the wrong relationship between two views in the 3D reconstructed model relatively accurately.Compared with the traditional RANSAC algorithm,the TO-RANSAC algorithm can adjust the threshold adaptively,which improves the accuracy of the 3D reconstruction results.The experimental results show that our BF-WGAN algorithm has a better deblurring effect and higher efficiency than do other representative algorithms.In addition,the TO-RANSAC algorithm yields a calculation accuracy considerably higher than that of the traditional RANSAC algorithm.
文摘3D reconstruction of worn parts is the foundation for remanufacturing system based on robotic arc welding, because it can provide 3D geometric information for robot task plan. In this investigation, a novel 3D reconstruction system based on linear structured light vision sensing is developed. This system hardware consists of a MTC368-CB CCD camera, a MLH-645 laser projector and a DH-CG300 image grabbing card. This system software is developed to control the image data capture. In order to reconstruct the 3D geometric information from the captured image, a two steps rapid calibration algorithm is proposed. The 3D reconstruction experiment shows a satisfactory result.
基金Supported by the Shanxi Provincial Key Research and Development Program,No.201903D321175.
文摘BACKGROUND Hernia is a common condition requiring abdominal surgery.The current standard treatment for hernia is tension-free repair using meshes.Globally,more than 200 new types of meshes are licensed each year.However,their clinical applications are associated with a series of complications,such as recurrence(10%-24%)and infection(0.5%-9.0%).In contrast,3D-printed meshes have significantly reduced the postoperative complications in patients.They have also shortened operating time and minimized the loss of mesh materials.In this study,we used the myopectineal orifice(MPO)data obtained from preoperative computer tomography(CT)-based 3D reconstruction for the production of 3D-printed biologic meshes.AIM To investigate the application of multislice spiral CT-based 3D reconstruction technique in 3D-printed biologic mesh for hernia repair surgery.METHODS We retrospectively analyzed 60 patients who underwent laparoscopic tension-free repair for inguinal hernia in the Department of General Surgery of the First Hospital of Shanxi Medical University from September 2019 to December 2019.This study included 30 males and 30 females,with a mean age of 40±5.6 years.Data on the MPO were obtained from preoperative CT-based 3D reconstruction as well as from real-world intraoperative measurements for all patients.Anatomic points were set for the purpose of measurement based on the definition of MPO:A:The pubic tubercle;B:Intersection of the horizontal line extending from the summit of the inferior edge of the internal oblique and transversus abdominis and the outer edge of the rectus abdominis,C:Intersection of the horizontal line extending from the summit of the inferior edge of the internal oblique and transversus abdominis and the inguinal ligament,D:Intersection of the iliopsoas muscle and the inguinal ligament,and E:Intersection of the iliopsoas muscle and the superior pubic ramus.The distance between the points was measured.All preoperative and intraoperative data were analyzed using the t test.Differences with P<0.05 were considered significant in comparative analysis.RESULTS The distance between points AB,AC,BC,DE,and AE based on preoperative and intraoperative data was 7.576±0.212 cm vs 7.573±0.266 cm,7.627±0.212 cm vs 7.627±0.212 cm,7.677±0.229 cm vs 7.567±0.786 cm,7.589±0.204 cm vs 7.512±0.21 cm,and 7.617±0.231 cm vs 7.582±0.189 cm,respectively.All differences were not statistically significant(P>0.05).CONCLUSION The use of multislice spiral CT-based 3D reconstruction technique before hernia repair surgery allows accurate measurement of data and relationships of different anatomic sites in the MPO region.This technique can provide precise data for the production of 3D-printed biologic meshes.
基金supported in part by the National Natural Science Foundation of China under Grant 61902311in part by the Japan Society for the Promotion of Science(JSPS)Grants-in-Aid for Scientific Research(KAKENHI)under Grant JP18K18044.
文摘With increasingly more smart cameras deployed in infrastructure and commercial buildings,3D reconstruction can quickly obtain cities’information and improve the efficiency of government services.Images collected in outdoor hazy environments are prone to color distortion and low contrast;thus,the desired visual effect cannot be achieved and the difficulty of target detection is increased.Artificial intelligence(AI)solutions provide great help for dehazy images,which can automatically identify patterns or monitor the environment.Therefore,we propose a 3D reconstruction method of dehazed images for smart cities based on deep learning.First,we propose a fine transmission image deep convolutional regression network(FT-DCRN)dehazing algorithm that uses fine transmission image and atmospheric light value to compute dehazed image.The DCRN is used to obtain the coarse transmission image,which can not only expand the receptive field of the network but also retain the features to maintain the nonlinearity of the overall network.The fine transmission image is obtained by refining the coarse transmission image using a guided filter.The atmospheric light value is estimated according to the position and brightness of the pixels in the original hazy image.Second,we use the dehazed images generated by the FT-DCRN dehazing algorithm for 3D reconstruction.An advanced relaxed iterative fine matching based on the structure from motion(ARI-SFM)algorithm is proposed.The ARISFM algorithm,which obtains the fine matching corner pairs and reduces the number of iterations,establishes an accurate one-to-one matching corner relationship.The experimental results show that our FT-DCRN dehazing algorithm improves the accuracy compared to other representative algorithms.In addition,the ARI-SFM algorithm guarantees the precision and improves the efficiency.
基金Supported by the National Natural Science Foundation of China(61876106)。
文摘Fabric pilling evaluation has been considered as an essential element for textile quality inspection. Traditional manual method is still based on human eyes and brain, which is subjective with low efficiency. This paper proposes an objective evaluation method based on semi-calibrated near-light Photometric Stereo(PS). Fabric images are digitalized by self-developed image acquisition system. The 3D depth information of each point could be obtained by PS algorithm and then mapped to 2D grayscale image. After that, the non-textured image could be filtered by using the Gaussian low-pass filter. The pilling segmentation is conducted by using global iterative threshold segmentation method,and then K-Nearest Neighbor(KNN) is finally selected as a tool for the grade classification of fabric pilling. Our experimental results show that the proposed evaluation system could achieve excellent judging performance for the objective pilling evaluation.
基金This work is funded by the Training Plan for Young Backbone Teachers in Colleges and Universities in Henan Province under Grant No.2021GGJS077.
文摘The three-dimensional(3D)reconstruction technology based on structured light has been widely used in the field of industrial measurement due to its many advantages.Aiming at the problems of high mismatch rate and poor real-time performance caused by factors such as system jitter and noise,a lightweight stripe image feature extraction algorithm based on You Only Look Once v4(YOLOv4)network is proposed.First,Mobilenetv3 is used as the backbone network to effectively extract features,and then the Mish activation function and Complete Intersection over Union(CIoU)loss function are used to calculate the improved target frame regression loss,which effectively improves the accuracy and real-time performance of feature detection.Simulation experiment results show that the model size after the improved algorithm is only 52 MB,the mean average accuracy(mAP)of fringe image data reconstruction reaches 82.11%,and the 3D point cloud restoration rate reaches 90.1%.Compared with the existing model,it has obvious advantages and can satisfy the accuracy and real-time requirements of reconstruction tasks in resource-constrained equipment.
基金Key R&D Program of Hebei Province,No.223777101D.
文摘BACKGROUND For treatment of hilar cholangiocarcinoma(HCCA),the rate of radical resection is low and prognosis is poor,and preoperative evaluation is not sufficiently accurate.3D visualization has the advantage of giving a stereoscopic view,which makes accurate resection of HCCA possible.AIM To establish precise resection of HCCA based on eOrganmap 3D reconstruction and full quantification technology.METHODS We retrospectively analyzed the clinical data of 73 patients who underwent HCCA surgery.All patients were assigned to two groups.The traditional group received traditional 2D imaging planning before surgery(n=35).The eOrganmap group underwent 3D reconstruction and full quantitative technical planning before surgery(n=38).The preoperative evaluation,anatomical classification of hilar hepatic vessels,indicators associated with surgery,postoperative complications,liver function,and stress response indexes were compared between the groups.RESULTS Compared with the traditional group,the amount of intraoperative blood loss in the eOrganmap group was lower,the operating time and postoperative intestinal ventilation time were shorter,and R0 resection rate and lymph node dissection number were higher(P<0.05).The total complication rate in the eOrganmap group was 21.05%compared with 25.71%in the traditional group(P>0.05).The levels of total bilirubin,Albumin(ALB),aspartate transaminase,and alanine transaminase in the eOrganmap group were significantly different from those in the traditional group(intergroup effect:F=450.400,79.120,95.730,and 13.240,respectively;all P<0.001).Total bilirubin,aspartate transaminase,and alanine transaminase in both groups showed a decreasing trend with time(time effect:F=30.270,17.340,and 13.380,respectively;all P<0.001).There was an interaction between patient group and time(interaction effect:F=3.072,2.965,and 2.703,respectively;P=0.0282,0.032,and 0.046,respectively);ALB levels in both groups tended to increase with time(time effect:F=22.490,P<0.001),and there was an interaction effect between groups and time(interaction effect:F=4.607,P=0.004).In the eOrganmap group,there was a high correlation between the actual volume of intraoperative liver specimen resection and the volume of preoperative virtual liver resection(t=0.916,P<0.001).CONCLUSION The establishment of accurate laparoscopic resection of hilar cholangiocarcinoma based on preoperative eOrganmap 3D reconstruction and full quantization technology can make laparoscopic resection of hilar cholangiocarcinoma more accurate and safe.
基金This project is supported by National Natural Science Foundation ofChina (No.50375047).
文摘A fast 3D reconstruction method based on structured light to measure various parameters of the raceway groove is presented. Digital parallel grating stripes distributed with sine density are projected onto the raceway groove by a DLP projector, and distorting of stripes is happened on the raceway. Simultaneously, aided by three-step phase-shifting approach, three images covered by different stripes are obtained by a high-resolution CCD camera at the same location, thus a more accuracy local topography can be obtained. And then the bearing is rotated on a high precision computer controlled rotational stage. Three images are also obtained as the former step at next planned location triggered by the motor. After one cycle, all images information is combined through the mosaics. As a result, the 3D information of raceway groove can be gained. Not only geometric properties but also surface flaws can be extracted by software. A preliminary hardware system has been built, with which some geometric parameters have been extracted from reconstructed local topography.
文摘While Kinect was originally designed as a motion sensing input device of the gaming console Microsoft Xbox 360 for gaming purposes, it＇s easy-to-use, low-cost, reliability, speed of the depth measurement and relatively high quality of depth measurement make it can be used for 3D reconstruction. It could make 3D scanning technology more accessible to everyday users and turn 3D reconstruction models into much widely used asset for many applications. In this paper, we focus on Kinect 3D reconstruction.
