Quality Of 3D Reconstruction Research Articles

MDCT angiographic findings of various congenital pulmonary artery anomalies in pediatric patients

BackgroundCongenital pulmonary artery anomalies are variable and need proper diagnosis and treatment. CT angiography with multiplanar reconstruction has the main role in the assessment of these anomalies and this non-invasive method should be the method of choice for preoperative planning and postoperative follow up.The aim of the study is to assess the value of MDCT in the detection of pulmonary arteries anomalies in the pediatric population with complex congenital heart disease in conjunction with echocardiography as an alternative to conventional angiography and to determine the superiority of MDCT in the assessment of other abnormalities such as airway anomalies.ResultsIn our retrospective study, 52 patients (28 male and 24 females, aged 1 day to 4 years: mean age 2 years) were examined with contrast-enhanced CT. CT examinations were done using a 128-section CT scanner (Siemens Somatom Definition AS) using non-ionic iodinated contrast media. 2D and 3D reconstructions were performed. The correlation was made with echocardiograms. All imaging studies were reviewed. The echo was done to all patients. Surgery and/or catheter angiography performed to all patients, their findings were reviewed and compared to CTA findings. Other abnormalities such as congenital airway anomalies are detected using axial MDCT images and reconstructed imaging techniques.MDCT was accurate in revealing pulmonary artery anomalies. The commonest pulmonary artery anomaly was atresia, stenosis then hypoplasia. These anomalies may be isolated or associated with other congenital heart diseases. In the current study, MDCT could diagnose all cases of pulmonary arterial anomalies with 96% sensitivity, 100% specificity, 98% accuracy, 100% positive, and 94% negative predictive values. CT scans provide accurate information to assess complex spatial relationships of vascular airway compression frequently associated with CHD in the pediatric population.ConclusionMDCT scanner can be an alternative to diagnostic conventional angiography for the non-invasive assessment of the pulmonary artery. Higher quality multiplanar and 3D reconstruction achieved by the MDCT scanners offer a rapid, reliable and non-invasive technique that can be used for the evaluation and preoperative assessment of thoracic vascular and extra-vascular anatomy in infants and children with suspected congenital heart disease. CT technologies are constantly developing collaboration between radiologists, pediatric cardiologists, and anesthesiologists, which is essential for improving CT performance.

MAPPING QUALITY EVALUATION OF MONOCULAR SLAM SOLUTIONS FOR MICRO AERIAL VEHICLES

Abstract. Monocular simultaneous localization and mapping (SLAM) attracted much attention in the mobile-robotics domain over the past decades along with the advancements of small-format, consumer-grade digital cameras. This is especially the case for micro air vehicles (MAV) due to their payload and power limitations. The quality of global 3D reconstruction by SLAM solutions is a critical factor in occupancy-grid mapping, obstacle avoidance, and map representation. Although several benchmarks have been created in the past to evaluate the quality of vision-based localization and trajectory-estimation, the quality of mapping products has been rarely studied. This paper evaluates the quality of three state-of-the-art open-source monocular SLAM solutions including LSD-SLAM, ORB-SLAM, and LDSO in terms of the geometric accuracy of the global mapping. Since there is no ground-truth information of the testing environment in existing visual SLAM benchmark datasets (e.g., EuRoC, TUM, and KITTI), an evaluation dataset using a quadcopter and a terrestrial laser scanner is created in this work. The dataset is composed of the image data extracted from the recorded videos by flying a drone in the test environment and the high-fidelity point clouds of the test area acquired by a terrestrial laser scanner as the ground truth reference. The mapping quality evaluation of the three SLAM algorithms was mainly conducted on geometric accuracy comparisons by calculating the deviation distance between each SLAM-derived point clouds and the laser-scanned reference. The mapping quality was also discussed with respect to their noise levels as well as further applications.

Simultaneous shape and camera-projector parameter estimation for 3D endoscopic system using CNN-based grid-oneshot scan.

For effective in situ endoscopic diagnosis and treatment, measurement of polyp sizes is important. For this purpose, 3D endoscopic systems have been researched. Among such systems, an active stereo technique, which projects a special pattern wherein each feature is coded, is a promising approach because of simplicity and high precision. However, previous works of this approach have problems. First, the quality of 3D reconstruction depended on the stabilities of feature extraction from the images captured by the endoscope camera. Second, due to the limited pattern projection area, the reconstructed region was relatively small. In this Letter, the authors propose a learning-based technique using convolutional neural networks to solve the first problem and an extended bundle adjustment technique, which integrates multiple shapes into a consistent single shape, to address the second. The effectiveness of the proposed techniques compared to previous techniques was evaluated experimentally.

The potential of Pléiades images with high angle of incidence for reconstructing the coastal cliff face in Normandy (France)

To monitor chalk cliff face along the Normandy coast (NW France) which is prone to erosion, we tested the potential of cliff face 3D reconstruction using pairs of images with high angle of incidence at different dates from the agile Pléiades satellites. The verticality aspect of the cliff face brings difficulties in the 3D reconstruction process. Furthermore, the studied area is challenging mainly because the cliff face is north-oriented (shadow). Pléiades images were acquired over several days (multi-date stereoscopic method) with requested incidence angles until 40°. 3D reconstructions of the cliff face were compared using two software: ASP® and ERDAS IMAGINE®. Our results are twofold. Firstly, despite ASP® provides denser point clouds than ERDAS IMAGINE® (an average of 1.60 points/m² from 40° incidence angle stereoscopic pairs on the whole cliff face of Varengeville-sur-Mer against 0.77 points/m² respectively), ERDAS IMAGINE® provides more reliable point clouds than ASP® (precision assessment on the Varengeville-sur-Mer cliff face of 0.31 m ± 2.53 and 0.39 m ± 4.24 respectively), with a better spatial distribution over the cliff face and a better representation of the cliff face shape. Secondly, the quality of 3D reconstructions depends mostly on the amount of noise from raw images and on the shadow intensity on the cliff face (radiometric quality of images).

Three-dimensional reconstruction based on multi-view photometric stereo fusion technology in movies special-effect

3D movies have become very popular in recent years. But there are vertical disparities between left and right views of 3D frames due to the lack of accuracy of the mechanical alignment during the shooting. In order to improve accuracy of reconstruction, a three-dimensional reconstruction technology based on multi-view photometric stereo fusion algorithm in movies special-effect production is present in this paper. The original normal is firstly replaced with the surface normal in the average normal, and the reconstructed normal is optimized so as to reduce the deviation of the original surface normal. And then, a reference-plane-based approach is applied to estimate the principle optical axis of each light source as well as its principle radiant energy. For each surface point on the target, the direction and intensity of its incident light ray can be precisely determined by the calibration parameters and the quasi-point light model. Finally, 3D reconstruction of the surface with a quasi-point light source is also implemented in two steps. By estimating the mean value of depth in the iterative process, the surface depth is projected into the physical coordinates. Qualitative and quantitative experimental results show that higher accuracy surface normal as well as better 3D reconstruction quality can be obtained by the proposed approach in comparison with conventional reconstruction methods.

Real-Time 3D Motion Tracking and Reconstruction System Using Camera and IMU Sensors

Jointly using the data of visual and inertial sensors to achieve higher accuracy and robustness is a problem of high computational complexity. In this work, this paper presents a 3D motion tracking and reconstruction system on a mobile device using camera and inertial measurement unit (IMU) sensors. The mobile device has very small active infrared projection depth sensors with high-performance IMU and wide field of view cameras. We utilize visual-inertial odometry (VIO) to integrate visual and IMU data. However, traditional VIO method is too complex to apply to a mobile device with limited computing resources in real time. We employ sliding window filter (SWF) for the proposed system to achieve accurate 3D motion tracking and high-quality reconstruction models in real time based on delayed state marginalization. Moreover, we also extensively evaluate in the dataset and real world. Many experiments show the accurate trajectories and high quality 3D reconstruction models by the proposed system. To the end, the qualitative and quantitative experimental results indicate the proposed system based on SWF has much better performance than existing methods in motion tracking and 3D reconstruction models.

Radiation Dose Reduction of Computed Tomography in Complex Distal Femur Fractures: A Cadaver Study to Develop a Low Dose Scanning Protocol

CONTEXTRecent advances in diagnostic imaging have made computed tomography (CT) a widely used test in trauma patients. Consequently, the collective radiation burdened sustained by this patient population has increased substantially. The purpose of this cadaveric study was to determine if a significantly lowered CT radiation dose protocol would provide adequate imaging studies for the surgeon, using the distal femur as a model.METHODSTen adult cadaveric knee specimens were used to create Orthopaedic Trauma Association (AO/OTA) 33-C3 distal femur fractures with associated coronal plane Hoffa fractures and varying intra-articular displacements. Using a single CT scanner, each cadaver was scanned at 5 separate protocols defined by sequentially lowered radiation doses, the highest of which was one-third the value of our institution’s current protocol. These images were then evaluated by fellowship-trained orthopedic surgeons, an orthopedic trauma fellow, and residents. Observer reliability and confidence levels were calculated for measuring fracture displacement, assessing the quality of 3D reconstructions, and developing treatment plans.RESULTSAcross all reviewers and specimens, there was an average difference of 0.66 millimeters (mm) between the measured fracture gap and true fracture gap. The highest intraclass correlation coefficient (ICC) calculated for the inter-rater reliability of gap measurements was 0.983 at 75 mAs (95% CI: 0.955-0.996), followed by 0.973 (95% CI: 0.930-0.993) and 0.958 (95% CI: 0.896-0.988) at 15 mAs and 60 mAs, respectively. All 3D reconstructions obtained at 75 mAs and 45 mAs values (N = 8) were of acceptable imaging quality to all reviewers, while only 3 of 4 3D reconstructions obtained at 15 mAs were considered acceptable. There was no difference in treatment plans across all reviewers, regardless of radiation dose.CONCLUSIONSIn summary, our results indicate that CT scans of complex distal femur fractures at one-third the amount of radiation exposure may provide adequate imaging necessary to develop an appropriate treatment plan. At significantly lowered doses, the reviewers were able to accurately measure the amount of fracture displacement and identify the presence of each Hoffa fracture. Future studies are necessary to compare this protocol’s diagnostic capacity and limitations in evaluating complex fractures with that of our institution’s standard protocol in a clinical setting.

A Method of Non‐textured Regions Matching

It is difficult for existing dense or quasidense matching algorithms to obtain accurate matching results due to the lack of effective feature information in non-textured regions. This may affect the quality of subsequent 3D reconstruction and super-resolution reconstruction. We propose a method of non-textured regions matching. To reduce the impact of noise and illumination, vector sampling normalized cross correlation is proposed to directly measure coherence between two colour images by the effective information of multi-channel features. Three mathematical properties are used in affine transformation: 1) Centroid location is not affected by affine transformation; 2) Affine transformation transforms a straight line into another straight line; 3) Affine transformation maintains the linear relation invariant. In non-textured regions, we can construct texture regions which have affine invariant properties to improve the accuracy of template matching. This could provide more information for the diffusion of dense or quasi-dense matching. In experiments, we demonstrate that the method has high accuracy in cases where there is no big distortion between the two viewing angles from two aspects of simulated images and photographic images.

A high-throughput neurohistological pipeline for brain-wide mesoscale connectivity mapping of the common marmoset.

Understanding the connectivity architecture of entire vertebrate brains is a fundamental but difficult task. Here we present an integrated neuro-histological pipeline as well as a grid-based tracer injection strategy for systematic mesoscale connectivity mapping in the common marmoset (Callithrix jacchus). Individual brains are sectioned into ~1700 20 µm sections using the tape transfer technique, permitting high quality 3D reconstruction of a series of histochemical stains (Nissl, myelin) interleaved with tracer labeled sections. Systematic in-vivo MRI of the individual animals facilitates injection placement into reference-atlas defined anatomical compartments. Further, by combining the resulting 3D volumes, containing informative cytoarchitectonic markers, with in-vivo and ex-vivo MRI, and using an integrated computational pipeline, we are able to accurately map individual brains into a common reference atlas despite the significant individual variation. This approach will facilitate the systematic assembly of a mesoscale connectivity matrix together with unprecedented 3D reconstructions of brain-wide projection patterns in a primate brain.

A reduced micro-dose protocol for 3D reconstruction of the spine in children with scoliosis: results of a phantom-based and clinically validated study using stereo-radiography.

The aim of this study was to validate the reproducibility of 3D reconstructions of the spine using a new reduced micro-dose protocol. First, semi-quantitative image analysis was performed using an anthropomorphic child phantom undergoing low-dose biplanar radiography. This analysis was used to establish a "lowest dose" allowing for acceptable visibility of spinal landmarks. Subsequently, a group of 18 scoliotic children, 12years of age or younger, underwent full-spine biplanar radiography with both micro-dose and the newly defined reduced micro-dose. An intra- and inter-observer reliability study of 3D reconstructions of the spine was performed according to the International Organization for Standardization (ISO)-5725 standard, with three operators. The reduced micro-dose setting corresponded to a theoretical reduction of radiation dose exposure of approximately 58%. In vivo results showed acceptable intra- and inter-observer reliability (for instance, 3.8° uncertainty on Cobb angle), comparable to previous studies on 3D spine reconstruction reliability and reproducibility based on stereo-radiography. A new reduced micro-dose protocol offered reliable 3D reconstructions of the spine in patients with mild scoliosis. However, the quality of 3D reconstructions from both reduced micro-dose and micro-dose was inferior to standard-dose protocol on most parameters. Standard-dose protocol remains the option of choice for most accurate assessment and 3D reconstruction of the spine. Still, this new protocol offers a preliminary screening option and a follow-up tool for children with mild scoliosis yielding extremely low radiation and could replace micro-dose protocol for these patients. • We investigated the reliability of 3D reconstructions of the spine based on a new stereo-radiography protocol reducing radiation dose by 58% compared with established micro-dose imaging protocol. • The new reduced micro-dose protocol offers a reproducible preliminary screening option and a follow-up tool in the necessarily frequent repeat imaging of children with mild scoliosis yielding extremely low radiation and could replace existing micro-dose protocol for these patients. • EOS standard-dose protocol remains the option of choice for exact radiographic assessment of scoliosis, offering more exact 3D reproducibility of the spine compared to both micro-dose and the new reduced micro-dose protocols.

THERMAL IR IMAGING: IMAGE QUALITY AND ORTHOPHOTO GENERATION

Abstract. This paper deals with two aspects of photogrammetric processing of thermal images: image quality and 3D reconstruction quality. The first aspect of the paper relates to the influence of day light on Thermal InfraRed (TIR) images captured by an Unmanned Aerial Vehicle (UAV). Environmental factors such as ambient temperature and lack of sun light affect TIR image quality. We acquire image sequences of the same object during day and night and compare the generated orthophotos according to different metrics like contrast and signal-to-noise ratio (SNR). Our experiments show that performing TIR image acquisition during night time provides a better thermal contrast, regardless of whether we compute contrast over the whole image or over small patches. The second aspect investigated in this work is the potential of using TIR images for photogrammetric tasks such as the automatic generation of Digital Surface Models (DSM) and orthophotos. Due to the low geometrical resolution of a TIR camera and the low image quality in terms of contrast and noise compared to RGB images, the TIR DSM suffers from reconstruction errors and an orthophoto generated using the TIR DSM and TIR images is visibly influenced by those errors. We therefore include measurements of the UAVs positions during image capturing provided by a Global Navigation Satellite System (GNSS) receiver to retrieve position and orientation of TIR and RGB images in the same world coordinate system. To generate an orthophoto from TIR images, they are projected onto the DSM reconstructed from RGB images. This procedure leads to a TIR orthophoto of much higher quality in terms of geometrical correctness.

A new calibration method and optimization of structure parameters under the non-ideal condition for 3D measurement system based on fiber-optic interference fringe projection

A precise and practical approach for parameters calibration of a three-dimensional (3-D) shape measurement system based on a fiber-optic interference fringe projection is proposed. Under the non-ideal condition, with lower requirements for the relative locations of each system module, we present an implicit parameter calibration to calibrate the structure parameters simply and rapidly by building the mathematical model, regardless of the environmental disturbances. And then the actual height of a measured object can be determined accurately via the derived phase-height relation. Furthermore, the measurement error model is deduced from the experimental results and the impact of that on the systematic structure parameters is also analyzed thoroughly, which both provide a reference for the optimization of a measurement system and enhance the practical applications concerning the optical profilometry techniques. The proposed method is validated by computer simulations and real data experiments, and a good 3D reconstruction quality is achieved using the proposed calibration system.

Processing apoferritin with the Appion pipeline.

The 3DEM map challenge provided an opportunity to test different algorithms and workflows for processing single particle cryo-EM data. We were interested in testing whether we could use the standard Appion workflow with minimal manual intervention to achieve similar or better resolution than other challengers. Another question we were interested in testing was what the influence of particle sorting and elimination would be on the resolution and quality of 3D reconstructions. Since apoferritin is historically a challenging particle for single particle reconstruction and the authors of the original map challenge data used only a fraction of the particles present in the dataset, we focused on the apoferritin dataset for our entry. We submitted a 3.7 Å map from 25,844 particles and a 3.6 Å map from 53,334 particles and after assessment were among the best of the apoferritin maps that were submitted. Here we present the details of our reconstruction strategy and compare our strategy to that of another high-scoring apoferritin map. Altogether, our results suggest that for a relatively conformationally homogeneous particle like apoferritin, including as many particles as possible after elimination of junk leads to the highest resolution, and the choice of parameters for custom mask creation can lead to subtle but significant changes in the resolution of 3D reconstructions.

Full 3D Reconstruction of Non-Rigidly Deforming Objects

In this article, we discuss enhanced full 360° 3D reconstruction of dynamic scenes containing non-rigidly deforming objects using data acquired from commodity depth or 3D cameras. Several approaches for enhanced and full 3D reconstruction of non-rigid objects have been proposed in the literature. These approaches suffer from several limitations due to requirement of a template, inability to tackle large local deformations and topology changes, inability to tackle highly noisy and low-resolution data, and inability to produce online results. We target online and template-free enhancement of the quality of noisy and low-resolution full 3D reconstructions of dynamic non-rigid objects. For this purpose, we propose a view-independent recursive and dynamic multi-frame 3D super-resolution scheme for noise removal and resolution enhancement of 3D measurements. The proposed scheme tracks the position and motion of each 3D point at every timestep by making use of the current acquisition and the result of the previous iteration. The effects of system blur due to per-point tracking are subsequently tackled by introducing a novel and efficient multi-level 3D bilateral total variation regularization. These characteristics enable the proposed scheme to handle large deformations and topology changes accurately. A thorough evaluation of the proposed scheme on both real and simulated data is carried out. The results show that the proposed scheme improves upon the performance of the state-of-the-art methods and is able to accurately enhance the quality of low-resolution and highly noisy 3D reconstructions while being robust to large local deformations.

Impact of Gamma Correction on Quality of Geospatial 3D Reconstructions though Photogrammetry

Impact of Gamma Correction on Quality of Geospatial 3D Reconstructions though Photogrammetry

Method of Key-Frame Selection for Video Images

In order to improve the quality and efficiency of image processing and 3D reconstruction, a key frame screening method for video image is proposed. Firstly, the image degradation is performed by the normalized image sharpness evaluation function based on the image degradation caused by camera motion and key frame segmentation. Secondly, in order to avoid the problem of low accuracy and motion degradation of triangulation, the geometric robust information criterion GRIC was used as the screening condition to screen the results of shallow screening. The experimental results show that the three-dimensional reconstruction of the selected image sequence as the source image reduces the number of feature points, shortens the time of feature matching, and improves the efficiency of 3D reconstruction by 3.6 times.

Single shot laser speckle based 3D acquisition system for medical applications

The state of the art techniques used by medical practitioners to extract the three-dimensional (3D) geometry of different body parts requires a series of images/frames such as laser line profiling or structured light scanning. Movement of the patients during scanning process often leads to inaccurate measurements due to sequential image acquisition. Single shot structured techniques are robust to motion but the prevalent challenges in single shot structured light methods are the low density and algorithm complexity. In this research, a single shot 3D measurement system is presented that extracts the 3D point cloud of human skin by projecting a laser speckle pattern using a single pair of images captured by two synchronized cameras. In contrast to conventional laser speckle 3D measurement systems that realize stereo correspondence by digital correlation of projected speckle patterns, the proposed system employs KLT tracking method to locate the corresponding points. The 3D point cloud contains no outliers and sufficient quality of 3D reconstruction is achieved. The 3D shape acquisition of human body parts validates the potential application of the proposed system in the medical industry.

Automatic Field-of-View Expansion using Deep Features and Image Stitching

Automatic photo enhancement, such field-of-view expansion, has become a challenging problem in computer graphics community. Due to the hardware limitation, image acquisition might get distracted by small field-of-view. Photo enhancement using internet photo collections has gained good performance in the past few years. However, it depends on the quality of 3D reconstruction. In this paper, we perform an automatic personal photo enhancement using the photo collection without any 3D reconstruction step. 2D global descriptor is used using NetVLAD deep architecture. Then, image stitching is applied for each similar candidate image. Experiment results show that the propose framework has promising results which could lead to further research.

3D Coronary Artery Reconstruction by 2D Motion Compensation Based on Mutual Information

Background3D reconstruction of the coronary arteries can provide more information in the interventional surgery. Motion compensation is one kind of the 3D reconstruction method. MethodsWe propose a novel and complete 2D motion compensated reconstruction method. The main components include initial reconstruction, forward projection, registration and compensated reconstruction. We apply the mutual information (MI) and rigidity penalty (RP) as registration measure. The advanced adaptive stochastic gradient descent (ASGD) is adopted to optimize this cost function. We generate the maximum forward projection by the simplified distance driven (SDD) projector. The compensated reconstruction adopts the MAP iterative reconstruction algorithm which is based on L0 prior. ResultsComparing with the ECG-gating reconstruction and other reference method, the evaluation indicates that the proposed 2D motion compensation leads to a better 3D reconstruction for both the rest and stronger motion phases. The algorithm compensates the residual motion and reduces the artifact largely. As the gating window width increases, the overall image noise decreases and the contrast of the vessels improves. ConclusionsThe proposed method improved the 3D reconstruction quality and reduced the artifact level. The considerable improvement in the image quality results from motion compensation increases the clinical usability of 3D coronary artery.

Split-bolus intravenous contrast material injection vs. single-bolus injection in patients following endovascular abdominal aortic repair (EVAR).

To investigate the feasibility and accuracy of a split-bolus contrast injection CTA (SB-CTA) compared to a single-bolus injection CTA (SI-CTA) protocol of post-EVAR patients. In this IRB-approved study, patients who underwent SB-CTA and SI-CTA after EVAR were evaluated. Attenuation and contrast-to-noise ratio (CNR) of vessels and parenchymal organs were assessed. Subjective enhancement of vessels, quality of 3D reconstructions, and enhancement phase of parenchymal organs were assessed by two independent readers. Endoleak visibility and incidental findings were evaluated. Fifty-six patients with SB-CTA and SI-CTA at two different time points after EVAR were included in the study (112 imaging studies, mean age 75±9years). There was no difference in attenuation and CNR for aorta between SB-CTA and SI-CTA (310±87 vs. 311±80HU, p=0.88; 31.1±12.6 vs. 29.0±12.2, p=0.20), although both were improved in the venous structures and parenchymal organs with SB-CTA (portal vein: 149±33 vs. 88±32HU, 11.6±5.6 vs. 4.3±3.4; liver: 91±19 vs. 68±20HU, 4.4±2.8 vs. 2.1±2.0; pancreas: 108±27 vs. 82±23HU, 6.4±3.1 vs. 3.6±2.4; p<0.0001). Subjectively, enhancement of veins, liver, spleen, bowel, and kidneys was improved (p<0.001), while there was no difference for the aorta (p=0.46) and 3D reconstructions (p=0.58). None of the 18/56 (32.1%) endoleaks was missed with the SB-CTA. More incidental findings were detected with SB-CTA (158/56) compared to SI-CTA (135/56), p<0.001. SB-CTA was able to adequately characterize 7/15 (47%) of the findings that needed further evaluation in SI-CTA. 4/15 (26%) incidental findings that were missed with SI-CTA but seen with SB-CTA required further evaluation. Split-bolus injection CTA in the follow-up of EVAR patients allows reliable visualization of endoleaks with improved subjective and objective image quality.