Search

In this paper, we study depth reconstruction via RGB-based, Sparse-Depth, and RGBd approaches. We showed that combination of RGB and Sparse Depth approach in RGBd scenario provides the best results. We also proved that the models performance can be further tuned via proper selection of architecture blocks and number of depth points guiding RGB-to-depth reconstruction. We also provide real-time architecture for depth estimation that is on par with state-of-the-art real-time depth reconstruction methods.

The 2019 fastMRI challenge was an open challenge designed to advance research in the field of machine learning for MR image reconstruction. The goal for the participants was to reconstruct undersampled MRI k-space data. The original challenge left an open question as to how well the reconstruction methods will perform in the setting where there is a systematic difference between training and test data. In this work we tested the generalization performance of the submissions with respect to various perturbations, and despite differences in model architecture and training, all of the methods perform very similarly.

Over the past few years, single-view 3D face reconstruction methods can produce beautiful 3D models. Nevertheless, the input of these works is unobstructed faces. We describe a system designed to reconstruct convincing face texture in the case of occlusion. Motivated by parsing facial features, we propose a complete face parsing map generation method guided by landmarks. We estimate the 2D face structure of the reasonable position of the occlusion area, which is used for the construction of 3D texture. An excellent anti-occlusion face reconstruction method should ensure the authenticity of the output, including the topological structure between the eyes, nose, and mouth. We extensively tested our method and its components, qualitatively demonstrating the rationality of our estimated facial structure. We conduct extensive experiments on general 3D face reconstruction tasks as concrete examples to demonstrate the method’s superior regulation ability over existing methods often break down. We further provide numerous quantitative examples showing that our method advances both the quality and the robustness of 3D face reconstruction under occlusion scenes.

Facial nerve reconstruction is defined as the surgical procedures that aim to restore facial nerve function of the mimetic muscles, therefore one of the requisites for the indication of nerve reconstruction procedures is the viability of the distal nerve branches, mimetic muscles, and the motor endplates. Herein, the authors describe the principles of facial nerve reconstruction including description of the types of nerve injuries, facial nerve reconstruction methods, timing for reconstruction, selection of distal nerve branches, and a new proposed classification of the level of facial nerve injuries in relation to the method of reconstruction.

Virtual clothing try-on, transferring a clothing image onto a target person image, is drawing industrial and research attention. Both 2D image-based and 3D model-based methods proposed recently have their benefits and limitations. Whereas 3D model-based methods provide realistic deformations of the clothing, it needs a difficult 3D model construction process and cannot handle the non-clothing areas well. Image-based deep neural network methods are good at generating disclosed human parts, retaining the unchanged area, and blending image parts, but cannot handle large deformation of clothing. In this paper, we propose CloTH-VTON that utilizes the high-quality image synthesis of 2D image-based methods and the 3D model-based deformation to the target human pose. For this 2D and 3D combination, we propose a novel 3D cloth reconstruction method from a single 2D cloth image, leveraging a 3D human body model, and transfer to the shape and pose of the target person. Our cloth reconstruction method can be easily applied to diverse cloth categories. Our method produces final try-on output with naturally deformed clothing and preserving details in high resolution.

In the beginning of 2013 I was a young psychology and economics student searching for a purpose in my research. I understood research as a way to create knowledge to find information to increase the chance for right decisions. So far so good, but the main question that was puzzling me was what “right decisions” are. After some discussions with my classmates, friends and family I realized, that decisions that lead to more happiness, well-being and quality of life might be the closest thing to “right decisions”. Therefore, I started to read about happiness research, which I have enjoyed ever since. After some reading time I realized that I found a lot of papers which described happiness differences from person to person but did not find a whole lot of papers that focused on happiness differences from moment to moment. Soon I started to realize that this is what I was mainly interested in: To understand which interventions have which effect on citizens’, employees’ or clients’ happiness. The main challenge I encountered was how to capture people’s happiness at different moments with methods such as the Experience Sampling Method or the Day Reconstruction Method without high efforts for the participants and the researchers and thus without high funding requirements. In order to solve this issue I teamed up with my best friend, the computer scientist Stephan Erdtmann, to develop the app “Happiness Analyzer”.

Dynamic contrast enhancement (DCE) MRI has been increasingly utilized in clinical practice. While machine learning (ML) applications are gaining momentum in MRI reconstruction, the dynamic nature of image acquisition for DCE-MRI limits access to a simultaneously high spatial and temporal resolution ground truth image for supervised ML applications. In this study, we introduced a pipeline to simulate the ground truth DCE-MRI k-space data from real breast perfusion images. Based on physical model and the clinical images, we estimate the perfusion parameters. Treating those as ground truth, we simulated the signal. Using our simulated images, we trained ML reconstruction models. We demonstrate the utility of our simulation pipeline using two ML models and one conventional reconstruction method. Our results suggest that, even though the image quality of the ML reconstructions seem to be very close to the simulated ground truth, the temporal pattern and its kinetic parameters may not be close to the ground truth data.

This work addresses the problem of the animal shape and pose estimation from an occluded image. Most of exisiting 3D animal reconstruction methods focus on an automatic and accurate framework in normal conditions, but ignore some exceptional occasions, such as occlusion, which limits the practical applications of estimating the animal shape and pose to a large extent. In this paper, we introduce a random elimination strategy from fully annotated joints and propose a deep neural network for SMAL parameters regression from the partial joints. Our proposed method can effectively deal with the reconstruction of animals under the scenario of an occluded image. We have conducted extensive experiments and results demonstrate that our 3D animal shape and pose estimation method can yield good performance on occluded images.

Volume Rendering (VR) from Digital Imaging and Communications in Medicine (DICOM®) data is a powerful source for the analysis and virtual representation of the human anatomy. To achieve this, accurate virtual dissections must be performed by applying the tools and different reconstruction methods offered by Volume Rendering Techniques (VRT).

Electron microscopy (EM) enables the reconstruction of neural circuits at the level of individual synapses, which has been transformative for scientific discoveries. However, due to the complex morphology, an accurate reconstruction of cortical axons has become a major challenge. Worse still, there is no publicly available large-scale EM dataset from the cortex that provides dense ground truth segmentation for axons, making it difficult to develop and evaluate large-scale axon reconstruction methods. To address this, we introduce the AxonEM dataset, which consists of two \(30\times 30\times 30~\mu \)m\(^3\) EM image volumes from the human and mouse cortex, respectively. We thoroughly proofread over 18,000 axon instances to provide dense 3D axon instance segmentation, enabling large-scale evaluation of axon reconstruction methods. In addition, we densely annotate nine ground truth subvolumes for training, per each data volume. With this, we reproduce two published state-of-the-art methods and provide their evaluation results as a baseline. We publicly release our code and data at https://connectomics-bazaar.github.io/proj/AxonEM/index.html to foster the development of advanced methods.

Soft tissue reconstruction of head and neck defects is a broad, challenging task. The key to a successful operation is choosing an appropriate reconstructive option based on the location, size, and characteristics of the defect. This chapter describes the main principles to consider when deciding on a primary reconstructive option for the defects created by oncologic resection. These principles include placing scars in relaxed skin tension line, minimizing donor defects, and reconstructing entire facial aesthetic subunits. This chapter also focuses on the reconstruction methods that maximize function and minimize aesthetic deformity.

Quantitative susceptibility mapping (QSM) utilizes MRI phase information to estimate tissue magnetic susceptibility. The generation of QSM requires solving ill-posed background field removal (BFR) and field-to-source inversion problems. Because current QSM techniques struggle to generate reliable QSM in clinical contexts, QSM clinical translation is greatly hindered. Recently, deep learning (DL) approaches for QSM reconstruction have shown impressive performance. Due to inherent non-existent ground-truth, these DL techniques use either calculation of susceptibility through multiple orientation sampling (COSMOS) maps or synthetic data for network training, which are constrained by the availability and accuracy of COSMOS maps or domain shift when training data and testing data have different domains. To address these limitations, we propose a weakly-supervised single-step QSM reconstruction method, denoted as wTFI, to directly reconstruct QSM from the total field without BFR. wTFI uses the BFR method RESHARP local fields as supervision to perform a multi-task learning of local tissue fields and QSM, and is capable of recovering magnetic susceptibility estimates near the edges of the brain where are eroded in RESHARP and realize whole brain QSM estimation. Quantitative and qualitative evaluation shows that wTFI can generate high-quality local field and susceptibility maps in a variety of neuroimaging contexts.

Gene tree reconstruction is an important problem in phylogenetics. However, gene sequences often lack sufficient information to confidently distinguish between competing gene tree topologies. To overcome this limitation, the best gene tree reconstruction methods use a known species tree topology to guide the reconstruction of the gene tree. While such species-tree-aware gene tree reconstruction methods have been repeatedly shown to result in vastly more accurate gene trees, the most accurate of these methods often have prohibitively high computational costs.

International audience; The current challenge for image treatment, targeting the heart ventricles (LV and RV), consists on ensuring an accurate automatic segmentation aiming to identify the myocardial contours. In this paper we propose an automatic watershed-based segmentation of the left ventricle (LV)based on topology, geometry and brightness priors. We attempted to demonstrate the accuracy of our proposed approach by comparing it with other manual segmentations carried out by two clinical experts, on 20 patients. In the aim of obtaining an accurate 3D reconstruction, we propose the volumetric object reconstruction method using The Visualization Toolkit VTK .Furthermore, the 3D+t assessment is used for the examination of the dynamic behavior of the LV to reveal regions of myocardial dysfunction.

We develop supporting arguments in favour of kernel-based reconstruction methods in the recovery step of particle simulations.We strongly recommend polyharmonic spline kernels, whose key features and advantages are briefly explained.

Traumatic injuries to the knee are common and can cause significant impairment in knee function in the short and long term. Instability caused by injuries to the medial and lateral side of the knee is a complex pathology that requires in-depth knowledge of the anatomy and biomechanics. The collateral structures not only provide stability against varus and valgus forces but also provide rotational stability to the knee. These injuries are commonly seen in multiligament injuries. The medial side of the knee contains three important structures: the superficial and deep medial collateral ligaments and the posterior oblique ligament. Bosworth, Lind, and Laprade have described reconstruction methods of the medial side. The lateral structures of the knee have a complex arrangement. The lateral collateral ligament, popliteus tendon, and popliteofibular ligament are the three main structures that provide stability. Stannard, Laprade, and Larson have described reconstruction techniques involving these structures. An in-depth understanding of the knee anatomy and biomechanics is key in being able to manage these patients effectively.

Grass is a very important element of nature and it could almost be found in every natural scene. Thus grass modeling, rendering as well as simulation becomes an important task for virtual scene creation. Existing manual grass modeling and reconstruction methods have researched on generate or reconstructing plants. However, these methods do not achieve a good result for grass blades for their extremely thin shape and almost invariant surface color. Besides, current simulation and rendering methods for grasses suffer from efficiency and computation complexity problems. This paper introduces a framework that reconstructs the grass blade model from the color-enhanced depth map, simplifies the grass blade model and achieves extremely large scale grassland simulation with individual grass blade response. Our method starts with reconstructing the grass blade model. We use color information to guide the refinement of captured depth maps from cameras based on an autoregressive model. After refinement, a high-quality depth map is used to reconstruct thin blade models, which cannot be well handled by multi-view stereo methods. Then we introduce a blade simplification method according to each vertex’s movement similarity. This method takes both geometry and movement characteristics of grass into account when simplifying blade mesh. In addition, we introduce a simulation technique for extremely large grassland that achieve tile management on GPU and allow individual response for each grass blade. Our method excels at reconstructing slender grass blades as well as other similar plants, and realistic dynamic simulation for large scale grassland.

For decades, CT images were reconstructed from the measured raw data using analytical reconstruction methods, such as filtered backprojection (FBP). While FBP is fast and accurate, the measured data are rarely ideal, and iterative and statistical methods can provide significant benefit particularly when the data quality is poor. They can produce images with lower noise and can also reduce artifacts from system imperfections. However, the resulting images are nonlinear and non-stationary and have other properties that are different from those produced by FBP that should be kept in mind when the images are interpreted.

Accurate and reliable water demand forecasting is important for effective and sustainable planning and use of water supply infrastructures. In this paper, a hybrid EEMD-Elman neural network model for hourly campus water demand forecast is proposed, aiming at improving the accuracy and reliability of water demand forecast. The proposed method combines the Elman neural network, EEMD method, and phase space reconstruction method providing favorable dynamic forecast characteristics and improving the forecasting accuracy and reliability. Simulation results show that the proposed model provides a better performance of hourly campus water demand forecast by using the real data of water usage of our campus.

Surgical treatment for cancers of the head and neck in general, and for cancers of the oral cavity specifically, will often require that reconstruction is undertaken to improve functional and cosmetic outcomes. Restoration of surgical ablations and the resultant defects that occur requires reconstitution of residual tissue and the rebuilding of lost anatomical elements. Fortunately, a variety of surgical reconstructive options are available to the surgeon. The type of reconstruction technique employed is dependent on the size and complexity, as well as the location of the surgical defect. Within the literature, a range of reconstructive techniques have been described. This includes procedures involving primary closure, local flaps, or regional flaps, to complex microvascular free tissue transfers. This chapter outlines and describes these reconstruction methods. In doing so, the chapter seeks to educate other professionals who may be involved in the posttreatment rehabilitation of those who require surgical ablation and reconstruction as part of their treatment for cancer of the oral structures.

Dynamic fluorescence molecular tomography (FMT) is a promising optical imaging technique for three-dimensionally demonstrating the metabolic process of fluorophore in small animals. Conventional FMT methods focus on reconstructing static distribution of fluorescent yield, and the reconstruction results may perform poorly if the boundary measurement data, acquired from time-varying fluorophore, were directly used in these methods. In this study, we apply joint \( \ell_{1} \) and Laplacian manifold regularization model to dynamic FMT. The \( \ell_{1} \)-norm regularization method is used to deal with the ill-posedness of FMT, and the Laplacian manifold regularization is introduced to obtain spatial structure information of boundary measurements. Then, we use gradient projection for sparse reconstruction (GPSR) method to solve the joint regularization model. Since the boundary measurements are obtained from different time points, the input data is converted from a vector to a matrix, and each column of the matrix corresponds to a time point. Thus, a sequence of fluorophore concentration images, corresponding to different time points, can be reconstructed in one step. Numerical simulation experiments are performed and the results indicate that the proposed method can recover the dynamic fluorophore well.

The presented paper describes a successfully parallel implementation of the statistical reconstruction method based on the continuous-to-continuous model using both CPU and GPU hardware approaches. Data were obtained from a commercial computer tomography device which were saved in DICOM standard file. The implemented reconstruction algorithm is formulated taking in two consideration the statistical properties of signals obtained by x-ray CT and the continuous-to-continuous data model. During our experiments, we tested the speed of the implemented algorithm and we optimized it in terms of the critical parameter which is very important regarding the potential use of this solution in clinical practice.

Understanding genotype-phenotype dependency is a universal aim for all life sciences. While the complete genotype-phenotype relations remain challenging to resolve, metabolic phenotypes are moving within the reach through genome-scale metabolic model simulations. Genome-scale metabolic models are available for commonly investigated yeasts, such as model eukaryote and domesticated fermentation species Saccharomyces cerevisiae, and automatic reconstruction methods facilitate obtaining models for any sequenced species. The models allow for investigating genotype-phenotype relations through simulations simultaneously considering the effects of nutrient availability, and redox and energy homeostasis in cells. Genome-scale models also offer frameworks for omics data integration to help to uncover how the translation of genotypes to the apparent phenotypes is regulated at different levels. In this chapter, we provide an overview of the yeast genome-scale metabolic models and the simulation approaches for using these models to interrogate genotype-phenotype relations. We review the methodological approaches according to the underlying biological reasoning in order to inspire formulating novel questions and applications that the genome-scale metabolic models could contribute to. Finally, we discuss current challenges and opportunities in the genome-scale metabolic model simulations.

Facial expression recognition (FER) using 2D images has been rapidly developed in the past decade. However, existing 2D-based FER methods seldom consider the impact of identity factors, and do not utilize shape features which have been proven to be effective complement to texture features. Built upon latest 3D face reconstruction methods, this paper proposes to generate expression-induced shape deformation map (ESDM) from the 3D face reconstructed from the input 2D face image, and then extract shape feature from ESDM by using a deep network. The shape feature is then combined with the texture feature on the input 2D face image, resulting in a fused feature, based on which the expression of the input 2D face image is recognized by using a softmax classifier. Evaluation experiments on BU-3DFE, MMI and CK+ databases show that our proposed shape feature effectively improves the 2D-based FER accuracy, and our method using the fused feature achieves state-of-the-art accuracy.

The method outlined in the Chaps. 15– 21 has been used for revealing nonlinear deterministic and stochastic behaviors in a variety of problems, ranging from physics, to neuroscience, biology and medicine. In most cases, alternative procedures with strong emphasis on deterministic features have been only partly successful, due to their inappropriate treatment of the dynamical fluctuations [1]. In this chapter, we provide a list of the investigated phenomena using the introduced reconstruction method. In the “outlook” possible research directions for future are discussed.

Domes first appeared in architecture in small structures such as round huts and tombs. As the need to accommodate more people became increasingly larger through the ages and the enhanced sense of symbolism of the dome became ever more important so the dome spans expanded. The construction materials switched from solid earth mounds to masonry hemispheres [1]. This growth in size has always been an engineering challenge and in many cases of masonry domed structures the dome, located at the highest point of the building, is considered to be one of the most vulnerable parts of the structure. Decreasing the weight concentrated on the highest part of the masonry structure increases the strength and durability of the structure. The aim of this paper is to present an alternative method for constructing a domed structure that was developed using computer aided design methods. This type of dome could serve as a replacement of a collapsed stone dome, usually of a church, or be fitted on a new structure.

End-stage temporomandibular joint (TMJ) disease due to a multitude pathophysiologic process can benefit from TMJ replacement. The traditional method has been used to reconstruct the TMJ using autogenous tissues from various donor sites in the body. However, as materials and understanding how to use those materials have advanced, the pendulum is now moving much more toward alloplastic reconstruction. This chapter will compare and contrast the two types of reconstruction methods for reconstructing the diseased TMJ.

For a graph signal in the low-frequency subspace, the missing data can be reconstructed through the sampled data by exploiting the smoothness of the graph signal. In this chapter, the concepts of local set and centerless local set are introduced and several iterative methods are presented to reconstruct bandlimited graph signal from decimated data or measured signal. These algorithms are built on frame theory and the concepts of (centerless) local sets, based on which several frames and contraction operators are provided. We then prove that the reconstruction methods converge to the original signal under certain conditions and demonstrate the new methods lead to a significantly faster convergence compared with the baseline method. Furthermore, the correspondence between graph signal sampling and time-domain irregular sampling is analyzed comprehensively, which may be helpful to future works on graph signals. Numerical experimental results demonstrate the effectiveness of the reconstruction methods in various sampling geometries, imprecise priori knowledge of cutoff frequency, and noisy scenarios.

The knowledge of the three-dimensional structure of proteins is crucial for understanding many important biological processes. Most of the biologically relevant protein systems are too large for classical, atomistic molecular modeling tools. In such cases, coarse-grained (CG) models offer various opportunities for efficient conformational sampling and thus prediction of the three-dimensional structure. A variety of CG models have been proposed, each based on a similar framework consisting of a set of conceptual components such as protein representation, force field, sampling, etc. In this chapter we discuss these components, highlighting ideas which have proven to be the most successful. As CG methods are usually part of multistage procedures, we also describe approaches used for the incorporation of homology data and all-atom reconstruction methods.

The dual-camera compressive hyperspectral imager (DCCHI) can capture 3D hyperspectral image (HSI) with a single snapshot. However, due to the high computation complexity of reconstruction methods, DCCHI cannot apply to the time-crucial applications. In this paper, we propose a GPU assisted towards real-time reconstruction framework for DCCHI. First, leveraging the fast convergence rate of the alternative direction multiplier method, we propose a reformative reconstruction algorithm which can achieve a fast convergence rate. Then, using the interpolation results of a low resolution reconstructed HSI as the warm start, we propose a fast reconstruction strategy to further reduce the computation burden. Last, a GPU parallel implementation is presented to achieve nearly real-time reconstruction. Evaluation experiments indicate our framework can obtain a significant promotion in reconstruction efficiency with a slight accuracy loss.

From 2007 to 2017, 34 patients with subacute wounds to lower limbs were treated by the assisted healing and delayed selective reconstruction method (AH-GSR). Sixteen patients (47%) presented with a concomitant fracture; 18 patients (53%) sustained degloving with a soft tissue injury only. Negative pressure wound therapy was used in 28 patients (82.3%). Antibiotics were given to all patients, in 12 (35.3%) as prophylaxis and in 22 (64.7%) as therapy. The reconstruction was performed by split-thickness skin grafts (SG) in 16 patients (47%), by dermal substitutes (DS) in 8 patients (23.5%), by local fascio-cutaneous flaps in 2 patients (5.9%), and by free flaps in 8 patients (23.5%). In this case series, three (8.8%) complications were recorded. Adhering to the AH-GSR method of treatment of lower extremities subacute wounds guarantees results comparable to the ones obtained with the treatment of acute wounds during the first week after injury.

We present an accuracy and performance analysis of Time Coherent 3D Animation Reconstruction methods from RGB-D video. We analyze the existing methods that can reconstruct a time coherent 3D animation using RGB-D video. We also present a modified algorithm using only the RGB data that extends the analysis of existing methods. We show that using all the methods it is possible to reconstruct a time-coherent 3D animation using either only the color data, color and depth data, or only the depth data. We compare all the methods using a number of error measures and analyze the strength and weaknesses of each method in terms of their accuracy and runtime performance. Our analysis demonstrates that given RGB-D video data, it is possible to select the best algorithm for time coherent 3D animation reconstruction under a number of constraints in terms of the required accuracy and runtime performance.

The increasing use of explosive devices in recent military operations has introduced complex patterns of injury requiring new reconstructive considerations. Proper initial management, staging, traditional and advanced reconstruction methods, and the use of regenerative medicine can have a drastic effect on outcomes. Even when limb salvage is not achieved, an understanding of residual limb length preservation and peripheral nerve techniques can improve functionality within the amputee patient. Surgery utilizing regenerative modalities and peripheral nerve techniques has shown promising results in military injuries with extensive soft tissue, orthopedic, and neurovascular damage. Improved understanding of the hybrid reconstructive ladder, the increased use of innovative surgical strategies, and further study into the benefits of regenerative medicine may lead to improvements in care for both military and nonmilitary trauma patients suffering from blast injuries.

Reconstruction Methods for Fractures with Bone Defects, Vascular Injury, and Salvage Procedures

Recently, many deep learning (DL) based MR image reconstruction methods have been proposed with promising results. However, only a handful of work has been focussing on characterising the behaviour of deep networks, such as investigating when the networks may fail to reconstruct. In this work, we explore the applicability of Bayesian DL techniques to model the uncertainty associated with DL-based reconstructions. In particular, we apply MC-dropout and heteroscedastic loss to the reconstruction networks to model epistemic and aleatoric uncertainty. We show that the proposed Bayesian methods achieve competitive performance when the test images are relatively far from the training data distribution and outperforms when the baseline method is over-parametrised. In addition, we qualitatively show that there seems to be a correlation between the magnitude of the produced uncertainty maps and the error maps, demonstrating the potential utility of the Bayesian DL methods for assessing the reliability of the reconstructed images.

Among a wide range of contactless 3D fingerprint acquisition methods available today, photometric stereo based method offers low-cost and accurate recovery of 3D fingerprints. This chapter discusses such online 3D fingerprint image acquisition approach. Systematic details on such photometric stereo based setup are detailed in this chapter, i.e. from setup hardware, offline system calibration, preprocessing the raw images, specular reflection removal methods, to the choice of reconstruction methods. This chapter also shares our insights and results on the attempts made to consider non-Lambertian nature of finger surface. Resulting computational complexity for respective online 3D fingerprint imaging system also appears in this chapter.

We propose a 3-dimensional sparse sampling reconstruction method, aiming for chemical shift imaging in magnetic resonance spectroscopy. The method is a Compressed Sensing (CS) method based on the interior point optimization technique that can substantially reduce the number of sampling points required, and the method has been tested successfully in hyperpolarized 13C experimental data using two different sampling strategies.

In order to solve the problem of high sampling rate in the wideband spectrum sensing of cognitive radio, this paper studies the method of cyclic spectrum detection based on the modulation wideband converter (MWC). A novel fast cyclic spectrum detection algorithm of MWC based on Lorentzian Norm is proposed to deal with the influence of some non-ideal factors on the performance of the existing MWC system reconstruction algorithm in physical implementation. Firstly, the objective function for sparse optimization is build based on smoothed L0-norm constrained Lorentzian norm regularization. Then a parallel reconstruction method is implemented in a unified parametric framework by combining the fixed-step formula and the conjugate gradient algorithm with sufficient decent property. Simulation results demonstrate that the proposed algorithm can not only improve the recovery probability of sparse signal, but also has a higher detection probability in low SNR environment compared with traditional reconstruction algorithms.

This chapter dedicates to the evolution of research hotspots and datasets on historical farmland reconstruction in China and abroad. Based on the statistical analysis and literature review, related studies are analyzed by research paradigms, assumptions, working methods and model validation. From the perspective of methodology, there are two main types, i.e. a top-down method based on historical records and a bottom-up reconstruction method based on geographic spatial model. For model validation, direct verification of reconstruction pattern is obviously a more precise method, but it is often restricted to the spatial-temporal scales of research and data source; indirect validation method provides a new idea for accuracy evaluation of the reconstruction results.

Direct inversion methods, also known as linear inversion methods, such as holography and diffraction tomography are the working horses of microwave imaging. They provide fast qualitative estimates of an object’s shape and electrical contrast. However, these traditional methods cannot be used as linearized solvers at the core of nonlinear iterative reconstruction schemes because of their inability to provide a quantitative estimate of the electromagnetic constitutive parameters. In the past decade, advances have led to two powerful approaches to linear quantitative inversion, specifically developed for microwave imaging based on scattering-parameter data. These two approaches, quantitative microwave holography (QMH) and scattered-power mapping (SPM), provide quantitative images in real time. Thus they add new capability to real-time microwave imaging and offer new linearized core solvers for nonlinear reconstruction schemes. The performance of QMH and SPM is compared utilizing three different strategies of acquiring the resolvent kernel in the forward model: analytical, simulated, and measured. The results ascertain that the quantitative reconstruction is attainable only with experimentally acquired resolvent kernel.

Genetic networks provide compact representations of interactions between genes, and offer a systems perspective into biological processes and cellular functions. Many algorithms have been developed to estimate such networks based on steady-state gene expression profiles. However, the estimated networks using different methods are often very different from each other. On the other hand, it is not clear whether differences observed between estimated networks in two different biological conditions are truly meaningful, or due to variability in estimation procedures. In this paper, we aim to answer these questions by conducting a comprehensive empirical study to compare networks obtained from different estimation methods and for different subtypes of cancer. We evaluate various network descriptors to assess complex properties of estimated networks, beyond their local structures, and propose a simple permutation test for comparing estimated networks. The results provide new insight into properties of estimated networks using different reconstruction methods, as well as differences in estimated networks in different biological conditions.

Negative survey is a promising method which can protect personal privacy while collecting sensitive data. Most of previous works focus on negative survey models with specific hypothesis, e.g., the probability of selecting negative categories follows the uniform distribution or Gaussian distribution. Moreover, as far as we know, negative survey is never conducted with manual selection in real world. In this paper, we carry out such a negative survey and find that the survey may not follow the previous hypothesis. And existing reconstruction methods like NStoPS and NStoPS-I perform poorly on the survey data. Therefore, we propose a method called NStoPS-MLE, which is based on the maximum likelihood estimation, for reconstructing useful information from the collected data. This method also uses background knowledge to enhance its performance. Experimental results show that our method can get more accurate aggregated results than previous methods.

In this paper, we are proposing a new energy-minimization reconstruction method for the multi gray level discrete tomography. The proposed reconstruction approach combines a gradient based algorithm with the graph cuts optimization. This new technique is able to reconstruct images that consist of an arbitrary number of gray levels. We present the experimental evaluation of the new method, where we compare its performance with performance of the already suggested methods for multi-level discrete tomography. The obtained experimental results give an advantage to the proposed approach, especially regarding the quality of the reconstructed test images.

In order to identify causes of civil aviation accidents, reconstruction and splice of wreckage is an important work for survey. This paper contributes to this research by proposing a wreckage 3D reconstruction method based on sequence image. According to the wreckage of fuselage and components, we respectively describe their reconstruction methods. Experiment results show that the algorithm can be well used to complete the mosaic reconstruction of aircraft wreckage and efficiency is greatly improved. The results show that, compared with physical splicing, the research result will help shortening the period and providing effective resolution to civil aviation accident investigation.

Since the emergence of the practice of transforming architectural structures into landscapes, the relationship between architecture and the environment has gradually changed from one of binary to multivariate opposition. This chapter introduces topography and field theory in cold-region architectural design and discusses the responses of architectural structures to spaces and landscapes that are influenced by the extreme climate in cold regions through relevant case studies by blanking, simulating, and mixing architectural forms with landforms.

This chapter reviews the techniques developed for positron emission tomography (PET) image reconstruction and image property analysis. Both mathematical theory and practical considerations are introduced. We focus on the commonly used methods on commercial PET scanners, in particular model-based statistical reconstruction methods. We also briefly describe data corrections necessary for PET image reconstruction, which are important for reducing artifacts and improving quantitative accuracy. Finally some recent developments are described, including the reconstruction of time-of-flight (TOF) PET data and direct parametric image reconstruction.

This chapter discusses the systematic uncertainties associated with the Monte Carlo (MC) simulated samples used in this analysis. The simulated signal samples are discussed in Sect. 7.3, and the background samples from simulation are given in Sect. 8.1.2 For definitions and reconstruction method of the physics objects used in this analysis, see Chap. 8 The systematic uncertainty figures are used in the final statistical interpretation and limit setting in Chap. 12