Geometric Texture Research Articles

A generic parallel computational framework of lifting wavelet transform for online engineering surface filtration

Nowadays, complex geometrical surface texture measurement and evaluation require advanced filtration techniques. Discrete wavelet transform (DWT), especially the second-generation wavelet (Lifting Wavelet Transform – LWT), is the most adopted one due to its unified and abundant characteristics in measured data processing, geometrical feature extraction, manufacturing process planning, and production monitoring. However, when dealing with varied complex functional surfaces, the computational payload for performing DWT in real-time often becomes a core bottleneck in the context of massive measured data and limited computational capacities. It is a more prominent problem for the areal surface texture filtration by using 2D DWT. To address the issue, this paper presents a generic parallel computational framework for lifting wavelet transform (GPCF-LWT) based on Graphics Process Unit (GPU) clusters and the Compute Unified Device Architecture (CUDA). Due to its cost-effective hardware design and the powerful parallel computing capacity, the proposed framework can support online (or near real-time) engineering surface filtration for micro- and nano-scale surface metrology through exploring a novel parallel method named LBB model, the improved algorithms of lifting scheme and three implementation optimizations on the heterogeneous multi-GPU systems. The innovative approach enables optimizations on individual GPU node through an overarching framework that is capable of data-oriented dynamic load balancing (DLB) driven by a fuzzy neural network (FNN). The paper concludes with a case study on filtering and extracting manufactured surface topographical characteristics from real surfaces. The experimental results have demonstrated substantial improvements on the GPCF-LWT implementation in terms of computational efficiency, operational robustness, and task generalization.

A 3D nodule candidate detection method supported by hybrid features to reduce false positives in lung nodule detection

Lungs cancer is a fatal disease. However, its early detection increases the chances of survival among patients. An automated nodule detection system provides the second opinion to radiologists in early diagnosis. In this paper, an automated technique for nodule detection and classification is presented. Firstly, the lung region is extracted on the basis of the optimal gray level threshold. In the next phase, a novel hybrid 3D nodule candidate detection method is presented, comprises of Active Contour Model (ACM), 3D neighborhood connectivity and geometric properties based rules. A hybrid feature vector is created, by combining geometric texture and Histogram of Oriented Gradient reduced by Principle Component Analysis (HOG-PCA) features, for each nodule candidate. After feature extraction, classification is performed by applying four different classifiers including k-Nearest Neighborhood (k-NN), Naive Bayesian, Support Vector Machine (SVM) and AdaBoost. The evaluation is performed over Lung Image Database Consortium (LIDC) database. It is evident that AdaBoost has outperformed all other classifiers regarding accuracy, sensitivity, specificity and FPs/scan. Moreover, the proposed technique has shown significantly better results as compared to other existing methods reported in the literature.

Advanced Volumetric Capture and Processing

Volumetric video is regarded worldwide as the next important development in media production, especially in the context of rapidly evolving virtual and augmented reality markets where volumetric video is becoming a key technology. Fraunhofer Heinrich Hertz Institute (HHI) has developed a novel technology for volumetric video. The 3D Human Body Reconstruction (3DHBR) technology captures real persons with our novel volumetric capture system and creates naturally moving dynamic 3D models, which can then be observed from arbitrary viewpoints in a virtual or augmented reality scene. The capture system consists of an integrated multicamera and lighting system for a full 360° acquisition. A cylindrical studio has been developed with a diameter of 6 m and consists of 32 20-MPixel cameras and 120 light-emitting-diode (LED) panels that allow for an arbitrary lit background. Hence, diffuse lighting and automatic keying are supported. The avoidance of green screen and provision of diffuse lighting offers the best possible conditions for relighting of the dynamic 3D models afterward at the design stage of the virtual reality (VR) experience. In contrast to classical character animation, facial expressions and moving clothes are reconstructed at high geometrical detail and texture quality. The complete workflow is fully automatic, requires about 12 hr/min of mesh sequence, and provides a high level of quality for immediate integration in virtual scenes. Meanwhile, a second, professional studio has been built up on the film campus of Potsdam Babelsberg. This studio is operated by VoluCap GmbH, a joint venture between Studio Babelsberg, ARRI, UFA, Interlake, and Fraunhofer Heinrich Hertz Institute (HHI).

Evaluation of the influence of selected parameters of Selective Laser Sintering technology on surface topography

The paper presents the preliminary research results of studies of the influence of basic technological parameters of the additive technology, Selective Laser Sintering (SLS), on the quality of geometric surface texture. Microstructural studies included the determination of two-dimensional as well as three-dimensional parameters of surface texture. The study includes the influence of technological parameters such as the orientation of the models on the virtual building platform, the energy density transmitted to the sintered powder layer and the building layer thickness. The results confirm that the above mentioned parameters have a significant influence on the geometrical surface texture. The most favorable variants of the location of the models on the building platform, the value of layer thickness and energy density applied to sintered powder were determined.

An appearance‐preserving simplification method for complex 3D building models

AbstractThe simplification of 3D building models to effectively reduce model complexity and improve rendering efficiency is an important component of 3D GIS. To reduce the data volume while preserving the model appearance, this article proposes a novel simplification method for complex 3D building models. Texture discontinuities are addressed by developing a new data model that records the mapping relation between the texture coordinates of each vertex and its neighboring triangles. The surface mesh of the building model is then segmented into regions, guided by topology and appearance. Finally, the mesh segmentation information is used to derive an improved error metric that considers both geometric and texture errors, and the texture coordinates are adjusted after each simplification operation. A series of comparative experiments alongside traditional methods demonstrates that our approach achieves a good balance between geometric fidelity and texture preservation, and produces simplified 3D building models with better visual quality.

Deep Residual Texture Network for Terrain Recognition

In terrain recognition, the terrain image not only has texture features but also contains spatial features. However, the traditional feature descriptors mainly focus on the texture features of terrain image, ignoring the spatial features of the image, and the convolutional neural network (CNN) can extract the spatial features of the image well due to the role of convolutional layer and the pooling layer. So how to extract these two features at the same time is a challenging problem. In this paper, we introduce a deep residual texture network (DrtNet) that builds a texture detail layer in the residual convolution network and becomes an end-to-end learning network. DrtNet, which simultaneously extracts the spatial geometric features and texture detail features of the terrain image, can successfully combine the traditional texture feature descriptor with the convolutional neural network. The experimental results show that the DrtNet achieves the accuracy of 97.85% on the SDU_Terrain16 dataset that was created by us and outperforms other traditional methods and current popular deep convolutional networks. In addition, DrtNet has achieved good results on two other material/texture datasets (GTOS and DTD).

The PAN and MS Image Pansharpening Algorithm Based on Adaptive Neural Network and Sparse Representation in the NSST Domain

How to improve the spatial resolution as much as possible while maintaining the spectral information of multi-spectral (MS) image in the field of image fusion is of great significance for practical applications, such as map updating, feature classification, and target recognition. To analyze the coefficients of the subband distribution characteristics, in this paper, we propose a new panchromatic (PAN) and MS image pansharpening model based on an adaptive neural network and sparse representation in the non-subsample shearlet transform (NSST) domain. First, this algorithm is specific to regional directional characteristics in the high-frequency subband of PAN and MS images, and we propose an adaptive pulse coupled neural network (PCNN) model. The model can adaptively calculate the link strength of a neural cell based on the region energy. Furthermore, we apply the model to the high-frequency fusing process with the corresponding fusion rule, and the rule can distinguish the high-frequency coefficients by ignition times, which can more effectively capture the geometric texture information and detailed information in the PAN image, enhancing the spatial resolution of the fused image. Second, because of the low-frequency sub-bands from the PAN image and I component obtained by intensity-hue-saturation (IHS) transformation of the MS images with high similarity to the original image but poor sparsity, we select a set of PAN images for learning, a more targeted over-complete dictionary for low-frequency sub-band sparse representation is obtained. Then, the larger absolute value of the sparse matrix is selected to obtain the low-frequency coefficients for the fusion image while maintaining the MS spectral information effectively, and the representation of characteristic information of low-frequency subband is more effective. A large number of simulation experiments verify the effectiveness of the proposed method.

A computer-aided method based on geometrical texture features for a precocious detection of fetal Hydrocephalus in ultrasound images.

Hydrocephalus is the most common anomaly of the fetal head characterized by an excessive accumulation of fluid in the brain processing. The diagnostic process of fetal heads using traditional evaluation techniques are generally time consuming and error prone. Usually, fetal head size is computed using an ultrasound (US) image around 20-22 weeks, which is the gestational age (GA). Biometrical measurements are extracted and compared with ground truth charts to identify normal or abnormal growth. In this paper, an attempt has been made to enhance the Hydrocephalus characterization process by extracting other geometrical and textural features to design an efficient recognition system. The superiority of this work consists of the reduced time processing and the complexity of standard automatic approaches for routine examination. This proposed method requires practical insidiousness of the precocious discovery of fetuses' malformation to alert the experts about the existence of abnormal outcome. The first task is devoted to a proposed pre-processing model using a standard filtering and a segmentation scheme using a modified Hough transform (MHT) to detect the region of interest. Indeed, the obtained clinical parameters are presented to the principal component analysis (PCA) model in order to obtain a reduced number of measures which are employed in the classification stage. Thanks to the combination of geometrical and statistical features, the classification process provided an important ability and an interesting performance achieving more than 96% of accuracy to detect pathological subjects in premature ages. The experimental results illustrate the success and the accuracy of the proposed classification method for a factual diagnostic of fetal head malformation.

FrankenGAN

Coarse building mass models are now routinely generated at scales ranging from individual buildings to whole cities. Such models can be abstracted from raw measurements, generated procedurally, or created manually. However, these models typically lack any meaningful geometric or texture details, making them unsuitable for direct display. We introduce the problem of automatically and realistically decorating such models by adding semantically consistent geometric details and textures. Building on the recent success of generative adversarial networks (GANs), we propose F ranken GAN, a cascade of GANs that creates plausible details across multiple scales over large neighborhoods. The various GANs are synchronized to produce consistent style distributions over buildings and neighborhoods. We provide the user with direct control over the variability of the output. We allow him/her to interactively specify the style via images and manipulate style-adapted sliders to control style variability. We test our system on several large-scale examples. The generated outputs are qualitatively evaluated via a set of perceptual studies and are found to be realistic, semantically plausible, and consistent in style.

Automatic estimation of mechanical properties from fractographs using optimal anisotropic diffusion and Voronoi tessellation

With the advent of materials informatics there is a high demand of establishing automatic structure-property correlationship of materials which is a popular research topics due to the advancement of image processing and pattern recognition algorithms. Therefore, in this work, an attempt is made to estimate mechanical properties (i.e. yield strength and ductility) of AISI 304LN stainless steel from the SEM images of fracture surfaces obtained from tensile tests carried out at different strain rates using image processing techniques. As the void morphologies of fracture surfaces change systematically with the change in strain rates, the automatic detection of voids and geometrical features estimation from detected voids from the obtained are key goals of this present study. Therefore, in this work, a novel method of optimal anisotropic diffusion technique along with contrast limited adaptive histogram equalization (CLAHE), Otsu's optimal thresholding and morphological thinning operation are applied over the fractographs for edge enhancement, overcoming inhomogeneous illumination, edge segmentation and thinning, respectively, to detect voids, automatically. Then, Voronoi tessellation technique, which is a geometrical texture analysis, is utilized on these edge images of fractographs to extract four features viz. Voronoi edges, mean area, mean elongation and mean perimeter of Voronoi polygons where the linear correlation values (R2) with mechanical properties are found in the range of 0.90–0.99. A high linear correlation of features (i.e. 0.97–0.99) with the ductility is noticed as ductility is a geometrical parameter measured during fracture.

Lung nodule detection and classification based on geometric fit in parametric form and deep learning

This study presents an automated detection and classification method to facilitate the radiologists in the diagnosis process. The major problem in these systems is the inclusion of false positives in the results, which may lead to inaccurate diagnosis. A nodule detection and classification method is proposed that consists of four major phases. First, the lung region extraction is performed based on optimal gray level threshold that is computed by fractional-order Darwinian particle swarm optimization. Then, a novel nodule candidate detection method, based on geometric fit in parametric form incorporating the geometric properties of the nodules, is proposed. In the next phase, a hybrid geometric texture feature descriptor is created for better representation of the candidate nodules, which is a combination of 2D as well as 3D information about nodule candidates. Finally, a deep learning approach based on stacked autoencoder and softmax for feature reduction and classification is applied to reduce false positives. Performance analysis on the largest publically available dataset, Lung Image Database Consortium and Image Database Resource Initiative, depicts that the proposed method has significantly reduced the number of false positives to 2.8 per scan with a promising sensitivity of 95.6%. The results demonstrate the significance of the methodology in automatic lung nodule detection and classification. Furthermore, it will facilitate and provide assistance to radiologists in precise nodule detection.

Thin Cap Fibroatheroma Detection in Virtual Histology Images Using Geometric and Texture Features

Atherosclerotic plaque rupture is the most common mechanism responsible for a majority of sudden coronary deaths. The precursor lesion of plaque rupture is thought to be a thin cap fibroatheroma (TCFA), or “vulnerable plaque”. Virtual Histology-Intravascular Ultrasound (VH-IVUS) images are clinically available for visualising colour-coded coronary artery tissue. However, it has limitations in terms of providing clinically relevant information for identifying vulnerable plaque. The aim of this research is to improve the identification of TCFA using VH-IVUS images. To more accurately segment VH-IVUS images, a semi-supervised model is developed by means of hybrid K-means with Particle Swarm Optimisation (PSO) and a minimum Euclidean distance algorithm (KMPSO-mED). Another novelty of the proposed method is fusion of different geometric and informative texture features to capture the varying heterogeneity of plaque components and compute a discriminative index for TCFA plaque, while the existing research on TCFA detection has only focused on the geometric features. Three commonly used statistical texture features are extracted from VH-IVUS images: Local Binary Patterns (LBP), Grey Level Co-occurrence Matrix (GLCM), and Modified Run Length (MRL). Geometric and texture features are concatenated in order to generate complex descriptors. Finally, Back Propagation Neural Network (BPNN), kNN (K-Nearest Neighbour), and Support Vector Machine (SVM) classifiers are applied to select the best classifier for classifying plaque into TCFA and Non-TCFA. The present study proposes a fast and accurate computer-aided method for plaque type classification. The proposed method is applied to 588 VH-IVUS images obtained from 10 patients. The results prove the superiority of the proposed method, with accuracy rates of 98.61% for TCFA plaque.

A new visco-plastic self-consistent formulation implicit in dislocation-based hardening within implicit finite elements: Application to high strain rate and impact deformation of tantalum

Modeling deformation processes of materials under high strain rate and impact conditions in which the strain rates vary spatiotemporally over several orders of magnitude is challenging, especially in terms of constitutive description. Visco-plastic power-law flow rule, commonly used within crystal plasticity constitutive models, usually introduces superfluous strain rate effects, entering numerically via the slip activation criterion, which result in artificially high flow stress predictions under high strain rate deformation conditions. This paper presents a novel, implicit finite element implementation of a visco-plastic self-consistent polycrystal model that is implicit in dislocation-based hardening with removed strain rate sensitivity introduced by the power-law flow rule. In the model, the strain rate dependence of the flow stress is defined solely by the rate sensitivity of Peierls stress and the evolution of dislocation density within the hardening law, enabling the prediction of high strain rate and impact deformation. As a result, the predicted flow stress is accurate in its magnitude. The physically based strain rate and temperature sensitive model also features, in addition to Schmid, the non-Schmid activation contribution for slip, which is necessary for understanding and modeling the plastic deformation of body-centered cubic metals. The model is calibrated to simulate the strain rate- and temperature-sensitive monotonic deformation of tantalum and is subsequently applied to a Taylor impact test of the same material. Since the impact simulation required remeshing and interpolation of state variables, the spherical linear interpolation algorithm in the space of quaternions is implemented to facilitate the interpolation of texture. Predictions of the simulation were found to compare favorably with experimental measurements of post-test geometrical changes and texture evolution. The implementation and insights from these predictions are presented and discussed in the paper.

Fast Cartoon-Texture Decomposition Filtering Based License Plate Detection Method

Vehicle license plate detection is an important step in automatic license plate recognition, which is prone to be influenced by the background interference and complex environment conditions. It is known that cartoon-texture decomposition split an image into geometric cartoon and texture component, which can remove background interference away from the vehicle image. In this paper, we introduce a fast cartoon-texture decomposition filter into the detection process. Combining the edge detection, morphological filtering and Radon transform based tilt correction method, we formulate a new license plate detection algorithm. Experiment results confirm that the proposed algorithm can remove background interference away, inhibit the emergence of fake license plates, and improve the detection accuracy. Moreover, there is no inner loop iteration in the new algorithm, so it is fast and high-efficiency.

Designing and Evaluating a Mesh Simplification Algorithm for Virtual Reality

With the increasing accessibility of the mobile head-mounted displays (HMDs), mobile virtual reality (VR) systems are finding applications in various areas. However, mobile HMDs are highly constrained with limited graphics processing units (GPUs) and low processing power and onboard memory. Hence, VR developers must be cognizant of the number of polygons contained within their virtual environments to avoid rendering at low frame rates and inducing simulator sickness. The most robust and rapid approach to keeping the overall number of polygons low is to use mesh simplification algorithms to create low-poly versions of pre-existing, high-poly models. Unfortunately, most existing mesh simplification algorithms cannot adequately handle meshes with lots of boundaries or nonmanifold meshes, which are common attributes of many 3D models. In this article, we present QEM 4VR , a high-fidelity mesh simplification algorithm specifically designed for VR. This algorithm addresses the deficiencies of prior quadric error metric (QEM) approaches by leveraging the insight that the most relevant boundary edges lie along curvatures while linear boundary edges can be collapsed. Additionally, our algorithm preserves key surface properties, such as normals, texture coordinates, colors, and materials, as it preprocesses 3D models and generates their low-poly approximations offline. We evaluated the effectiveness of our QEM 4VR algorithm by comparing its simplified-mesh results to those of prior QEM variations in terms of geometric approximation error, texture error, progressive approximation errors, frame rate impact, and perceptual quality measures. We found that QEM 4VR consistently yielded simplified meshes with less geometric approximation error and texture error than the prior QEM variations. It afforded better frame rates than QEM variations with boundary preservation constraints that create unnecessary lower bounds on overall polygon count reduction. Our evaluation revealed that QEM 4VR did not fair well in terms of existing perceptual distance measurements, but human-based inspections demonstrate that these algorithmic measurements are not suitable substitutes for actual human perception. In turn, we present a user-based methodology for evaluating the perceptual qualities of mesh simplification algorithms.

An efficient morphology generation and level set representation of cementitious microstructures with arbitrarily shaped aggregates and cracks via extended finite elements

Concrete has a random and complex microstructure, the mesh generation of which could be challenging and time-consuming. This research presents a numerical procedure for morphology generation and level set representation of realistic cementitious microstructures without meshing the internal material interfaces. Central to the proposed numerical method is the exploitation of a vector level set function in the description of arbitrarily shaped aggregate inclusions, which allows bypassing the orthogonal projection operation on the inclusion interface as well as the determination of the outward normal. The proposed level set method is then coupled to the extended finite element method to simulate crack propagation in cement-based materials. To realistically simulate the microstructure of concrete, an improved approach that is capable of reproducing the complex geometric shape and surface texture of aggregates is proposed. To enhance the computational efficiency, a packing algorithm that mimics the settling of soil particles and enables dense aggregate packing patterns is developed to speed up the process of placing the generated aggregates into representative volume elements. After that, a series of numerical experiments are conducted to evaluate the accuracy and efficiency of the new technique. The evaluation demonstrates the effectiveness and potential of the proposed method.

Retinal Vessel Segmentation Using Minimum Spanning Superpixel Tree Detector.

The retinal vessel is one of the determining factors in an ophthalmic examination. Automatic extraction of retinal vessels from low-quality retinal images still remains a challenging problem. In this paper, we propose a robust and effective approach that qualitatively improves the detection of low-contrast and narrow vessels. Rather than using the pixel grid, we use a superpixel as the elementary unit of our vessel segmentation scheme. We regularize this scheme by combining the geometrical structure, texture, color, and space information in the superpixel graph. And the segmentation results are then refined by employing the efficient minimum spanning superpixel tree to detect and capture both global and local structure of the retinal images. Such an effective and structure-aware tree detector significantly improves the detection around the pathologic area. Experimental results have shown that the proposed technique achieves advantageous connectivity-area-length (CAL) scores of 80.92% and 69.06% on two public datasets, namely, DRIVE and STARE, thereby outperforming state-of-the-art segmentation methods. In addition, the tests on the challenging retinal image database have further demonstrated the effectiveness of our method. Our approach achieves satisfactory segmentation performance in comparison with state-of-the-art methods. Our technique provides an automated method for effectively extracting the vessel from fundus images.

Directed Self-Assembly of Star-Block Copolymers by Topographic Nanopatterns through Nucleation and Growth Mechanism.

Exploring the ordering mechanism and dynamics of self-assembled block copolymer (BCP) thin films under confined conditions are highly essential in the application of BCP lithography. In this study, it is aimed to examine the self-assembling mechanism and kinetics of silicon-containing 3-arm star-block copolymer composed of polystyrene (PS) and poly(dimethylsiloxane) blocks as nanostructured thin films with perpendicular cylinders and controlled lateral ordering by directed self-assembly using topographically patterned substrates. The ordering process of the star-block copolymer within fabricated topographic patterns with PS-functionalized sidewall can be carried out through the type of secondary (i.e., heterogeneous) nucleation for microphase separation initiated from the edge and/or corner of the topographic patterns, and directed to grow as well-ordered hexagonally packed perpendicular cylinders. The growth rate for the confined microphase separation is highly dependent upon the dimension and also the geometric texture of the preformed pattern. Fast self-assembly for ordering of BCP thin film can be achieved by lowering the confinement dimension and also increasing the concern number of the preformed pattern, providing a new strategy for the design of BCP lithography from the integration of top-down and bottom-up approaches.

Relief generation from 3D scenes guided by geometric texture richness

Typically, relief generation from an input 3D scene is limited to either bas-relief or high-relief modeling. This paper presents a novel unified scheme for synthesizing reliefs guided by the geometric texture richness of 3D scenes; it can generate both basand high-reliefs. The type of relief and compression coefficient can be specified according to the user’s artistic needs. We use an energy minimization function to obtain the surface reliefs, which contains a geometry preservation term and an edge constraint term. An edge relief measure determined by geometric texture richness and edge z-depth is utilized to achieve a balance between these two terms. During relief generation, the geometry preservation term keeps local surface detail in the original scenes, while the edge constraint term maintains regions of the original models with rich geometric texture. Elsewhere, in highreliefs, the edge constraint term also preserves depth discontinuities in the higher parts of the original scenes. The energy function can be discretized to obtain a sparse linear system. The reliefs are obtained by solving it by an iterative process. Finally, we apply non-linear compression to the relief to meet the user’s artistic needs. Experimental results show the method’s effectiveness for generating both bas- and high-reliefs for complex 3D scenes in a unified manner.

Integration of aerial oblique imagery and terrestrial imagery for optimized 3D modeling in urban areas

Photorealistic three-dimensional (3D) models are fundamental to the spatial data infrastructure of a digital city, and have numerous potential applications in areas such as urban planning, urban management, urban monitoring, and urban environmental studies. Recent developments in aerial oblique photogrammetry based on aircraft or unmanned aerial vehicles (UAVs) offer promising techniques for 3D modeling. However, 3D models generated from aerial oblique imagery in urban areas with densely distributed high-rise buildings may show geometric defects and blurred textures, especially on building façades, due to problems such as occlusion and large camera tilt angles. Meanwhile, mobile mapping systems (MMSs) can capture terrestrial images of close-range objects from a complementary view on the ground at a high level of detail, but do not offer full coverage. The integration of aerial oblique imagery with terrestrial imagery offers promising opportunities to optimize 3D modeling in urban areas. This paper presents a novel method of integrating these two image types through automatic feature matching and combined bundle adjustment between them, and based on the integrated results to optimize the geometry and texture of the 3D models generated from aerial oblique imagery. Experimental analyses were conducted on two datasets of aerial and terrestrial images collected in Dortmund, Germany and in Hong Kong. The results indicate that the proposed approach effectively integrates images from the two platforms and thereby improves 3D modeling in urban areas.