Abstract Cross‐virtuality analytics (XVA) is a novel field of research within immersive and visual analytics. A broad range heterogeneous devices across the reality–virtuality continuum, along with respective metaphors analysis techniques, are currently becoming available. The goal XVA to enable that use transitional collaborative interfaces seamlessly integrate different support multiple users. In this work, we take closer look at analyse existing body work for an overview its current...

This research investigates the effectiveness of image quality metrics in predicting human perception superresolution computed tomography images. By comparing multiple and correlating their scores with expert ratings on attributes such as noise, contrast, sharpness, we aim to identify that closely align visual judgment. Our findings highlight limitations relying solely a single metric for assessment emphasize importance considering specific tasks content when selecting appropriate metrics....

In this work, we apply and adapt established probability of detection (POD) methods on the in-line inspection aluminium cylinder heads using X-ray computed tomography (XCT). We propose to use XCT simulation tool SimCT simulate virtual radiographs from specimen including artificial defects, which avoids manufacturing specimens with calibrated defects known type (e.g. pores, inclusions, cracks) characteristics size, shape, location). To quantify POD, these images are analysed ZEISS automated...

open_iA is a platform for visual analysis and processing of volumetric datasets.The main driver behind its development to provide common framework performing analytics on industrial Computed Tomography (CT) data.In contrast general volume visualization or software, it offers specialized tools, which address domain-specific scenarios such as porosity determination, fiber characterization image parameter space analysis.The wide range building blocks, these tools consist of, facilitate the new...

The comparison of many members an ensemble is difficult, tedious, and error-prone, which aggravated by often just subtle differences. In this paper, we introduce Dynamic Volume Lines for the interactive visual analysis sets 3D volumes. Each volume linearized along a Hilbert space-filling curve into 1D line plot, depicts intensities over indices. We present nonlinear scaling these plots based on intensity variations in volumes, enables more effective use available screen space. builds basis...

Abstract We present GEMSe, an interactive tool for exploring and analyzing the parameter space of multi‐channel segmentation algorithms. Our targeted user group are domain experts who not necessarily specialists. GEMSe allows exploration possible combinations a framework its ensemble results. Users start with sampling computing corresponding segmentations. A hierarchically clustered image tree provides overview variations in resulting label images. Details provided through exemplary images...

X-ray computed tomography (XCT) is one of the most powerful imaging techniques in non-destructive testing (NDT) for detecting, analysing and visualising defects such as pores, fibres, cracks etc. industrial specimens. Detecting images, however, still a challenging problem, it strongly depends on quality XCT images. Numerical simulation proved to be valuable order increase both image detection performance. In this work, we thus analyse differences between traditional segmentation (i.e.,...

Abstract Industrial X-ray computed tomography (XCT) is a crucial non-destructive testing method for quality control in various branches of industry. Accurate segmentation XCT data aids defect identification and material characterization. In this paper we present our results applying the Segment Anything Model (SAM) context. SAM, an unsupervised approach that automates without manual annotations, combines deep convolutional neural networks generative adversarial networks. We used SAM on...

Long fibre reinforced thermoplastics (LFTs) have become important in recent years, due to outstanding mechanical properties (such as high strength, stiffness, and excellent impact behaviour). Nevertheless, there is still a lack of knowledge regarding the failure micromechanics LFTs. This publication focuses on qualitative quantitative analysis influencing factors for initiation long thermoplastics. Existing new methodologies are combined detailed local microstructure micromechanical...

Abstract This paper addresses the increasing demand in industry for methods to analyze and visualize multimodal data involving a spectral modality. Two modalities are used: high‐resolution X‐ray computed tomography (XCT) structural characterization low‐resolution fluorescence (XRF) elemental decomposition. We present InSpectr, an integrated tool interactive exploration visual analysis of multimodal, multiscalar data. The has been designed around set tasks identified by domain experts fields...

We demonstrate a collaborative visualisation framework using traditional desktop-based setup together with virtual reality-based system for the analysis of rich X-Ray Computed Tomography Data (XCT) from composite materials. The supports experts in time-consuming and cognitively demanding task analysing objects, particular, curved fibres pores, along their characteristics, such as length, orientation, or shape. Our employs methods reality focusing on detailed spatial impression, combined...

Abstract We present visual analysis methods for the evaluation of tomographic fiber reconstruction algorithms by means analysis, debugging and comparison reconstructed fibers in materials science. The are integrated a tool (FIAKER) that supports entire workflow. It enables various algorithms, differently parameterized individual steps iterative algorithms. Insight into performance is obtained list‐based ranking interface. A 3D view offers interactive visualization techniques to gain deeper...

We present visual methods for the analysis and comparison of results curved fibre reconstruction algorithms, i.e., algorithms extracting characteristics fibres from X-ray computed tomography scans. In this work, we extend previous different or parametrisations to fibres. propose dissimilarity measures such apply these compare multiple a specified reference. further visualisation analyse differences between quantitatively qualitatively. two case studies, show that presented provide valuable...

This work introduces a novel Augmented Reality (AR) approach to visualize material data alongside real objects in order facilitate detailed analyses based on spatial non-destructive testing (NDT) as generated X-ray computed tomography (XCT) imaging. For this purpose, we introduce framework that leverages the potential of AR devices, visualization and interaction techniques seamlessly explore complex primary secondary XCT matched with real-world objects. The overall goal proposed analysis...

Fiber-reinforced polymers (FRPs) are of great importance in various industries because their superior properties as compared to conventional materials, versatile processing, and wide application possibilities. To fulfil the high-quality standards its respective applications, industrial 3D X-ray computed tomography (XCT) is increasingly used. It enables an accurate, non-destructive characterization material features such inclusions, voids, fibers, or other reinforcements, which core for...

This work illustrates the use of deep learning methods applied on X-ray computed tomography (XCT) datasets to segment pores and fibres in reinforced composite components from aeronautic industry by binary semantic segmentation. We first apply data pre-processing, then employ a modified 3D U-Net, representing convolutional neural network. Tweaking hyper-parameters, we have reached an optimal model for our datasets. One models has 99% segmentation accuracy when testing using Dice function. In...

The novel research field of Cross-Virtuality Analytics, where users move or interconnect stages the Reality-Virtuality continuum to analyse data, can profit in multiple ways from well researched area visual coherence. In this position paper, we present initial ideas how coherence approaches and data gathered by them be used improve general, what transition techniques could developed analysis process these approaches.

Even though it is a crucial step for achieving suitable results, the preprocessing of data before used as input to deep neural networks often only described side note. This work elaborates on required steps in this procedure. Specifically, we provide insights into selection appropriate segmentation algorithms generate reference volumes from X-ray computed tomography (XCT) scans training data. Furthermore, evaluates criteria an learning network architecture, and quantitative comparison...

This work introduces methods for analyzing the three imaging modalities delivered by Talbot-Lau grating interferometry X-ray computed tomography (TLGI-XCT). The first problem we address is providing a quick way to show fusion of all modalities. For this purpose tri-modal transfer function widget introduced. controls mixing that uses output functions modalities, allowing user create one customized fused image. A second prevalent in processing TLGI-XCT data lack tools segmentation process such...

This work explores combining classical desktop-based analysis systems with virtual reality. In this context, we provide a system for exploring objects, in particular fibers and pores, along their characteristics, such as length, orientation, or shape. employs visualization methods reality focusing on the detailed spatial impression, combined analyzing characteristics distributions. We two case studies based fiber-reinforced polymer (FRP) datasets, showcasing potential of our system. One...