WRAP - latest items
Sheet metal stamping is widely used for high-volume production. However, it is susceptible to the occurrence of defects. These defects have far-reaching implications, such as compromised product quality, increased costs, and potential safety hazards. Therefore, developing an accurate automated defect detection method is vital. This thesis addresses the specific challenges observed during split defect detection in prototype stamping components, aiming to advance the deep learning (DL) based inspection in sheet metal stamping.
The study begins by evaluating the effectiveness of DL models for classifying split defects and suggests a route toward the creation of a reliable vision-based DL model for stamping defect inspection. The visibility of defects in captured images is crucial for successful vision-based defect detection. In real stamping environments, the shiny metallic surfaces interact with overhead electric lights, sunlight, and skylights, leading to unpredictable specular reflections. To address this challenge, the thesis introduces a method that combines DL models with high-dynamic-range (HDR) imaging. This approach outperforms traditional imaging-based models, achieving enhanced accuracy while reducing false-positive predictions.
The low failure rates of stamping parts result in limited data for training DL models. This thesis introduces a method for generating a synthetic dataset of sheet metal stamping split defects to train DL models. This approach combines physics-based simulation and real defect texturing. By leveraging finite element simulation, plausible split locations are determined using a forming limit curve, and defect features are generated by mapping fine details of real splits. Incorporating synthetic images improves the detection performance and achieves a similar accuracy with significantly fewer real samples.
The transportation Lp distance, denoted TLp, has been proposed as a generalisation of Wasserstein Wp distances motivated by the property that it can be applied directly to colour or multi-channelled images, as well as multivariate time-series without normalisation or mass constraints. These distances, as with Wp, are powerful tools in modelling data with spatial or temporal perturbations. However, their computational cost can make them infeasible to apply to even moderate pattern recognition tasks. We propose linear versions of these distances and show that the linear TLp distance significantly improves over the linear Wp distance on signal processing tasks, whilst being several orders of magnitude faster to compute than the TLp distance.
Optimal transport and its related problems, including optimal partial transport, have proven to be valuable tools in machine learning for computing meaningful distances between probability or positive measures. This success has led to a growing interest in defining transport-based distances that allow for comparing signed measures and, more generally, multi-channeled signals. Transport
distances are notable extensions of the optimal transport framework to signed and possibly multi-channeled signals. In this paper, we introduce partial transport
distances as a new family of metrics for comparing generic signals, benefiting from the robustness of partial transport distances. We provide theoretical background such as the existence of optimal plans and the behavior of the distance in various limits. Furthermore, we introduce the sliced variation of these distances, which allows for faster comparison of generic signals. Finally, we demonstrate the application of the proposed distances in signal class separability and nearest neighbor classification.
The flow around a bubble, precipitously expanding in a thin gap between flat walls, was found to have a peculiar feature: distinct fingering occurs at the bubble wall, which was observed through the ultra-high speed optical visualization. The effect is attributed to the reversal of the flow within boundary layers, which provokes the growth of instabilities at the inflection point and, when the surface tension is low enough (the local Weber number is high enough), leads to the fingering. In this paper, we show the high speed recordings of the fingering and model the evolution of the radial velocity to quantitatively confirm feasibility of the proposed instability mechanism.
Objectives: To examine randomised controlled trials (RCTs) of 'hospital at home' (HAH) for admission avoidance in adults presenting with acute physical illness to identify the use of vital sign monitoring approaches and evidence for their effectiveness.
Design: Systematic review
Setting and participants: This review compared strategies for vital sign monitoring in admission avoidance HAH for adults presenting with acute physical illness. Vital sign monitoring can support HAH acute multidisciplinary care by contributing to safety, determining requirement of further assessment and guiding clinical decisions. There are a wide range of systems currently available, including reliable and automated continuous remote monitoring using wearable devices.
Methods: Eligible studies were identified through updated database and trial registries searches (2 March 2016 to 15 February 2023), and existing systematic reviews. Risk of bias was assessed using the Cochrane risk of bias 2 tool. Random effects meta-analyses were performed, and narrative summaries provided stratified by vital sign monitoring approach.
Results: 21 eligible RCTs (3459 participants) were identified. Two approaches to vital sign monitoring were characterised: manual and automated. Reporting was insufficient in the majority of studies for classification. For HAH compared to hospital care, 6-monthly mortality risk ratio (RR) was 0.94 (95% CI 0.78 to 1.12), 3-monthly readmission to hospital RR 1.02 (0.77 to 1.35) and length of stay mean difference 1.91 days (0.71 to 3.12). Readmission to hospital was reduced in the automated monitoring subgroup (RR 0.30 95%CI 0.11 to 0.86).
Conclusions and Implications: This review highlights gaps in the reporting and evidence-base informing remote vital sign monitoring in alternatives to admission for acute illness, despite expanding implementation in clinical practice. Although continuous vital sign monitoring using wearable devices may offer added benefit, its use in existing RCTs is limited. Recommendations for the implementation and evaluation of remote monitoring in future clinical trials are proposed.
Contact
Envelope publications at warwick dot ac dot uk
Quick links
- HEFCE’s Open Access Policy
- University of Warwick Open Access Policy
- Other Funder’s Policies
- Funding available for Open Access