Smart photogrammetry for three-dimensional shape measurement

Eastwood, Joe; Zhang, Hui; Isa, Mohammed; Sims-Waterhouse, Danny; Leach, Richard; Piano, Samanta

doi:10.1117/12.2556462

Cited by 13 publications

(9 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The idea is that a smart measurement system should be able to autonomously build the statistical model from a set of repeated measurements (which is already possible, as the model building step is fully automated) and use such model to a) construct a map of local bias and variance to serve as guidance to assess where measurement performed better and where it performed worse, in order to plan further measurement actions; b) estimate error associated to linear dimensions, and thus assess reliability of each in a part inspection scenario. The development of smart measurement systems relying on the statistical model presented in this work is currently explored as part of ongoing research work [64].…”

Section: Discussionmentioning

confidence: 99%

Statistical point cloud model to investigate measurement uncertainty in coordinate metrology

Senin

Catalucci

Moretti

et al. 2021

Precision Engineering

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Statistical point cloud model to investigate measurement uncertainty in coordinate metrology

Senin

Catalucci

Moretti

et al. 2021

Precision Engineering

View full text Add to dashboard Cite

“…Although the proposed technique is shown to be beneficial, there are still some issues that require further investigation; among them, improving the time for conducting the optimisation and investigating the effects of non-uniform lighting on visibility. These suggested improvements will form the basis for future work that we are conducting to develop an automated optical coordinate measurement pipeline [42,43].…”

Section: Discussionmentioning

confidence: 99%

Optimisation of camera positions for optical coordinate measurement based on visible point analysis

Zhang

Eastwood

Isa

et al. 2021

Precision Engineering

Self Cite

View full text Add to dashboard Cite

“…Simulated image data, specifically synthetic image dataset generation from 3D modelling or rendering software (Blender, Unity and Unreal), have shown promise for successfully training AI systems for a range of applications [ 19 , 20 ]. This approach is low-cost and easily adaptable to a range of situations relevant to different applications.…”

Section: Introductionmentioning

confidence: 99%

Training generative adversarial networks for optical property mapping using synthetic image data

Osman

Crowley

Gordon

2022

Biomed. Opt. Express

View full text Add to dashboard Cite

We demonstrate the training of a generative adversarial network (GAN) for the prediction of optical property maps (scattering and absorption) using spatial frequency domain imaging (SFDI) image data sets that are generated synthetically with a free open-source 3D modelling and rendering software, Blender. The flexibility of Blender is exploited to simulate 5 models with real-life relevance to clinical SFDI of diseased tissue: flat samples containing a single material, flat samples containing 2 materials, flat samples containing 3 materials, flat samples with spheroidal tumours and cylindrical samples with spheroidal tumours. The last case is particularly relevant as it represents wide-field imaging inside a tubular organ e.g. the gastro-intestinal tract. In all 5 scenarios we show the GAN provides an accurate reconstruction of the optical properties from single SFDI images with a mean normalised error ranging from 1.0-1.2% for absorption and 1.1%-1.2% for scattering, resulting in visually improved contrast for tumour spheroid structures. This compares favourably with the ∼10% absorption error and ∼10% scattering error achieved using GANs on experimental SFDI data. Next, we perform a bi-directional cross-validation of our synthetically-trained GAN, retrained with 90% synthetic and 10% experimental data to encourage domain transfer, with a GAN trained fully on experimental data and observe visually accurate results with an error of 6.3%-10.3% for absorption and 6.6%-11.9% for scattering. Our synthetically trained GAN is therefore highly relevant to real experimental samples but provides the significant added benefits of large training datasets, perfect ground-truths and the ability to test realistic imaging geometries, e.g. inside cylinders, for which no conventional single-shot demodulation algorithms exist. In the future, we expect that the application of techniques such as domain adaptation or training on hybrid real-synthetic datasets will create a powerful tool for fast, accurate production of optical property maps for real clinical imaging systems.

show abstract

Smart photogrammetry for three-dimensional shape measurement

Cited by 13 publications

References 0 publications

Statistical point cloud model to investigate measurement uncertainty in coordinate metrology

Statistical point cloud model to investigate measurement uncertainty in coordinate metrology

Optimisation of camera positions for optical coordinate measurement based on visible point analysis

Training generative adversarial networks for optical property mapping using synthetic image data

Contact Info

Product

Resources

About