Toward photoelastic sensors: a hybrid proposal for imaging the stress field through load stepping methods

León, Juán; Branch, John W.; Restrepo-M., Alejandro

doi:10.1364/cosi.2020.cth3c.4

Cited by 8 publications

(4 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the lack of public collections of multipolarized digital photoelasticity images and the complex configuration and cost of polariscopes, it has been necessary to move towards hybrid approaches where the images are derived from computational simulations but include experimental data. 17 Among the analytical image generation methodology, it was possible to find the Isochromatic-Art dataset 18 , a collection of over 100,000 isochromatic (single polarization state) images and their corresponding greyscale images of 224 x 224 pixel stress maps for classical digital photoelasticity models. This collection was also used to train the StressNet and VGG networks, which will be adapted and evaluated in this study.…”

Section: Analytical Images Generationmentioning

confidence: 99%

“…In the case of isochromatic imaging (with a single polarization state), one-shot strategies based on hybrid load stepping 3 , regularized phase tracking 4 and LUT comparison 5 have been used. However, there are still difficulties in quantifying the strain information when the strain geometries produce complex stripe distributions, when the strain magnitudes cause high stripe densities and color overmodulation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Photoelastic and Stokes images through deep convolutional neural networks: a comparison of stress fields

Eusse-Naranjo,

Restrepo-Martínez

2023

Optical Measurement Systems for Industrial Inspection XIII

View full text Add to dashboard Cite

Photoelasticity is a non-destructive optical testing technique that focuses on stress analysis. Traditional methods of demodulating stress fields are limited by various conditions, such as the image acquisition set, material properties, load values, light sources and isoclinics. As an alternative, deep convolutional neural networks (DCNNs) have been used to recover stress fields in automated and predictive methods. In this study, different DCNNs architectures are trained by means of two datasets, each one with 45000 images. First dataset has images with four polarization states (0°, 45°, 90° and 135°). Second dataset has images with 3-channel, each one corresponding to a Stokes parameter (s0, s1, s2). The quality of predicted images is evaluated with quality metrics such as MSE, SSIM, and PSNR. MSE and Adam are used as loss function and optimizer, respectively. Results show that on average, the use of DCNNs with images with four polarization states achieve better quality metrics than DCNNs with Stokes images. These results indicate that it is possible to obtain real-time stress fields using different representations of polarized images in deep networks and opens new opportunities for representing polarized images in deep learning models and extending its applications of stress analysis.

show abstract

Section: Analytical Images Generationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Photoelastic and Stokes images through deep convolutional neural networks: a comparison of stress fields

Eusse-Naranjo,

Restrepo-Martínez

2023

Optical Measurement Systems for Industrial Inspection XIII

View full text Add to dashboard Cite

show abstract

“…It is expensive to obtain enough photoelastic fringe patterns and corresponding stress field images through the photoelastic experiment, which is due to the complicated experimental environment configuration and tedious post-data processing. Briñez-de León et al (2020d) propose a hybrid scheme that includes real experimental data and computational simulation. Different types of light sources, the range of loading external forces, rotation angles of optical elements, and various types of camera sensors are fully considered in this method.…”

Section: Photoelasticity and The Neural Network Modelmentioning

confidence: 99%

Photoelastic Stress Field Recovery Using Deep Convolutional Neural Network

Tao

Wang

Qian

et al. 2022

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

Recent work has shown that deep convolutional neural network is capable of solving inverse problems in computational imaging, and recovering the stress field of the loaded object from the photoelastic fringe pattern can also be regarded as an inverse problem solving process. However, the formation of the fringe pattern is affected by the geometry of the specimen and experimental configuration. When the loaded object produces complex fringe distribution, the traditional stress analysis methods still face difficulty in unwrapping. In this study, a deep convolutional neural network based on the encoder–decoder structure is proposed, which can accurately decode stress distribution information from complex photoelastic fringe images generated under different experimental configurations. The proposed method is validated on a synthetic dataset, and the quality of stress distribution images generated by the network model is evaluated using mean squared error (MSE), structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR), and other evaluation indexes. The results show that the proposed stress recovery network can achieve an average performance of more than 0.99 on the SSIM.

show abstract

“…All these facts limit a wide use of this phase-shifting method. This paper has as contribution the development of an algorithm that uses the structure of a standard load-stepping method [14][15][16][17] but incorporates digital processing for generating a special color image that allows to produce phase shifting, avoiding the problems associated with the limitations of the load, the number of images, movement, or deformation of the sample. Thus, this article proposes a new three-wavelength chromatic-corrected image hybrid phase shift method that works for single-camera digital photoelasticity applications.…”

Section: Introductionmentioning

confidence: 99%

A new three-wavelength chromatic-corrected hybrid phase shifting method for single-camera digital photoelasticity

Restrepo-Martínez

2023

Modeling Aspects in Optical Metrology IX

View full text Add to dashboard Cite

Digital photoelasticity is a non-contact inspection technique, that requires new strategies to unwrap the stresses map based on color images. Therefore, this paper presents a new three-wavelength chromatic-corrected hybrid phase-shift method for single-camera digital photoelasticity applications. For the experiments, the intensities of isochromatic images are simulated considering a birefringent sample compressed and a circular polariscope configured to produce a bright field image that avoids the effects caused by isoclinics. To increase the range of the fringe order, three LEDs with peaks of close wavelengths were used. Additionality, for each LED a RGB color image is simulated. The red channel of each image is used to generate a new synthetic chromatic-corrected image (CCI) thereby: the red channel of λ3 is the red channel of the new CCI, and the green and blue channels of the CCI use the respective red channels of λ2 and λ1; additionally, the wavelengths must satisfy the following condition (λ3 > λ2 >λ1). An inverted image of the CCI is computed. Thus, with the CCI and its inverted image, six images are stored, and with these images and some trigonometric relations proposed by Ekman and Nurse a wrapped phase map is extracted. Finally, an unwrapping algorithm is applied to reconstruct the stresses map. The results show that the method improves the detected maximum order and reduces stress map distortions compared to similar color phase shifting approaches. Furthermore, since the algorithm requires only a camera and a circular polariscope setup, it can be implemented in dynamic experimental applications.

show abstract

Toward photoelastic sensors: a hybrid proposal for imaging the stress field through load stepping methods

Abstract: Extending photoelasticity studies to industrial applications, such as photoelastic sensors, implies to overcome limitations like need of experts, experimental over-calibration, or reduced number of experiments. This paper introduces a computational hybrid technique toward this goal.

Cited by 8 publications

References 4 publications

Photoelastic and Stokes images through deep convolutional neural networks: a comparison of stress fields

Photoelastic and Stokes images through deep convolutional neural networks: a comparison of stress fields

Photoelastic Stress Field Recovery Using Deep Convolutional Neural Network

A new three-wavelength chromatic-corrected hybrid phase shifting method for single-camera digital photoelasticity

Contact Info

Product

Resources

About