A Distributed Reflector Localization Approach to Ultrasonic Array Imaging in Non-Destructive Testing Applications

Moallemi, Nasim; Shahbazpanahi, Shahram

doi:10.1109/tsp.2014.2329276

Cited by 11 publications

(6 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table II shows the false positive rates obtained using the three models along with different types of images or classifiers. It can be seen that (1) softmax usually provided the higher false positive rate than the RF [13] classifier; (2) given that softmax was used, the lower false positive rate was produced using the z-normalised images than that generated using the original images. But this is not the case for the RF classifier; and (3) the model derived using fine-tuning always yielded the lower false positive rate than those produced by its two counterparts when softmax was used.…”

Section: A Defect Detectionmentioning

confidence: 99%

See 1 more Smart Citation

Defect Detection and Classification by Training a Generic Convolutional Neural Network Encoder

Dong

Taylor

Cootes

2020

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

Non-destructive Testing (NDT) often involves analysing images to identify (rare) defects. We propose a method for locating and classifying abnormalities using Convolutional Neural Networks (CNNs). A particular problem is that it is often difficult to get large numbers of examples of images of defects, making training a classifier challenging. To address this problem we generate large numbers of synthetic images by combining real defects with different backgrounds. These images are used to train a U-Net style network to perform defect detection at the pixel level. We also demonstrate that the encoder of the network produces features which can be applied to the defect classification task at the image level. Both the defect detection and classification modules are tested on multiple small data sets. Our results show that these modules are able to fulfil the industrial component inspection task at the pixel level (locating defect regions) and image level (identifying if an image contains a defect).

show abstract

Section: A Defect Detectionmentioning

confidence: 99%

“…N ON-DESTRUCTIVE Testing (NDT) [1], [2] is widely used in quality control and monitoring of manufactured components. Different techniques can be used for NDT purposes, including ultrasonic testing (UT) [3], radiographic testing (RT) [4], [5], penetrant testing (PT) [6] and so on.…”

Section: Introductionmentioning

confidence: 99%

Defect Detection and Classification by Training a Generic Convolutional Neural Network Encoder

Dong

Taylor

Cootes

2020

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

show abstract

“…To study such a problem, we model the interface between the two layers as a spatially distributed source. Assuming a two-dimensional model, this interface can be modeled as a line consisting of infinite number of point sources [28]. At each point on the interface, the propagating sound wave is refracted into the second layer, therefore, each point on the interface acts as a point source for all the points inside the second layer.…”

Section: Data Modelmentioning

confidence: 99%

A Novel Spatio-Temporal Frequency-Domain Imaging Technique for Two-Layer Materials Using Ultrasonic Arrays

Moallemi

Shahbazpanahi

2018

IEEE Access

Self Cite

View full text Add to dashboard Cite

Ultrasonic imaging of multi-layer materials with parallel interfaces is a challenging problem in non-destructive testing. In multi-layer materials, since the sound velocity and the propagation path change when sound travels from one layer into another, calculating the sound travel time is complicated. In this paper, we develop a frequency-domain imaging algorithm for estimating the scattering coefficients of all the points inside the second layer of a two-layer (liquid-solid) medium in order to obtain an image of the region of interest. To do so, we first introduce our data model for the array received signals by modeling the interfaces between the layers of a two-layer medium as a spatially distributed source. Then, we introduce a mapping relationship between the two-dimensional image of the region of interest and the three-dimensional Fourier transform of the received signals. This proposed algorithm has relatively lower computational complexity and it can be used for online imaging. Computer simulations as well as experimental data show the accuracy of the proposed algorithm.INDEX TERMS Distributed source, array signal processing, non-destructive testing, immersion ultrasonic imaging, multi-layer imaging.

show abstract

“…An acoustic-electric channel is formed by coaxially aligning and acoustically coupling a pair of piezoelectric transducers to both sides of the plate. Glycerin is used to couple the transducers into the medium, which is evidently used in ultrasonic nondestructive testing [8]. In [9], the influence of different coupling methods between the transducer and the metal wall on power transfer efficiency was investigated.…”

Section: Introductionmentioning

confidence: 99%

Performance Enhancement of an Ultrasonic Power Transfer System Through a Tightly Coupled Solid Media Using a KLM Model

Kar

Wallrabe

2020

Micromachines

View full text Add to dashboard Cite

Contactless ultrasonic power transmission (UPT) through a metal barrier has become an exciting field of research, as metal barriers prevent the use of electromagnetic wireless power transfer due to Faraday shielding effects. In this paper, we demonstrate power transfer through a metal wall with the use of ultrasonic waves generated from a piezoelectric transducer. Accurate characterization and modeling of the transducer and investigation of the influence of the acoustic properties of the transmitting medium are instrumental for the performance prediction and optimal design of an ultrasonic power link. In this work, we applied the KLM model for the emitting and receiving transducers, with respect to the transmitting medium and model for both the emission and reception function. A practical UPT system was built by mechanically coupling and co-axially aligning two composite transducers on opposite sides of a transmitting medium wall. The optimal transmission performance of the ultrasonic power link through thickness-stretch vibrations of the wall together with two piezoelectric transducers working in TE mode was determined. Eventually, the operating frequency and ohmic loading condition for maximum power transmission were obtained for two different media, aluminium and polyoxymethylene (POM), with contrasting specific acoustic impedances. The results showed that the measured optimal electric loads and operating frequency for maximum power transfer agreed well with the theoretical predictions.

show abstract

A Distributed Reflector Localization Approach to Ultrasonic Array Imaging in Non-Destructive Testing Applications

Cited by 11 publications

References 22 publications

Defect Detection and Classification by Training a Generic Convolutional Neural Network Encoder

Defect Detection and Classification by Training a Generic Convolutional Neural Network Encoder

A Novel Spatio-Temporal Frequency-Domain Imaging Technique for Two-Layer Materials Using Ultrasonic Arrays

Performance Enhancement of an Ultrasonic Power Transfer System Through a Tightly Coupled Solid Media Using a KLM Model

Contact Info

Product

Resources

About