Time-Frequency Analysis of Image Based on Stockwell Transform

Choraś, Ryszard S.

doi:10.1007/978-3-319-01622-1_11

Cited by 6 publications

(9 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…, N − 1, is obtained from x(t) by sampling. By replacing τ → k and f → n N, the discrete ST for x[k], S[k, n], for n = 0 is calculated as [29]:…”

Section: A 1d Stockwell Transformmentioning

confidence: 99%

“…It is worth mentioning that the total number of points DOST results is equal to that of the input image. By integrating over all values p x , p y , a local spatial frequency domain consists of the positive and negative frequency components from information about the frequencies ( f u , f v ) in the bandwidth of 2 p x −1 × 2 p y −1 frequencies [29].…”

Section: B 2d Dostmentioning

confidence: 99%

See 1 more Smart Citation

Two-Dimensional Stockwell Transform and Deep Convolutional Neural Network for Multi-Class Diagnosis of Pathological Brain

Soleimani

Vahidi

Vaseghi

2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

Since the brain lesion detection and classification is a vital diagnosis task, in this paper, the problem of brain magnetic resonance imaging (MRI) classification is investigated. Recent advantages in machine learning and deep learning allows the researchers to develop the robust computer-aided diagnosis (CAD) tools for classification of brain lesions. Feature extraction is an essential step in any machine learning scheme. Time-frequency analysis methods provide localized information that makes them more attractive for image classification applications. Owing to the advantages of two-dimensional discrete orthonormal Stockwell transform (2D DOST), we propose to use it to extract the efficient features from brain MRIs and obtain the feature matrix. Since there are some irrelevant features, twodirectional two-dimensional principal component analysis ((2D) 2 PCA) is used to reduce the dimension of the feature matrix. Finally, convolution neural networks (CNNs) are designed and trained for MRI classification. Simulation results indicate that the proposed CAD tool outperforms the recently introduced ones and can efficiently diagnose the MRI scans.

show abstract

“…, N − 1, is obtained from x(t) by sampling. By replacing τ → k and f → n N, the discrete ST for x[k], S[k, n], for n = 0 is calculated as [29]:…”

Section: A 1d Stockwell Transformmentioning

confidence: 99%

Section: B 2d Dostmentioning

confidence: 99%

Two-Dimensional Stockwell Transform and Deep Convolutional Neural Network for Multi-Class Diagnosis of Pathological Brain

Soleimani

Vahidi

Vaseghi

2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

show abstract

“…This technique helps a signal to display frequency information over time and produces different patterns for different operating conditions. Techniques such as short-time Fourier transform [29] stransform [30,31], Wigner-Ville distribution [32], and continuous-time wavelet [33,34] are used to obtain TF representation. Continuous wavelet analysis is more effective as it represents the signal with multiple resolutions compared with other methods [35,36].…”

Section: Time-frequency Representationmentioning

confidence: 99%

A New Hybrid Diagnosis of Bearing Faults Based on Time-Frequency Images and Sparse Representation

Aydın¹,

Kaner²

2020

View full text Add to dashboard Cite

Induction motors are an essential component of many applications in industry due to their robust and simple construction. Since bearing faults are the most occurred fault type in the induction motors, it is important to implement the fault detection procedure at an early stage to prevent a sudden interruption of industrial systems. In recent years, deep learning-based techniques have become important tools for converting raw data into images and for producing high-quality images. However, deep learning-based techniques are still difficult to apply in real-time because the techniques require large training data, which slows down the learning process. In the present study, we propose a novel bearing faults diagnosis method at different operating speeds and load conditions. We obtain the time-frequency (TF) representation by applying continuous wavelet analysis to the raw vibration signals. The results of TF representation is recorded as an image. We apply co-occurrence Histograms of Oriented Gradients (coHOG) to the image to obtain features and classify the features with extreme learning machine with a sparse classifier (ELMSRC) to diagnose faults. We obtained better results in terms of time and performance compared with the proposed method of other classification and deep learning techniques.

show abstract

“…The Stockwell transform can be seen as a mix between Short-Time Fourier Transform, sometimes called Gabor Transform, and the Wavelet Transform. Due to its flexibility, the Stockwell transform has been used for image filtering, texture recognition, noise reduction, image compression and, in general, as a tool in image processing [4], [5], [6].…”

Section: The Stockwell Transformmentioning

confidence: 99%

On the application of the Stockwell transform to GPR data analysis

Riba

Piro

Battisti

et al. 2015

2015 8th International Workshop on Advanced Ground Penetrating Radar (IWAGPR)

View full text Add to dashboard Cite

For most archaeological prospections with GPR, target detection is the most important aspect of the surveys. The main efforts of GPR data processing are committed to increase the signal to noise ratio in radargrams. The usual processing aims to: filter the radar section from clutters and attenuate the in-band noises; enhance coherent reflections that are likely due to the target response; transform the acquired time section into a depth section; build a 3D image of the reflection targets in the subsoil by correlating reflections from adjacent radargrams. In this paper we show some application of the Stockwell transform, both to synthetic and field data, to enhance the signal to noise ratio in radargrams.

show abstract

Time-Frequency Analysis of Image Based on Stockwell Transform

Cited by 6 publications

References 7 publications

Two-Dimensional Stockwell Transform and Deep Convolutional Neural Network for Multi-Class Diagnosis of Pathological Brain

Two-Dimensional Stockwell Transform and Deep Convolutional Neural Network for Multi-Class Diagnosis of Pathological Brain

A New Hybrid Diagnosis of Bearing Faults Based on Time-Frequency Images and Sparse Representation

On the application of the Stockwell transform to GPR data analysis

Contact Info

Product

Resources

About