Work place: Electrical and Electronic Engineering Department, Islamic Azad University, Tehran South Branch, Tehran, Iran
Research Interests: Signal Processing, Graph and Image Processing, Image Processing
Sedigheh Ghofrani was born 1968 in Ghochan, recieved BSc degree in Electronic Engineering from Tehran University, Iran, in 1991, the MSc. Degree in communication from Islamic Azad University, South Tehran Branch, Iran, in 1997 and Ph.D. in Electronic from Iran University of Science and Technology, in 2004. She has been the assistant professor of Electronic and Electrical Engineering Department at the Islamic Azad University, South Tehran Branch from 2004 to 2011 and associate professor since 2012. Her area of research includes image processing and signal processing. In 2003, she spent eight months at the School of Electronic and Electrical Engineering, the University of Leeds, UK, supported by British Council foundation. In 2012, she spent eight months at the Center for Advanced Communications (CAC) at Villanova University, PA, USA, as visiting research professor.
DOI: https://doi.org/10.5815/ijigsp.2019.11.04, Pub. Date: 8 Nov. 2019
Removing the brain part, as the epilepsy source attack, is a surgery solution for those patients who have drug resistant epilepsy. So, the epilepsy localization area is an essential step before brain surgery. The Electroencephalogram (EEG) signals of these areas are different and called as focal (F) whereas the EEG signals of other normal areas are known as non-focal (NF). Visual inspection of multi-channels for F EEG detection is time-consuming along with human error. In this paper, an automatic and adaptive method is proposed based on second order difference plot (SODP) of EEG rhythms in empirical wavelet transform (EWT) domain as an adaptive signal decomposition. SODP provides the data variability rate or gives a 2D projection for rhythms. The feature vector is obtained using the central tendency measure (CTM). Finally, significant features, chosen by Kruskal–Wallis statistical test, are fed to K nearest neighbor (KNN) and support vector machine (SVM) classifiers. The achieved results of the proposed method in terms of three objective criteria are compared with state-of-the-art papers demonstrating an outstanding algorithm here in.[...] Read more.
DOI: https://doi.org/10.5815/ijigsp.2018.12.02, Pub. Date: 8 Dec. 2018
Now a days application of sparse representation are widely spreading in many fields such as face recognition. For this usage, defining a dictionary and choosing a proper recovery algorithm plays an important role for the method accuracy. In this paper, two type of dictionaries based on input face images, the method named SRC, and input extracted features, the method named MKD-SRC, are constructed. SRC fails for partial face recognition whereas MKD-SRC overcomes the problem. Three extension of MKD-SRC are introduced and their performance for comparison are presented. For recommending proper recovery algorithm, in this paper, we focus on three greedy algorithms, called MP, OMP, CoSaMP and another called Homotopy. Three standard data sets named AR, Extended Yale-B and Essex University are used to asses which recovery algorithm has an efficient response for proposed methods. The preferred recovery algorithm was chosen based on achieved accuracy and run time.[...] Read more.
DOI: https://doi.org/10.5815/ijigsp.2016.10.07, Pub. Date: 8 Oct. 2016
The main contribution of this paper is using compressive sensing (CS) theory for crypto steganography system to increase both the security and the capacity and preserve the cover image imperceptibility. For CS implementation, the discrete Cosine transform (DCT) as sparse domain and random sensing matrix as measurement domain are used. We consider 7 MRI images as the secret and 7 gray scale test images as cover. In addition, three sampling rates for CS are used. The performance of seven CS recovery algorithms in terms of image imperceptibility, achieved peak signal to noise ratio (PSNR), and the computation time are compared with other references. We showed that the proposed crypto steganography system based on CS works properly even though the secret image size is greater than the cover image.[...] Read more.
DOI: https://doi.org/10.5815/ijigsp.2015.07.01, Pub. Date: 8 Jun. 2015
In this paper a new application-oriented method for automatic Farsi license plate detection (ALPD), based on morphology and a modified edge clustering algorithm is proposed. Access control (AC), law enforcement (LE), and road patrol (RP) are mainly three applications for ALPD. After image enhancement by preprocessing, the edge statistics analysis and the morphology filter are used to decrease the search regions and remove the unwanted edges. Then the expectation-maximization (E-M) algorithm is used to estimate the corresponding Gaussian components for edges of remained regions. In this way the results of edge clustering and Gaussian components estimation are deterministic whereas the processing time in comparison with similar approaches in literature, is decreased significantly. Candidate regions are obtained by applying application-oriented thresholds to the properties of estimated Gaussian components. Finally for each candidate region, the maximally stable extremal region (MSER) detector is used to detect character-like regions and then select the region(s) of interest containing license plates. The proposed method is evaluated by using a database which includes images for the three groups AC, LE and RP applications, whereas some images suffer of being low quality, low contrast and blur and some images have complex background through existing multiple license plates. The experimental results show that our proposed method is reliable for images of different quality and illumination condition and it is able to detect the rotated and skewed license plates even in images containing multiple license plates and complex backgrounds.[...] Read more.
DOI: https://doi.org/10.5815/ijigsp.2015.02.03, Pub. Date: 8 Jan. 2015
Ultrasound images suffer of multiplicative noise named speckle. Bayesian shrinkage in transform domain is a well-known method based on finding threshold value to suppress the speckle noise. The main problem of applying Bayesian shrinkage is finding the optimum threshold value in appropriate transform domain. In this paper, we compare the performance of adaptive Bayesian thresholding when nonsubsampled Wavelet, Contourlet and Shearlet transforms are used. We processed two synthetic test images and three original ultrasound images as well to demonstrate the efficiency of the designed filters. In order to compare the performance of Bayesian shrinkage when employing the three mentioned transform domain, we used peak signal to noise ratio (PSNR), mean square error (MSE), and structural similarity (SSIM) as the full-reference (FR) objective criteria parameters and noise variance (NV), mean square difference (MSD), and equivalent number of looks (ENL) as the no-reference (NR) objective criteria parameters.[...] Read more.
Subscribe to receive issue release notifications and newsletters from MECS Press journals