arXiv subjects

Ali Amiri

Publications and source records attributed to Ali Amiri.

9 recordsLinked to original sources

arXiv2026

Trans-stenotic pressure drop estimation from PC-MRI and ultrasound imaging velocimetry using a modified Bernoulli equation

Accurate non-invasive estimation of trans-stenotic pressure gradients remains a challenge. In clinical practice, pressure gradients are often estimated from velocity measurements using Bernoulli-based formulas, but these simplified relations do not explicitly account for how pressure losses change with the flow regime. Here, we introduce a modified Bernoulli (MB) formulation that incorporates regime-dependent pressure losses through a Reynolds-number-dependent loss coefficient. Steady in-vitro experiments were performed in an idealized stenosis model over physiologically relevant flow rates (0.65-3.9 L/min), combining direct pressure measurements with ultrasound imaging velocimetry and phase-contrast magnetic resonance imaging (PC-MRI) to measure velocities. The MB model was calibrated from the measured pressure drops and then evaluated against the simplified Bernoulli (SB) and extended Bernoulli (EB) formulations. Over the tested flow regime, MB agreed best with the measurements. SB and EB showed larger biases, with errors of roughly 10-55% (SB) and -15 to 25% (EB), and overestimated the pressure drop in the clinically relevant range. We additionally quantified the effect of PC-MRI in-plane pixel size on MRI-based pressure estimates. Coarse sampling of the stenosis throat led to systematic underestimation of flow rate and bulk velocity and, consequently, of the MB-predicted pressure drop. In contrast, the peak throat velocity was substantially less sensitive to pixel size, resulting in smaller estimation errors when used as input for the MB. Overall, the results demonstrate that accounting for flow-regime-dependent loss mechanisms enhances pressure drop estimation, and that sufficient sampling of the stenotic throat is crucial for MRI-based flow rate and pressure drop estimation. In addition, peak-velocity-based MB pressure drop estimations are less sensitive to pixel size.

physics.flu-dyn↗

arXiv2024

Developing Convolutional Neural Networks using a Novel Lamarckian Co-Evolutionary Algorithm

Neural Architecture Search (NAS) methods autonomously discover high-accuracy neural network architectures, outperforming manually crafted ones. However, The NAS methods require high computational costs due to the high dimension search space and the need to train multiple candidate solutions. This paper introduces LCoDeepNEAT, an instantiation of Lamarckian genetic algorithms, which extends the foundational principles of the CoDeepNEAT framework. LCoDeepNEAT co-evolves CNN architectures and their respective final layer weights. The evaluation process of LCoDeepNEAT entails a single epoch of SGD, followed by the transference of the acquired final layer weights to the genetic representation of the network. In addition, it expedites the process of evolving by imposing restrictions on the architecture search space, specifically targeting architectures comprising just two fully connected layers for classification. Our method yields a notable improvement in the classification accuracy of candidate solutions throughout the evolutionary process, ranging from 2% to 5.6%. This outcome underscores the efficacy and effectiveness of integrating gradient information and evolving the last layer of candidate solutions within LCoDeepNEAT. LCoDeepNEAT is assessed across six standard image classification datasets and benchmarked against eight leading NAS methods. Results demonstrate LCoDeepNEAT's ability to swiftly discover competitive CNN architectures with fewer parameters, conserving computational resources, and achieving superior classification accuracy compared to other approaches.

cs.NE↗

arXiv2024

A Comprehensive Survey on Deep-Learning-based Vehicle Re-Identification: Models, Data Sets and Challenges

Vehicle re-identification (ReID) endeavors to associate vehicle images collected from a distributed network of cameras spanning diverse traffic environments. This task assumes paramount importance within the spectrum of vehicle-centric technologies, playing a pivotal role in deploying Intelligent Transportation Systems (ITS) and advancing smart city initiatives. Rapid advancements in deep learning have significantly propelled the evolution of vehicle ReID technologies in recent years. Consequently, undertaking a comprehensive survey of methodologies centered on deep learning for vehicle re-identification has become imperative and inescapable. This paper extensively explores deep learning techniques applied to vehicle ReID. It outlines the categorization of these methods, encompassing supervised and unsupervised approaches, delves into existing research within these categories, introduces datasets and evaluation criteria, and delineates forthcoming challenges and potential research directions. This comprehensive assessment examines the landscape of deep learning in vehicle ReID and establishes a foundation and starting point for future works. It aims to serve as a complete reference by highlighting challenges and emerging trends, fostering advancements and applications in vehicle ReID utilizing deep learning models.

cs.CV↗

arXiv2021

UGRWO-Sampling for COVID-19 dataset: A modified random walk under-sampling approach based on graphs to imbalanced data classification

This paper proposes a new RWO-Sampling (Random Walk Over-Sampling) based on graphs for imbalanced datasets. In this method, two schemes based on under-sampling and over-sampling methods are introduced to keep the proximity information robust to noises and outliers. After constructing the first graph on minority class, RWO-Sampling will be implemented on selected samples, and the rest will remain unchanged. The second graph is constructed for the majority class, and the samples in a low-density area (outliers) are removed. Finally, in the proposed method, samples of the majority class in a high-density area are selected, and the rest are eliminated. Furthermore, utilizing RWO-sampling, the boundary of minority class is increased though the outliers are not raised. This method is tested, and the number of evaluation measures is compared to previous methods on nine continuous attribute datasets with different over-sampling rates and one data set for the diagnosis of COVID-19 disease. The experimental results indicated the high efficiency and flexibility of the proposed method for the classification of imbalanced data

cs.LG↗

arXiv2020

QLMC-HD: Quasi Large Margin Classifier based on Hyperdisk

In the area of data classification, the different classifiers have been developed by their own strengths and weaknesses. Among these classifiers, we propose a method that is based on the maximum margin between two classes. One of the main challenges in this area is dealt with noisy data. In this paper, our aim is to optimize the method of large margin classifiers based on hyperdisk (LMC-HD) and combine it into a quasisupport vector data description (QSVDD) method. In the proposed method, the bounding hypersphere is calculated based on the QSVDD method. So our convex class model is more robust compared with the support vector machine (SVM) and less tight than LMC-HD. Large margin classifiers aim to maximize the margin and minimizing the risk. Since our proposed method ignores the effect of outliers and noises, so this method has the widest margin compared with other large margin classifiers. In the end, we compare our proposed method with other popular large margin classifiers by the experiments on a set of standard data which indicates our results are more efficient than the others

cs.CV↗

arXiv2019

Weighted second-order cone programming twin support vector machine for imbalanced data classification

We propose a method of using a Weighted second-order cone programming twin support vector machine (WSOCP-TWSVM) for imbalanced data classification. This method constructs a graph based under-sampling method which is utilized to remove outliers and reduce the dispensable majority samples. Then, appropriate weights are set in order to decrease the impact of samples of the majority class and increase the effect of the minority class in the optimization formula of the classifier. These weights are embedded in the optimization problem of the Second Order Cone Programming (SOCP) Twin Support Vector Machine formulations. This method is tested, and its performance is compared to previous methods on standard datasets. Results of experiments confirm the feasibility and efficiency of the proposed method.

cs.CV↗

arXiv2016

Extraction of Active Regions and Coronal Holes from EUV Images Using the Unsupervised Segmentation Method in the Bayesian Framework

The solar corona is the origin of very dynamic events that are mostly produced in active regions (AR) and coronal holes (CH). The exact location of these large-scale features can be determined by applying image-processing approaches to extreme-ultraviolet (EUV) data. We here investigate the problem of segmentation of solar EUV images into ARs, CHs, and quiet-Sun (QS) images in a firm Bayesian way. On the basis of Bayes' rule, we need to obtain both prior and likelihood models. To find the prior model of an image, we used a Potts model in non-local mode. To construct the likelihood model, we combined a mixture of a Markov-Gauss model and non-local means. After estimating labels and hyperparameters with the Gibbs estimator, cellular learning automata were employed to determine the label of each pixel. We applied the proposed method to a Solar Dynamics Observatory/ Atmospheric Imaging Assembly (SDO/AIA) dataset recorded during 2011 and found that the mean value of the filling factor of ARs is 0.032 and 0.057 for CHs. The power-law exponents of the size distribution of ARs and CHs were obtained to be -1.597 and -1.508, respectively, with the maximum likelihood estimator method. When we compare the filling factors of our method with a manual selection approach and the SPoCA algorithm, they are highly compatible.

astro-ph.SR↗

arXiv2015

An efficient parallel algorithm for computing determinant of non square matrices

One of the most significant challenges in Computing Determinant of Rectangular Matrices is high time complexity of its algorithm. Among all definitions of determinant of rectangular matrices, used definition has special features which make it more notable. But in this definition, C(n m) sub matrices of the order m*m needed to be generated that put this problem in NP hard class. On the other hand, any row or column reduction operation may hardly lead to diminish the volume of calculation. Therefore, in this paper we try to present the parallel algorithm which can decrease the time complexity of computing the determinant of non-square matrices to O(pow(n,2)).

cs.DC↗

arXiv2014

Automatic Method for Identifying Photospheric Bright Points and Granules Observed by Sunrise

In this study, we propose methods for the automatic detection of photospheric features (bright points and granules) from ultra-violet (UV) radiation, using a feature-based classifier. The methods use quiet-Sun observations at 214 nm and 525 nm images taken by Sunrise on 9 June 2009. The function of region growing and mean shift procedure are applied to segment the bright points (BPs) and granules, respectively. Zernike moments of each region are computed. The Zernike moments of BPs, granules, and other features are distinctive enough to be separated using a support vector machine (SVM) classifier. The size distribution of BPs can be fitted with a power-law slope -1.5. The peak value of granule sizes is found to be about 0.5 arcsec^2. The mean value of the filling factor of BPs is 0.01, and for granules it is 0.51. There is a critical scale for granules so that small granules with sizes smaller than 2.5 arcsec^2 cover a wide range of brightness, while the brightness of large granules approaches unity. The mean value of BP brightness fluctuations is estimated to be 1.2, while for granules it is 0.22. Mean values of the horizontal velocities of an individual BP and an individual BP within the network were found to be 1.6 km/s and 0.9 km/s, respectively. We conclude that the effect of individual BPs in releasing energy to the photosphere and maybe the upper layers is stronger than what the individual BPs release into the network.

astro-ph.SR↗