arXiv subjects

Muhammad Tahir

Publications and source records attributed to Muhammad Tahir.

29 records · Page 2Linked to original sources

arXiv2022

A Deep-Unfolded Spatiotemporal RPCA Network For L+S Decomposition

Low-rank and sparse decomposition based methods find their use in many applications involving background modeling such as clutter suppression and object tracking. While Robust Principal Component Analysis (RPCA) has achieved great success in performing this task, it can take hundreds of iterations to converge and its performance decreases in the presence of different phenomena such as occlusion, jitter and fast motion. The recently proposed deep unfolded networks, on the other hand, have demonstrated better accuracy and improved convergence over both their iterative equivalents as well as over other neural network architectures. In this work, we propose a novel deep unfolded spatiotemporal RPCA (DUST-RPCA) network, which explicitly takes advantage of the spatial and temporal continuity in the low-rank component. Our experimental results on the moving MNIST dataset indicate that DUST-RPCA gives better accuracy when compared with the existing state of the art deep unfolded RPCA networks.

cs.CV↗

arXiv2022

Current-induced quasiparticle magnetic multipole moments

Magnetic ordering beyond the standard dipolar order has attracted significant attention in recent years, but it remains an open question how to effectively manipulate such nontrivial order parameters using external perturbations. In this context, we present a theory for Cartesian magnetic multipole moments and their currents created by electric currents based on a general gauge-invariant formula for arbitrary-order spin magnetic multipole moments of Bloch wave packets. As a concrete example, we point out that the low-energy quasiparticles in phosphorene subject to a perpendicular electric field have a valley structure that hosts magnetic octupole moments. The quasiparticle magnetic octupole moments can be exhibited by an in-plane electric current and lead to accumulation of staggered spin densities at the corners of a rectangular-shaped sample. A current carrying the octupole moments can further be induced through nonlinear response. Our work paves the way to systematically searching for and utilizing quasiparticles with higher-order magnetic multipole moments in crystal materials.

cond-mat.mtrl-sci↗

arXiv2022

Variational-Based Nonlinear Bayesian Filtering with Biased Observations

State estimation of dynamical systems is crucial for providing new decision-making and system automation information in different applications. However, the assumptions on the standard computational models for sensor measurements can be violated in practice due to different types of data abnormalities such as outliers and biases. In this work, we focus on the occurrence of measurement biases and propose a robust filter for their detection and mitigation during state estimation of nonlinear dynamical systems. We model the presence of bias in each dimension within the generative structure of the state-space models. Subsequently, employing the theory of Variational Bayes and general Gaussian filtering, we devise a recursive filter which we call the Bias Detecting and Mitigating (BDM) filter. As the error detection mechanism is embedded within the filter structure its dependence on any external detector is obviated. Simulations verify the performance gains of the proposed BDM filter compared to similar Kalman filtering-based approaches in terms of robustness to temporary and persistent bias presence.

eess.SY↗

arXiv2022

A Survey on Image Aesthetic Assessment

Automatic image aesthetics assessment is a computer vision problem dealing with categorizing images into different aesthetic levels. The categorization is usually done by analyzing an input image and computing some measure of the degree to which the image adheres to the fundamental principles of photography such as balance, rhythm, harmony, contrast, unity, look, feel, tone and texture. Due to its diverse applications in many areas, automatic image aesthetic assessment has gained significant research attention in recent years. This article presents a review of the contemporary automatic image aesthetics assessment techniques. Many traditional hand-crafted and deep learning-based approaches are reviewed, and critical problem aspects are discussed, including why some features or models perform better than others and the limitations. A comparison of the quantitative results of different methods is also provided.

cs.CV↗

arXiv2022

Outlier-Robust Filtering For Nonlinear Systems With Selective Observations Rejection

Considering a common case where measurements are obtained from independent sensors, we present a novel outlier-robust filter for nonlinear dynamical systems in this work. The proposed method is devised by modifying the measurement model and subsequently using the theory of Variational Bayes and general Gaussian filtering. We treat the measurement outliers independently for independent observations leading to selective rejection of the corrupted data during inference. By carrying out simulations for variable number of sensors we verify that an implementation of the proposed filter is computationally more efficient as compared to the proposed modifications of similar baseline methods still yielding similar estimation quality. In addition, experimentation results for various real-time indoor localization scenarios using Ultra-wide Band (UWB) sensors demonstrate the practical utility of the proposed method.

eess.SY↗

arXiv2021

Deep localization of protein structures in fluorescence microscopy images

Accurate localization of proteins from fluorescence microscopy images is challenging due to the inter-class similarities and intra-class disparities introducing grave concerns in addressing multi-class classification problems. Conventional machine learning-based image prediction pipelines rely heavily on pre-processing such as normalization and segmentation followed by hand-crafted feature extraction to identify useful, informative, and application-specific features. Here, we demonstrate that deep learning-based pipelines can effectively classify protein images from different datasets. We propose an end-to-end Protein Localization Convolutional Neural Network (PLCNN) that classifies protein images more accurately and reliably. PLCNN processes raw imagery without involving any pre-processing steps and produces outputs without any customization or parameter adjustment for a particular dataset. Experimental analysis is performed on five benchmark datasets. PLCNN consistently outperformed the existing state-of-the-art approaches from traditional machine learning and deep architectures. This study highlights the importance of deep learning for the analysis of fluorescence microscopy protein imagery. The proposed deep pipeline can better guide drug designing procedures in the pharmaceutical industry and open new avenues for researchers in computational biology and bioinformatics.

cs.CV↗

arXiv2021

A Systematic Review of Online Exams Solutions in E-learning: Techniques, Tools and Global Adoption

E-learning in higher education is exponentially increased during the past decade due to its inevitable benefits in critical situations like natural disasters, and pandemic. The reliable, fair, and seamless execution of online exams in E-learning is highly significant. Particularly, online exams are conducted on E-learning platforms without the physical presence of students and instructors at the same place. This poses several issues like integrity and security during online exams. To address such issues, researchers frequently proposed different techniques and tools. However, a study summarizing and analyzing latest developments, particularly in the area of online examination, is hard to find in the literature. In this article, an SLR for online examination is performed to select and analyze 53 studies published during the last five years. Subsequently, five leading online exams features targeted in the selected studies are identified and underlying development approaches for the implementation of online exams solutions are explored. Furthermore, 16 important techniques and 11 datasets are presented. In addition, 21 online exams tools proposed in the selected studies are identified. Additionally, 25 leading existing tools used in the selected studies are also presented. Finally, the participation of countries in online exam research is investigated. Key factors for the global adoption of online exams are identified and investigated. This facilitates the selection of right online exam system for a particular country on the basis of existing E-learning infrastructure and overall cost. To conclude, the findings of this article provide a solid platform for the researchers and practitioners of the domain to select appropriate features along with underlying development approaches, tools and techniques for the implementation of a particular online exams solution as per given requirements.

cs.CY↗

arXiv2020

Identification and Quantification of Aerosol Hot-spots over Lahore Region using MODIS Data

The increased concentration of aerosols in the air caused by ever-rising urbanization and the development of various industries has horrendous consequences on human health, environment and climate. The first step to counter adverse effects of air pollution in any region is to identify locations with a high concentration of aerosols, termed aerosol hot-spots. We specifically focus on the region of Lahore, Pakistan, a city that has consistently been ranked among the top ten most polluted cities in the world. In order to identify aerosol hot-spots in the city, we utilize two-year (2017-18) Aerosol Optical Thickness (AOT) data from MODerate resolution Imaging Spectroradiometer (MODIS) on the Aqua satellite. We propose a method based on the Glowworm Swarm Optimization (GSO) algorithm that discovers several aerosol hot-spots over Lahore comprising of major highways and industrial areas. Furthermore, we formulate two quantification metrics to gauge the amount of aerosol content over each hot-spot. We also analyze the temporal variation of aerosol content that suggests the addition or suppression of pollution sources in the hot-spots. Unlike previous studies, our work provides novel insight into the regional aerosol concentration of Lahore and calls for maintenance of air quality of the regions falling under the identified aerosol hot-spots.

eess.SP↗

arXiv2016

Proof of bijection for combinatorial number system

Combinatorial number system represents a non-negative natural numbers as sum of binomial coefficients. This paper presents an induction proof that there exists unique representation of every non-negative natural number $m$ as sum of $r$ binomial coefficients.

math.CO↗

arXiv2015

Electrically Engineered Band Gap in Two-Dimensional Ge, Sn, and Pb: A First-Principles and Tight-Binding Approach

First-principles calculations were performed to investigate the electronic structure of two-dimensional (2-D) Ge, Sn, and Pb without and with the presence of an external electric field in combination with spin-orbit coupling. Tight-binding calculations based on four orbitals per atom and an effective single orbital are presented to match with the results obtained from first-principles calculations. In particular, the electronic band structure and the band splitting are investigated with both models. Moreover, the simple $k\cdot p$ model is also considered in order to understand the band splitting in the presence of an external electric field and spin-orbit coupling. A large splitting is obtained, which is expected to be useful for spintronic devices. The fair agreement between the first-principle, $k\cdot p$ model, and tight-binding approaches leads to a table of parameters for future tight-binding studies on hexagonal 2-D nanostructures. By using the tight binding parameters, the transport properties of typical 0-D triangular quantum dots between two semi-infinite electrodes in the presence of spin-orbit coupling are addressed.

cond-mat.mtrl-sci↗

arXiv2011

Time-dependent transport via a quantum shuttle

We present a theoretical study of time-dependent transport via a quantum shuttle within the non-equilibrium Green's function technique. An arbitrary voltage is applied to the tunnel junction and electrons in the leads are considered to be at zero temperature. The transient and the steady state behavior of the system is considered here in order to explore the quantum dynamics of the shuttle device as a function of time and applied bias. The properties of the phonon distribution of the oscillating dot coupled to the electrons are investigated using a non-perturbative approach. We derive a relation for the oscillator momentum charge density correlation function which is an interesting physical example for the visualization of shuttling phenomenon. We consider the crossover between the tunneling and shuttling regimes for different values of the key parameters as a function of applied bias and time. We also consider the energy transferred from the electrons to the oscillating dot as a function of time. This will provide useful insight for the design of experiments aimed at studying the quantum behavior of a shuttling device.

cond-mat.mes-hall↗