Find the paper and code here. 

Seismic monitoring in hazardous environments, such as active volcanic regions, presents significant challenges due to restricted accessibility. Additionally, remote sensing methods are often compromised by atmospheric turbulence, which distorts measurements and reduces accuracy.

Our Moiré-based system, enhanced with Active Convolved Illumination (ACI), leverages the displacement-magnifying properties of Moiré patterns to detect subtle ground movements with high precision.

This cost-effective, non-invasive, and scalable solution enhances remote seismic sensing, with broad applications in volcanology, geophysics, structural analysis, and metrology, enabling high-precision displacement measurements in extreme conditions. 

Find the paper and code here. 

Atmospheric turbulence significantly distorts optical wave propagation, posing major challenges for imaging and remote sensing applications. Traditional mitigation strategies include adaptive optics, post-processing deconvolution techniques, and machine learning-based restoration methods. Among these, ACI has shown promise in preserving spatial coherence and improving remote sensing performance. Concurrently, deep learning (DL) has emerged as a powerful tool for turbulence correction, to enhance image restoration.

In this work, we propose a novel hybrid approach that integrates ACI with a DL-based restoration model to more effectively compensate for atmospheric distortions. We first train a CNN on diverse noisy datasets before adapting it to atmospheric imaging. To evaluate our method, we conduct numerical simulations of turbulence effects and compare the performance of ACI alone, DL-based restoration alone, and their combined application. Our results demonstrate that integrating ACI with deep learning significantly enhances image fidelity, surpassing existing standalone techniques. This study highlights the potential of merging active optical compensation with AI-driven restoration for robust and scalable turbulence correction in imaging and remote sensing.

Find the paper and code here. 

Our Moiré-based system leverages the displacement-magnifying properties of Moiré patterns to precisely detect subtle ground movements in environments where seismic monitoring is challenging due to restricted accessibility, such as active volcanic regions.

In this work, we present the experimental setup of the Moiré apparatus and remote sensing techniques, discuss its design and implementation, and evaluate its challenges and limitations. This method improves remote seismic monitoring and has wide-ranging applications in volcanology, geophysics, structural analysis, and metrology, allowing for highly accurate displacement measurements even in challenging environments.  

Find the paper and code here. 

Imaging through the atmosphere is significantly affected by random perturbations, making it challenging to model atmospheric turbulence accurately. This paper provides a comprehensive framework for simulating atmospheric turbulence in imaging systems, detailing the necessary steps, considerations, and assumptions. We present a thorough analysis of atmospheric parameters, covering both vertical and horizontal path propagation under varying and constant refractive index structure parameters.

Our study examines air refractive index fluctuations in relation to the turbulence model, utilizing phase power spectral density to characterize distortions. Additionally, we address constraints imposed by finite simulation planes and analyze the conditions for an admissible point source to retrieve impulse responses effectively. The sampling process for vacuum and atmospheric propagation is discussed, with numerical simulations illustrating the impact of turbulence on a coherent point source propagating over a long horizontal path. This work serves as a foundational guide for modeling atmospheric imaging, offering valuable insights into turbulence-induced distortions and simulation methodologies.

Find the paper and code here. 

Objective: Designed a machine learning model to facilitate the selection of valuable stocks by leveraging historical data and advanced algorithms. Conducted extensive data preprocessing, feature engineering, and model training to achieve high accuracy and robustness. Validated the model using various metrics and evaluated its performance against other models and benchmarks.

Utilized the Long Short-Term Memory (LSTM) and Autoregressive Integrated Moving Average (ARIMA) algorithms to predict stock prices with high accuracy and consistency.

Find the code here. 

Objective: Designed a machine learning model to facilitate the selection of valuable stocks by leveraging historical data and advanced algorithms. Conducted extensive data preprocessing, feature engineering, and model training to achieve high accuracy and robustness. Validated the model using various metrics and evaluated its performance against other models and benchmarks.

Utilized the Long Short-Term Memory (LSTM) and Autoregressive Integrated Moving Average (ARIMA) algorithms to predict stock prices with high accuracy and consistency.

Find the code here. 

Objective: Investigated the use of metamaterials to enhance the characteristics of patch antennas, which are widely used in wireless communication systems. Developed a novel methodology to improve the performance of patch antennas by incorporating metamaterials into their structures. Conducted experiments to validate the effectiveness of the methodology and analyzed the results to gain insights into the underlying mechanisms.

Objective: Gained hands-on experience in fabricating solar cells using microfabrication techniques in a state-of-the-art laboratory. Received training on the various steps involved in the process, including photolithography, photoresist deposition, wafer development, and etching.

Objective: Reproduced and validated the results of a research paper using COMSOL Multiphysics to simulate the nanofocusing of circularly polarized Bessel-type plasmon polaritons. Gained expertise in numerical modeling, boundary conditions setup, and electromagnetic field analysis within hyperbolic metamaterials.