Browsing by Author "Zahin, Labiba"

Now showing 1 - 2 of 2

Agricultural analysis and crop yield prediction of Habiganj using multispectral bands of satellite imagery with machine learning
(BRAC University, 2020-09) Shahrin, Fariha; Zahin, Labiba; Rahman, Ramisa; Hossain, A S M Jahir; Azad, A.K.M Abdul Malek
Bangladesh is predominately an agriculture-based country, which faces uncertain crop yields and inefficient farming infrastructure resulting in adverse effect in food security. Habiganj is selected as the study area because of its vulnerability to floods and drought due to its unique terrain. This paper aims to present a combinational agricultural mapping and monitoring of Habiganj with crop growth and yield prediction. Multi-spectral band images of Habiganj from Landsat 8 are processed and remote sensing indices are extracted. With options of K-means and Mask R-CNN methods, crop growth is evaluated using both Python and MATLAB. Then using two type of machine learning algorithms crop yield of Habiganj is predicted from its existing parameters and the datasets are predicted by using two type of time series model. Furthermore, comparative studies are concluded between two platforms and time series model to determine the most suited environment for this research purpose.
Physics-informed variational autoencoders for cosmological field reconstruction and parameter inference
(BRAC University, 2026-01) Zahin, Labiba; Tasnim, Zarin; Zaman, Tasnim; Islam, Mehrabul; Zahir, Safkat; Hossain, Muhammad Iqbal
In modern cosmology, predicting cosmological parameters is key to understanding the fundamental physical laws of the universe and dictating how cosmic structures are formed, evolve, and are observed. Parameters such as the total matter density (Ωm) and the amplitude of matter fluctuations (σ8) cannot be measured directly; instead, they have to be predicted based on complicated, high-dimensional simulation maps or observational data. Since, due to the complexity of the high-dimensional maps, traditional deep learning models often fail to give meaningful results, as they often learn shortcuts to statistical patterns that may appear correct but completely ignore the actual laws of physics. In this work, a Physics Informed Variational Autoencoder (PI-VAE) has been proposed as a combined framework for learning the compact representations of cosmological fields while directly imposing fundamental physical constraints to accurately reconstruct multi-channel cosmological fields, and directly infer key cosmological parameters from the learned latent space. By attaching a lightweight parameter regression head with VAE, the research looks into how physical information stored in the latent representation can be used for parameter predictions. To conduct this research, the CAMELS (Cosmology and Astrophysics with Machine Learning Simulations) dataset has been used, which comprises thousands of hydrodynamical simulations intended to systematically vary cosmological and astrophysical parameters. Our findings, supported by the ablation experiments, highlight the potential of physics-guided deep generative models for cosmological analysis by showing that physics-informed latent representations can simultaneously achieve meaningful cosmological parameter inference and accurate field reconstruction.