arXiv subjects

Arijit Bhattacharya

Publications and source records attributed to Arijit Bhattacharya.

3 recordsLinked to original sources

arXiv2026

GourNet: A CNN-Based Model for Mango Leaf Disease Detection

Mango cultivation is crucial in the agricultural sector, significantly contributing to economic development and food security. However, diseases affecting mango leaves can significantly reduce both the production and overall fruit grade. Detecting leaf diseases at an early stage with precision is key to effective disease prevention and sustaining crop productivity. In this paper, we introduce a "deep learning" model named "GourNet", which leverages "Convolutional Neural Networks" to identify infections in mango leaves. We utilize the "MangoLeafBD" (MBD) dataset to train and assess the effectiveness of the presented model. The MBD dataset contains seven disease classes and a Healthy class, making a total of eight classes. To enhance model performance, the images are preprocessed through steps like resizing, rescaling, and data augmentation prior to training. To properly evaluate the model, the dataset is separated into 80% for training, with the remaining 20% equally split between validation and testing. Our model uses only 683,656 total parameters and achieves a classification accuracy of 97%. This research's source code can be found at: https://github.com/ekramalam/GourNet-Repo.

cs.CV↗

arXiv2025

Data-driven stability analysis in a multi-element supercritical Liquid Oxygen-methane combustor

Thermoacoustic instability (TAI) is a pressing problem in rocket combustors. TAI can cause significant damage to a combustor, resulting in mission failure. Therefore, stability analysis is crucial during the design and development phases of a rocket combustor. Stability analysis during the design phase can be substantially aided by the rocket combustor's large eddy simulation (LES). However, the computational cost of LES for full-scale rocket combustors is high. Therefore, using a small set of data from a large eddy simulation of a multi-element full-scale combustor, we investigated the effectiveness and computational needs of many data-driven and physics-driven tools for the classification of the stable and unstable regimes in the current study. Recurrence network analysis (RNA), reservoir computing (RC), and multi-scale permutation entropy (MPEA) analysis are the instruments employed in this study. The regime categorization task is unsuitable for RNA and MPEA, according to the results. With little input data, RC-based metrics may map the stable and unstable regimes and are thought to be computationally inexpensive and straightforward to use. In order to help with the design and development of rocket combustors, the combined LES-RC method to stability analysis is therefore anticipated to result in a notable decrease in processing needs.

physics.flu-dyn↗

arXiv2025

Insights of Transitions to Thermoacoustic Instability in Inverse Diffusion Flame using Multifractal Detrended Fluctuation Analysis

The inverse diffusion flame (IDF) can experience thermoacoustic instability due to variations in power input or flow conditions. However, the dynamical transitions in IDF that lead to this instability when altering control parameters have not been thoroughly investigated. In this study, we explore the control parameters through two different approaches and employ multifractal detrended fluctuation analysis to characterize the transitions observed prior to the onset of thermoacoustic instability in the inverse diffusion flame. Our findings reveal a loss of multifractality near the region associated with thermoacoustic instability, which suggests a more ordered behavior. We determine that the singularity exponent, the width of the multifractal spectrum, and the Hurst exponent are reliable indicators of thermoacoustic instability and serve as effective classifiers of dynamical states in inverse diffusion flames.

physics.flu-dyn↗