Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
751
result(s) for
"Sharma, Richa"
Sort by:
Herbify: an ensemble deep learning framework integrating convolutional neural networks and vision transformers for precise herb identification
Herbs have historically been central to medicinal practices, representing one of the earliest forms of therapeutic intervention. While synthetic drugs are often highly effective in treating acute conditions, their use is frequently accompanied by adverse side effects. In addition, the growing dependence on synthetic pharmaceuticals has raised concerns regarding affordability, thereby fostering a renewed interest in herbal medicine as a cost-effective and holistic alternative. In response to this need, the current study introduces a computer vision framework for accurate herb identification. A novel dataset,
Herbify
, was compiled from two different herb datasets and refined through rigorous cleaning, preprocessing, and quality control procedures. The resulting dataset underwent standardization via the
Preprocessing Algorithm for Herb Detection
(PAHD), producing a refined dataset of 6104 images, representing 91 distinct herb species, with an average of about 67 images per species. Utilizing transfer learning, the research harnessed pre-trained Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), then integrated these models into an ensemble framework that leverages the unique strengths of each architecture. Experimental results indicate that EfficientNet v2-Large achieved a noteworthy F₁-score of 99.13%, while the ensemble of EfficientNet v2-Large and ViT-Large/16, termed
EfficientL-ViTL
, attained an even higher F₁-score of 99.56%. Additionally, the research also introduces ‘
Herbify
’ application, an AI-driven framework designed to identify herbs using the developed model. By directly tackling the principal obstacles in herb identification, the proposed system achieves a highly accurate and operationally viable classification mechanism. The experimental outcomes showcase top-tier performance in herb identification and emphasize the transformative potential of AI-based solutions in supporting botanical applications.
Journal Article
How to tackle complexity in urban climate resilience? Negotiating climate science, adaptation and multi-level governance in India
As the world’s population is expected to be over 2/3 rd urban by 2050, climate action in cities is a growing area of interest in the inter-disciplines of development policy, disaster mitigation and environmental governance. The climate impacts are expected to be quite severe in the developing world, given its urban societies are densely packed, vastly exposed to natural elements while possessing limited capabilities. There is a notable ambiguity and complexity that inhibits a methodical approach in identifying urban resilience measures. The complexity is due to intersection of large number of distinct variables in climate geoscience (precipitation and temperature anomalies at different locations, RCPs, timeline), adaptation alternatives (approach, priority, intervention level) and urban governance (functional mandate, institutional capacity, and plans & policies). This research examines how disparate and complex knowledge and information in these inter-disciplines can be processed for systematic ‘negotiation’ to situate, ground and operationalize resilience in cities. With India as a case, we test this by simulating mid-term and long-run climate scenarios (2050 & 2080) to map regional climate impacts that shows escalation in the intensity of climate events like heat waves, urban flooding, landslides and sea level rise. We draw on suitable adaptation measures for five key urban sectors- water, infrastructure (including energy), building, urban planning, health and conclude a sleuth of climate resilience building measures for policy application through national/ state policies, local urban plans and preparation of city resilience strategy, as well as advance the research on ‘negotiated resilience’ in urban areas
Journal Article
Maintenance of cytoplasmic and membrane densities shapes cellular geometry in Escherichia coli
Microbes precisely control their composition and geometry across diverse growth conditions, yet the mechanisms coordinating these processes remain unclear. Here, we integrate quantitative proteomics, microscopy, and biochemical measurements to reveal a biophysical principle linking these properties in
Escherichia coli
: cytoplasmic and membrane protein densities maintain a tightly conserved ratio across growth conditions, while the periplasmic density varies. Building on this observation, we develop a mathematical model demonstrating that maintaining this density ratio constrains the surface-to-volume ratio as a nonlinear function of proteome composition, specifically the ribosomal proteome fraction and partitioning between cellular compartments. The model holds under guanosine tetraphosphate perturbations that alter ribosome levels, further demonstrating that cellular geometry is not strictly determined by growth rate. These findings provide a biophysical framework for geometry control, underscoring density maintenance as a key physiological constraint that shapes cellular phenotypes.
Chure et al. analyse experimental data to show that E. coli bacteria maintain stable protein density ratios between cytoplasm and membranes. In addition, they develop a biophysical model that predicts surface-to-volume ratio from ribosomal content and protein partitioning across cell compartments.
Journal Article
Water quality prediction using Machine Learning Models
The quality of water is a vital determinant of environmental sustainability, economic development, and general welfare. India has substantial water quality issues, with different areas facing varying levels of pollution. Industrial effluents introduce toxic chemicals and heavy metals into water bodies, while agricultural runoff carries pesticides, fertilizers, and sediments, causing eutrophication and water pollution. The Ganges, Yamuna, and Godavari rivers have elevated amounts of pollution. According to the Central Pollution Control Board, the levels of biochemical oxygen demand, which is a measure of organic pollution, often above the acceptable thresholds in many sections of these rivers. Conventional techniques for monitoring water quality are often arduous, time-consuming, and incapable of delivering real- time evaluations. The objective of this study is to create a precise classification model that can accurately forecast water quality by using a range of indicators. The aim is to use machine learning techniques, including decision trees, K-Nearest Neighbor (KNN), and Random Forest, to develop prediction models that can effectively assess water quality and identify possible pollution incidents before they become major issues. This research used a comprehensive dataset of water quality metrics, including pH, turbidity, dissolved oxygen, temperature, phosphates, and nitrates, to assess the accuracy of each algorithm in forecasting water potability. The Random Forest method attained a superior accuracy of 70.4%, successfully handling intricate interactions and mitigating overfitting by using ensemble learning. The KNN method, which achieved an accuracy of 59%, had challenges arising from its susceptibility to the selection of k and distance measures, as well as processing inefficiencies. The Decision Tree approach, despite its speed and interpretability, had the lowest accuracy of 58% mostly owing to overfitting, which impeded its ability to generalize. This study highlights the better performance of the Random Forest model in predicting water quality because of its ability to capture complex non-linear relationships, handle noisy data, and prevent overfitting by aggregating multiple decision trees.
Journal Article