Explainable AI

Explainable AI focuses on making AI systems transparent and understandable. It aims to improve trust by providing clear reasoning behind AI decisions, particularly critical in sensitive applications like healthcare and finance.

Trustworthy AI

Trustworthy AI emphasizes reliability, safety, and ethical alignment of AI systems. This approach ensures AI behaves predictably, responsibly, and aligns with societal values, addressing concerns around data privacy and accountability.

Precision Scalable DNN accelerator Chips

With the increasing demand for reconfigurable and precision scalable hardware architectures, we design and develop DNN accelerator Hardware.

Computing in memory based DNN accelerators

With the energy-efficiency bottleneck created by von-neumann-architectures. We design and develop SRAM/NVM based Computing in memory DNN accelerators

Nutrient Deficiency Detection System for Agricultural Drones

Timely detection of deficiencies like nitrogen, potassium, zinc, magnesium, and iron ensures healthier plants and higher crop yields to the farmer. We specialize in using light weight Deep Learning models on IoT edge devices deployed in agricultural drones for accurate and real time diagnosis of diverse range of crops. This will improve the soil health and enhance crop quality with a reduction in overall and costs associated with overuse of fertilizers, etc.

Crop Disease Detection System for Agricultural Spraying Drones

Crop diseases pose a significant threat to global food security, often leading to massive losses for farmers. Traditional methods of disease detection rely on manual inspections, which are time-consuming and prone to errors. We specialize in using light weight Deep Learning models on IoT edge devices deployed in agricultural drones for accurate and real time diagnosis and simultaneous spraying by drones of diverse range of crops.

Air Quality Monitoring System

Integrating drones with air quality monitoring sensors offers a dynamic solution to collect real-time environmental data across urban landscapes. These drones can be deployed to monitor pollutant levels, track emission hotspots, and respond to environmental emergencies effectively. Simultaneously, embedding air quality sensors into smart street lights creates a distributed, energy-efficient network for continuous data collection.

Smart Waste Management system

Drones equipped with cameras and sensors can monitor garbage levels in bins, detect illegal dumping sites, and identify pollution hotspots with real-time aerial data. This allows for precise and swift response in waste collection and disposal. Additionally, drones with load-carrying capabilities can transport waste from hard-to-access areas to designated collection points, reducing the need for human involvement in hazardous locations.

Inspection of Power Lines and Power Grid

Drones can be transforming power grid inspection. They swiftly and efficiently access remote and hazardous areas, reducing the need for human personnel in dangerous tasks. Equipped with high-resolution cameras, drones can capture detailed images of power lines, towers, and infrastructure.

Detection of Structural Damages and Wear

Detecting structural damages in roads, buildings, and bridges using AI has revolutionized maintenance processes. AI-powered systems analyze high-resolution images and videos, captured by drones, to identify cracks, deformations, and surface irregularities.

Seed Quality Inspection System

Poor seed quality leads to uneven germination and lower crop yields, resulting in financial losses for farmers. Our seed quality assessment system utilizes advanced imaging and AI-powered algorithms to analyze seed attributes such as size, shape, color, and viability.

Food Quality and Grading System

Accurate grading systems ensure that fruits, vegetables, and other produce meet established quality standards. Our AI-powered food quality and grading system uses computer vision and machine learning to evaluate parameters like size, color, shape, texture, and freshness. This automated approach eliminates the subjectivity of manual grading, providing consistent and reliable results.

PCB Defects Detection

Identifying and addressing PCB defects is crucial to ensure reliability and safety of Electronic systems. Common issues include connectivity problems, component misplacement, and thermal instability, all of which impact performance and lifespan. Lightweight Deep learning models will be deployed for real time PCB inspection systems.

Wafer Defects, How to Proceed

Handling wafer defects involves inspection, classification, and repair strategies to maintain semiconductor quality. Techniques such as machine vision and deep learning help in defect detection, improving yield and reducing manufacturing costs.