Embedded Files
Automated Driver License Testing
Automated Driver License Testing
Watch Microsoft CEO Satya Nadella explain the project in 2 sentences!
Supervisor: Dr. Akshay Nambi
The project is about administrating driving skills based on a single smartphone. A windshield mounter smartphone camera is used to detect the evaluate driving maneuvers using back camera and driver's gaze using front camera.
The project has been deployed in IDTR Dehradun, IDTR Aurangabad, and has been scaled up for multiple RTO sites.
Hyperspectral Imaging
Hyperspectral Imaging
This project is based on the classification and regeneration of Hyperspectral images based on Infrared(LWIR) and Visible range satellite image datasets using Deep Learning and Data Fusion techniques.
The task is to super-resolve LWIR (Hyperspectral Image) using Visible RGB image and indentify the number of classes based on the dataset.
Supervisor: Dr. Ganesh Bagler
RecipeDB is a website of recipes and ingredients from over 22 world regions, intended to enable data-driven explorations of recipes.
It facilitates multi-level analysis of traditional recipes (dietary classifications, ingredient composition, nutritional profile, recipes, etc).
Developed statistics and visualizations between the relationships and statistics in the database.
Denoising and Super-Resolution of Images
Denoising and Super-Resolution of Images
This project includes the implementation of architectures for Super-resolution of images and videos for the NTIRE Challenge 2019.
Developed models for the implementation which include conditional auto-encoders and deep back-projection networks. Achieved the PSNR of 36.xx for the image denoising task.
Secured 11th rank for Video Super-resolution challenge and got the architecture published.
The project was led by Dr. Manoj Sharma, former Scientist Fellow at CSIR-CEERI.
A bot that traverses a square grid using white line sensors and plucks cardboard made fruits and drops them into various deposition zones for different fruits.
Worked on Digital Image Processing using OpenCV.
Built Path Planning algorithms for the traversal of the FIREBIRD V bot.
Built hardware for sensors and mechanisms.
Dehazing of Underwater images and Videos
Dehazing of Underwater images and Videos
This project includes algorithm of dehazing of underwater images and videos using both OpenCV and Deep Learning algorithms seperately.
Worked on White Balancing and Thresholding techniques.
Worked on Deep Learning Architectures for Dehazing Datasets.
The project was led by Dr. Rubeena Vohra, Assistant Professor, BVCOE
Escalade 7.0, IIT Guwahati
Escalade 7.0, IIT Guwahati
Escalade is one of the robotics event organised by Robotics module of a techno-managerial fest of IIT Guwahati called "Techniche"
The task was to develop two bots:
Manual bot contains hardware mechanisms for picking, dropping of the boxes and lifting mechanism for climbing the wall.
The Autonomous bot consists of Line Following, Wall Following, and Colour Sensing algorithms using sensors and Arduino as microcontroller.
Google Sites
Report abuse