I am a computer science engineering graduate with a specialization in intelligent systems, currently pursuing a Master's in Data Science from the University of Hertfordshire. With a strong proficiency in Python, I have successfully completed multiple projects that demonstrate my expertise in the language. My interests lie in the dynamic fields of machine learning and computer vision, where I am passionate about exploring diverse techniques and their practical applications.
Intern, Company-PINNOTECH January 2023- June 2023
Plant Disease Detection System Developed and fine-tuned algorithms for identifying plant diseases through image processing techniques. Improved system accuracy and performance through meticulous attention to detail and adaptability.
Safety Helmet Detection System Developed an efficient system to identify and ensure the usage of safety helmets within an industrial setting. Enhanced workplace safety by showcasing strong problem-solving skills and the ability to implement real-world solutions.
This project is about the extraction of the Digital Elevation Model (DEM) from a satellite images to get higher accuracy. Used U-net model for DEM extraction and semantic segmentation of satellite image. Images for dataset were taken from Google earth engine. Images were of 512x512 pixel size but were rescaled to 160x160 pixel size for computational reason during preprocessing. I was involved in Dataset creation and implemented U-net model. Technologies: Python, U-NET model, Matplotlib, Numpy, OpenCV, Google Earth Engine
This project was about recognising a chess board and its pieces from an image and generating the FEN code for that image. I was responsible for selection of Dataset and led preprocessing part of the project. OpenCV library was used for image segmentation and feature extraction. Utilized computer vision techniques to extract and recognize chess pieces from images. Technologies: Python, OpenCV, Keras, Tensorflow, Numpy, Matplotlib.
"Sudoku Solver" could solve a Sudoku Puzzle from its image as input. Led preprocessing steps on image dataset. To reduce noise in image dataset used image denoising filters. Input image is transformed into binary image using thresholding and contours which is used to read sudoku from binary image. Imported model using Keras to predict a solution. Technologies: Python,PyCharm CE, OpenCV, Keras, Tensorflow, Numpy, Pandas
M.Sc Data Science University of Hertfordshire | 2024 - Current
B.Tech Computer Science Engineering, Specialisation in Intelligent Systems MIT SOE, ADT, Pune | 2019 - 2024
Python, Linear regression, Logistic regression, Decision trees, Random forest, Artificial neural network, Machine Learning, Deep Learning, Convolutional neural network , Recurrent neural network Libraries and Framework:- Tensorflow ,PyTorch ,Pandas ,Numpy , Scikit ,Matplotlib ,Seaborn ,Keras
Problem solving, Critical thinking, Teamwork, multitasking and attention to detail
Coder's Python course Global Cert | 2021
APPWEB (web development) course Global Cert| 2021
Artificial Intelligence course Skillvertex | 2021
Computer Vision course OpenCV|2022
Parth Tawde, High-Resolution Satellite Imagery Analysis for Terrain and Surface Data Extraction: Techniques and Applications, 2023