In-Pipe Bot

A Long-Term Project done as a Core of Robotix Club ( The Robotics and Automation Club of IITH). The project focuses on identifying|inspecting the cracks at the inner surface of industrial pipes. OpenCv.ipynb is the implementation of crack detection using in-built functions of OpenCV library. While CNNs.ipynb is the implementation using Convolutional Neural Network, trained on 20,000 Positive and 20,000 Negative images obtained from Kaggle The above implementation using CNNs is trained to identify cracks on Concrete Surfaces.