The project focuses on recognizing handwritten digits using machine learning. The project uses supervised machine learning and the training data consists of more than 100 training images of 4 digits (0 to 3). All these images are 28 x 28 pixels. The user has to feed a 28 x 28 pixels image with white background and black handwritten digit, which is then processed, and using machine learning the most appropriate result is displayed.
YouTube Link: https://youtu.be/_XdQbfou450