This repository contains the source code of a simple digit recognizer. The training and testing data comes from the the MNIST database of handwritten digits. The image classifier is based on the k-nearest neighbors algorithm.
Each image is 'unwrapped' pixel by pixel to form a one-dimensional array, which is used as a feature vector. Distances between feature vectors are computed in a Euclidean n-dimensional space (n = 28^2)
A better way to obtain features is to extract them directly from the shape (for example the amount of holes, open areas etc). Such an implementation is being developed in the digit-features-v2 branch.
gradle run