Application of K-Nearest-Neighbors Classification for Number Recognition
Akshay Nagpal · International journal of engineering and future technology · 2016
In this paper, I have presented an application of k-nearest-neighbors classification algorithm to single digit recognition. A classifier function was developed which could recognize the number present in the image. The input to the classifier function consists of PNG and JPEG files containing a single digit. The classifier recognized the number by using the dataset of text files given to it. It was observed that the accuracy of the classifier improved with every iteration of the program as the program learnt from its earlier errors. Finally, further improvements and future scope of the model has been discussed.