This is a small project demostrating the concepts of Image Processing, Feature Extraction and Pattern Recognisation. Features are extracted of the English alphabets using the 7 invariant moments algorithm also known has Hu moments. The feature extraction is commonly done in every machine learning project where datasets are used. So this extraction of the features can be further on used in digitization of the text. For every alphabet there are 7 features extracted from it that are horizontal features, vertical features and such features comes under translation, rotation and orthogonal transformation of a 2-D image.
Optical character recognition(OCR), invariant moments, K-nearest neighbor, OpenCV (cv2), numpy, skimage, Segmentation, Glaussian Blur, Erode, Dialation, Sobel Edge Detection, Distance Mactrix, Confusion Matrix, tkinter, imagetk.
Drafted Research Paper: (Click here to view paper)