Handwriting Recognition using Convolutional Neural Network and Support Vector Machine Algorithms

The goal of the Handwritten Text Recognition system is to transform human handwritings into digital text. Handwriting Recognition has been one of the active and challenging research areas in the field of image processing and pattern recognition. Going back to the history of handwriting recognition, the first step created was a special pen device which was also known as'S RI pen’ established in 1964. From that, the technology has evolved so much better. This technology will improve further in the future; one among such trials is this proposed system. The proposed research employs Convolutional N eural Network and Support Vector Machine as classifiers, EMNIS T as dataset with suitable parameters for training and testing for Handwriting Recognition and providing 95.41 % accuracy with lesser computational time to recognize English alphabets from A-Z, along with digits from 0–9. The algorithms, Convolutional Neural Network and Support Vector Machine are used here in this research.