This paper implements a handwritten Kannada character recognition machine with the use of the SVM classifier. It utilizes k utilizes SVM classifiers for the usage of pixel values as a measure to recognize the characters. SVM gives good results when computing with a small sample size because it has the upper hand in nonlinear and high-dimensional pattern recognition and has better approximation ability and generalization ability. Dataset of all the Kannada letters was prepared and accuracy was around 90%
Website link: https://ieeexplore.ieee.org/document/9376883