摘要:The handwritten character recognition is potentially an active area of research due to the presence of several challenging issues. Due to a large variation in writing styles, development of optical handwritten character reader is a challenging task. In order to decrease the burden of computation and to improve the recognition accuracy, several measures need to be taken in the overall process of recognition. The main objective of this review was to recognize and analyze handwritten document images. There are wide varieties of classification techniques available for the problem of pattern recognition . These techniques include Support Vector Machine (SVM), Back Propagation Neural Networks (BPNN), Probabilistic Neural Networks (PNN) and many more. In this study, a survey has been performed on some of these machine learning techniques for the identification of various handwritten north Indian scripts. This study attempts to address the most significant results obtained so far and then all the gathered data is represented in the form of tables so as to have a clear idea by visualizing data at once. This research paper provides a comprehensive survey on various machine learning techniques involved in north Indian script recognition and the study also highlights the crucial aspects of the research till date.