India is a multi-linguisticcountry and Hindi is a national language of India. There is no such work has been done on offline recognition of Hindi characters so that the Hindi data is stored digitally and the paper work will reduce and the data also store safely for the long period of time because as we know that the data on the paper is not secure, paper may lost or may get faded. Therefore in this paper an algorithm is proposed torecognize the Hindi character optically using k-means clustering algorithm. HCR is not same as the English character recognition because Hindi characters are joined together with the shirorekha which is the line on the upper part of the characters and in English language there is no shirorekha. So in English there is no need to remove that shirorekha but for recognize Hindi character it is necessary.


K-means clustering algorithm is used for cluster the same data into their respective clusters and for classification.The objective of this paper is to provide a high performance OCR solution for Devanagari script that can help in exploring future applications such as navigation, for ex. traffic sign recognition in foreign lands etc.