An Approach Script Identification from Multilingual Documents using KNN and CNN

In the context of our country, script recognition is a complex task because of more number of prevailing scripts and that many of the official documents are multiscripts in nature, English is being commonly used along with regional script. Hence, it is pre-requisite to identify the script in the document and then feed the document to the respective OCR for further processing. In this work, we explore the significance of Gabor and SIFT operator for script identification at the word level respectively. Identification of script type is done by using K-NN and CNN classifiers