Detection of Dot Matrix Printed Documents Using Component Analysis

Document Quality Analysis is one of the most prominent areas of interest these days. It is used for managing of huge number of documents, detection of fraud document, recognizing identity documents etc. A lot of work has been done in the areas of coloured photograph recognition, improving the quality of documents and printing technology. However, to the best of our knowledge no work has been done to recognize the dot matrix prints from other high quality prints using principle component analysis. This is useful for automatically recognizing the text on a certain print. Post the detection of the quality of the print, recognition of the text can be easily done using specific algorithm for each quality. In this paper, the above objective is achieved by applying principle component analysis before and after smearing of the text on a particular print and then classifying it into dot matrix or non-dot matrix print. Moreover, this approach has also been compared with prior works related to classification and detection of print quality.