The Simple Image Processing Scheme for Document Retrieval Using Date of Issue as Query

Storing paper-based documents converted into digital image format is one of effective solutions to preserve content in a document. Searching the Document Images stored in a repository by using time as queries is one of importance tasks of document retrieval. This paper presents a simple scheme of image processing to retrieve the documents, which contain the desired date of issue printed in Thai alphabets, from a repository. The procedure of this retrieval scheme consists of 4 stages: image acquisition, pre-processing, zone identification, and pattern recognition. The proposed machine vision scheme gave the excellent result over 365 dummy dates of issue of documents in the tested repository. It is a simple and uncomplex approach for the document image retrieval of Thai government letters, printed in Thai alphabets, using date of issue as query.