Abstract

In the business transactions of large corporations such as utility companies and banks, many cheques are being processed on a regular basis. Automatic reading of bank cheques is an active topic in Document Analysis. A machine capable of reading bank cheques will have wide applications in banks and those companies where huge quantities of cheques have to be processed. To realize such a system, many image processing, pattern recognition and OCR techniques must be involved. However, to develop an effective item extraction system is a difficult task, especially when the cheques contain complicated and colourful background pictures. In this thesis, a novel approach is proposed to locate the baselines of the cheque quickly and efficiently from a grey-scale cheque image. Based on that, we can extract the items such as the legal amount; courtesy amount and date respectively via a layout-driven extraction method. We have developed a system that can extract cheque items effectively and automatically combining all the current methods. The result is quite encouraging and a reliability of 98% has been achieved when tested on IRIS database which contains 600 gray-level cheque images. The performance analysis also indicates that our program is less time-consuming than the existing extraction system