EFEKTIFITAS PEMBERKASAN DOKUMEN DIGITAL MELALUI METODE TEMPLATE MATCHING CORRELATION

Scanning printed documents in the form of digital documents has become a necessity for many institutions and companies, as well as in LP3I Polytechnic Jakarta. One document that is often on over the media, i.e. diplomas and transcripts. With an average number of graduates reaching 1,900 per year, certainly many documents must be transferred to provide and be filed well, but there is no application that can manage documents over media that have already been given file names and unstructured filing, so the files are it will be difficult to search again. Generally, official documents include a unique number as an identity with a certain code structure, such as NIM, diploma number and transcript number. The purpose of this research is to determine the effectiveness of document scanning results by utilizing optical character recognition techniques known as OCR (Optical Character Recognition) on the unique number. The method used in this optical character recognition process is to use template matching correlation, which is a technique to get the highest value in the comparison of characters in input images, which have been segmented into letters and processed in binary image formats, and characters in template images. The results of this study are in the form of a prototype application using Matlab that can detect unique numbers in documents as a reference in filing files in the form of naming new files and placing them in the desired folder. Key words: documents, digital, template matching correlation