Reply to thread

Message: <blockquote data-quote="nelik" data-source="post: 6836561" data-attributes="member: 60402">Optical Character Recognition System  Previous works at the UCSC concerning OCRs had concentrated on developing a technique best suited for detecting printed Sinhala characters. This component of work focused on converting that research into a real product by making it robust for variations in font size, particularly those commonly used by the majority of the people including newspaper prints and government publications. Later it will be developed into a font-independent OCR software. You can download this software from the <a href="http://www.ucsc.cmb.ac.lk/ltrl/?page=downloads&lang=en&style=default" target="_blank">download</a> section.                  The methodology used for construction of the OCR consisted of following steps <ol> <li data-xf-list-type="ol"> Preprocessing activities:<ol> <li data-xf-list-type="ol"> Scanning documents and skew detection</li> <li data-xf-list-type="ol"> Noise detection and removal</li> <li data-xf-list-type="ol"> Extraction of text characteristics and individual characters</li> </ol>   </li> <li data-xf-list-type="ol">Data collection:<ol> <li data-xf-list-type="ol"> Identification of representative texts</li> <li data-xf-list-type="ol"> Separation of training, validation and testing sets</li> </ol>   </li> <li data-xf-list-type="ol">Processing:<ol> <li data-xf-list-type="ol"> Feature extraction and pattern matching</li> <li data-xf-list-type="ol"> Testing of competing algorithms</li> <li data-xf-list-type="ol"> Optimization of algorithms</li> <li data-xf-list-type="ol"> Application development</li> </ol>   </li> </ol>                  <a href="http://www.ucsc.cmb.ac.lk/ltrl/images/panl10n_p1_ocr_full.jpg" target="_blank"><img src="http://www.ucsc.cmb.ac.lk/ltrl/images/panl10n_p1_ocr_sample.jpg" alt="" class="fr-fic fr-dii fr-draggable " style="" /></a></blockquote>

Verification: Nawa warak dahaya keeyada? (Namaya wadi kireema dahaya)

Top Bottom