A New Method for Line Segmentation of Persian Handwritten Documents
عنوان مقاله: A New Method for Line Segmentation of Persian Handwritten Documents
شناسه ملی مقاله: TDCONF01_212
منتشر شده در اولین همایش ملی الکترونیکی پیشرفت های تکنولوژی در مهندسی برق، الکترونیک و کامپیوتر در سال 1393
شناسه ملی مقاله: TDCONF01_212
منتشر شده در اولین همایش ملی الکترونیکی پیشرفت های تکنولوژی در مهندسی برق، الکترونیک و کامپیوتر در سال 1393
مشخصات نویسندگان مقاله:
S. Parisa Hosseyni Moghadam - Computer and IT Department Faculty of Engineering, Bandar-Abbas branch, Islamic Azad University,Bandar-Abbas, Iran
Abbas Akkasi - Computer and IT Department Faculty of Engineering, Bandar-Abbas branch, Islamic Azad University,Bandar-Abbas, Iran
خلاصه مقاله:
S. Parisa Hosseyni Moghadam - Computer and IT Department Faculty of Engineering, Bandar-Abbas branch, Islamic Azad University,Bandar-Abbas, Iran
Abbas Akkasi - Computer and IT Department Faculty of Engineering, Bandar-Abbas branch, Islamic Azad University,Bandar-Abbas, Iran
In this paper, a new method for finding of the lines segmentation in Persian handwritten texts, is presented. At first, we remove the slants and dots that, which are located between the lines, then cut the connected components height. The next step we connect the words in each line, after connecting words we consider the lowest black pixel in each line as a separate line. The performances of proposed method have been evaluated in standard data base that includes all the challenges in lines segmentation. The results show that this method has good potential
کلمات کلیدی: Textual line segmentation; Persian handwritten recognition
صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/362574/