Using a Hybrid Model for Image Recognition of Handwritten Text

No Thumbnail Available
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Samarkand branch of TUIT
Abstract
Description
In the recognition of Uzbek handwritten letters, an important issue is segmentation and its recognition. The curvature of the handwriting, cursive and word-to-word problems, incorrect placement of diacritics, and the presence of ups and downs lead to the processing of the segmentation problem. In the manuscripts, the performance of linear and word segmentation is shown based on results obtained from software based on the CNN+LSTM+CTC neural network model.
Oʻzbek matni qoʻlyozma harflarni tanib olishda segmentatsiya masalasi va uni tanib olish muhim hisoblanadi. Qo’lyozma matnning egriligi, bir-birini takrorlashi va tegishi bilan bog'liq muammolar, kursiv bog'lanish, noto'g'ri pozitsiya diakritik belgilar, ko'tarilish va tushish borligi segmentatsiya masalasini qayta ishlashga olib keladi. Qo’lyozmalarda CNN+LSTM+CTC neyron tarmoq modeli asosida dasturiy ta’minotdan olingan natija asosida qator segmentatsiyasi va soʻzlarga ajratish samaradorligi koʻrsatib berilgan.
Keywords
segmentation, neural network, filtering, topology, handwriting, python, segmentatsiya, neyron tarmoq, filtratsiya, topologiya, qo’l yozma, python
Citation