Design and implementation of android application to extract text from images by using tesseract for English and Hindi

Abstract

The proposed Implementation is on the Android Application to extract using Tesseract OCR in which the following concepts will be used, which are Adaptive Thresholding, Connected Component, Fine Lines, and Recognize Word. Using this Optical Character Recognition (OCR) Technology, an Application generated text which is printed on a clean, B/W or colourful background can be converted into a computer readable form ASCII. With the help of this Android Application using Tesseract OCR, the system has two ways for Text Extraction. The first one is to capture a photo while the second one uploads an image from the gallery after that system can proceed for as per the user requirement which portion of the image they want to crop or edit. After editing the picture, it converts into the text. This Android Application is for two languages, English and Hindi.

Implementasi Hirarki Dataset Dalam Membangun Model Language Aksara Bali Menggunakan Framework Tesseract OCR

“…Containers memiliki sejumlah keunggulan yang membuatnya populer di kalangan alat pelatihan data, termasuk konfigurasi yang sederhana, tingkat keamanan yang tinggi, kemampuan untuk berjalan di berbagai platform cloud, kemampuan diagnostik, dan dukungan untuk berbagai sistem operasi [15]. Prosedur pelatihan dataset terdiri dari dua proses utama: pelatihan bentuk karakter dan kompilasi kamus bahasa [16]. Hasil dari pelatihan ini adalah model language Tesseract yang telah dilatih secara khusus untuk mengenali karakter dalam dataset tersebut.…”