• Raj Bharath. R
• Nagaraj. R
• Yuvaraj. M
• Aadith U.K

Text is one of the amazing inventions of human. It acts as a bridge between knowledge gainer and knowledge giver. Text can give us the life for what we see. In a field of Computer Vision, text detection and recogoniton plays a vital role in achieving the intelligence for machine. As a human text detection is an easy task. But an achieving that machine is an crucial task. The main objective is to integrate text detection and recognition as one thing. The System consists of 4 modules. They are 1)Text Detection 2)Preprocess the Detected part 3)Text Recogonition 4)Exporting as Documents. The first part is text detection by using Efficiency Accuracy Scene Text Detector and obtaining region of interest. The efficiency accuracy scene text detector contains of two steps only. A first step is to apply Fully Convolution Network to produce text regions. A second step is to apply non-maxima suppression to avoid multiple region of interest. The second part is preprocess the region of interest for increasing the efficiency of recogoniton. This is done by applying threshold to easily separate text from background and apply Gaussian blur for it. The third part is text recogonition by using tesseract OCR. The fourth part is exporting text into documents format by using docx, pypdf package.

Text Detection, Text Recogoniton

"Text Detection and Text Recogonition in MultiScript", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.7, Issue 7, page no.458-461, July-2020, Available :http://www.jetir.org/papers/JETIR2007357.pdf

"Text Detection and Text Recogonition in MultiScript", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.7, Issue 7, page no. pp458-461, July-2020, Available at : http://www.jetir.org/papers/JETIR2007357.pdf