sachan99/engrafo_manuscript-digitizer

ENGRAFO_MANUSCRIPT-DIGITIZER

Overview:

The Manuscript Digitization Project aims to digitally preserve handwritten manuscripts with significant scientific, historical, or aesthetic value dating back at least 75 years. Manuscripts are primary sources of historic information and represent an ancient and rich cultural heritage, particularly in India, where the largest collection of manuscripts exists. Unfortunately, these manuscripts are diminishing over time due to natural decay and other factors. To counteract this issue, the project focuses on digitizing the manuscripts using a mobile application. The application allows users to capture images of handwritten manuscripts using a camera or upload them from their gallery. The text from the captured images is then detected and extracted, and the content is converted into PDF format for preservation.

Key Features:

Digitize handwritten manuscripts using a user-friendly Web application. Multiple manuscript images can be uploaded. Detect and extract text from the manuscript images. Offers text extraction for six Indian languages (Sanskrit, Telugu, Marathi, Bengali, Arabic, and Tamil). Convert the extracted text into PDF format for preservation. Help preserve the scientific, historical, and aesthetic knowledge contained in these manuscripts.

Motivation:

India possesses an extensive collection of manuscripts written in various languages and scripts, including Sanskrit. These manuscripts hold invaluable knowledge in medicine, science, mathematics, literature, art, architecture, theology, philosophy, music, dance, and more. Preserving these manuscripts is vital not only for preserving the information they contain but also for understanding and appreciating the history and culture of the nation. Traditional methods of preservation have been utilized, but modern techniques of digital preservation are becoming increasingly important. The government, along with institutions like the National Library, National Informatic Center, and National Archives, is actively involved in preserving cultural heritage through strategies and policies. However, the manuscripts are scattered among different organizations and individuals, necessitating collaborative efforts to digitize and preserve them using modern technologies.

Requirements:

-f https://download.pytorch.org/whl/torch_stable.html

Flask==1.1.2

ipython==7.8.0

ipython-genutils==0.2.0

easyocr==1.2.2

numpy==1.16.5

numpydoc==0.9.1

gunicorn==19.9.0

torch==1.8.1+cpu

torchvision==0.9.1+cpu

opencv-python-headless==4.2.0.32

openpyxl==3.0.0

Website pages:

Home page:

Help page:

Digitizer Page:

Archive page:

Feedback Page: