Advances in OCR: Template-Driven Data Extraction from Government Documents

— Today, almost everyone has at least one form of personal identification, be it an ID card or a driver's license. Engaging with documents or needing to present them in various scenarios has become a commonplace aspect of our daily lives. In light of the ongoing pandemic, we have increasingly turned to digital communication methods, adapting to new ways of working and handling matters generally through online tools and platforms. In this scenario, we have made some headway against bureaucratic hurdles, as many documents that previously had to be physically presented can now be scanned and emailed, or uploaded via website forms. This shift suggests that the pandemic may have instilled some positive changes that could persist beyond the current crisis, helping to eliminate unnecessary time expenditures and reduce overcrowding. The aim of this project is to develop an application that can extract textual data from an image provided by the user. This capability will facilitate the sharing of digital document versions, significantly enhancing the convenience of our everyday tasks.