Experimental Evaluation of Smart Camera based Reading Assistance for Visually Impaired People using Optical Character Recognition Logic

Braille is used by the majority of visually impaired people to read books and papers that are not easily produced or accessible in any other way. The invention of technologies that might alleviate the painful chores endured by the sight handicapped is thus necessitated. Thanks to book digitalization, many brilliant minds in academia, business, and government have tried to construct a powerful document analysis system, but thus far, only the visually impaired have been able to benefit. Using a Raspberry Pi as a platform, this project seeks to research picture recognition technologies that incorporate voice synthesis and create an affordable, user-friendly image-to-speech conversion system. This research presents a novel method called Modified Optical Character Recognition Logic (MOCRL). To test how well it works, it is cross-validated with the tried-and-true model called Neural Text Analyzer Logic (NTAL). In the suggested system, a tiny built-in camera scans printed text, transforms it to audio files utilizing a synthesized voice, and subsequently reads audibly the scanned information. This makes it possible to instantaneously translate books, papers, and other common items, even while you’re not at home or working. In addition to saving effort and time, this improves the quality of life for the visually impaired by giving them more freedom to accomplish things on their own.