The diploma thesis presents the process of implementing a system to assist the blind and
visually impaired using a video camera, speakers, and Python software. Initially, the daily
challenges faced by the blind and visually impaired are discussed. Based on these challenges,
it is determined that the main emphasis of the practical work will be text recognition and human face detection. The paper begins by presenting the history of technological solutions, followed by an overview of assistive solutions available on the market. An analysis of these tools is given, and an assessment of their positive and negative attributes is presented. Based on the results of the analysis, it is determined that the solution will be developed for the Slovenian language, as most of the analysed tools are focused on larger linguistic communities. The technologies used and necessary for the system’s implementation are then described and presented.
Based on the findings, goals were set for the development of a prototype system. The first
functional part uses the video camera to detect and recognize various faces in the camera’s field of view. It informs the user about who and where the detected persons are located. The second functional part reads Slovenian text in the camera’s view, moreover it guides the user on how to direct the camera. In the end the system is tested, and an evaluation of the final solution is provided.
|
