Zhenping CuiFollow


This publication describes a text scanner and touch reader application software for visually-impaired users (e.g., blind or partially blind) that may help them gain a higher degree of independence and privacy. As the visually-impaired user moves a user equipment (UE) around a scene with text, the application software records the scene with text. To aid the visually-impaired user record the whole text of interest, the application software verbally instructs the visually-impaired user on how to move the UE. Once the application software collects an adequate input, it instructs the visually-impaired user to stop recording. The application software uses machine learning to analyze the captured scene with text to generate a full picture of the scene with text. Then, the application software runs optical character recognition (OCR) over the full picture of the scene with text. The visually-impaired user may choose to have the UE read aloud the whole text or may slide his or her finger across the UE’s screen to hear certain parts of the text, similar to reading Braille.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.