Publication Date: 2023/05/29
Abstract: Millions of people worldwide have visual impairments, and integrating them into society is a crucial ongoing goal. To support their quality of life, various guidance systems have been developed, often for specific purposes. However, these solutions can significantly enhance the mobility and safety of visually impaired individuals. To address this, a vision-based platform using Python and OpenCV library functionalities has been developed to recognize realworld objects indoors and outdoors. YOLO is a novel approach to object detection that has been used in the software. The image is transformed into a scan image for further interpretation of its contents. Efforts continue to support visually impaired individuals and enable their full participation in society. The detected image is scanned and fed into Tesseract OCR for conversion to text. Additionally, facial emotion recognition is applied to determine a person's mood. Finally, a Text-to-Speech (TTS) engine is utilized to convert the detected text and objects into audible speech.
Keywords: YOLO, Open-CV, Tesseract OCR, TTS, facialbased emotion recognition
DOI: https://doi.org/10.5281/zenodo.7981190
PDF: https://ijirst.demo4.arinfotech.co/assets/upload/files/IJISRT23MAY1475.pdf
REFERENCES