| Autor: |
Yıldırım, İrem, Demir, İlyas, Peker, İsa, Yılmaz, Mehmet Emin, Güneyli, Ozan, Aktaş, Ezgi Şevval, Özer, Elif, Kantaroğlu, Emrah |
| Předmět: |
|
| Zdroj: |
Journal of Computer & Electrical & Electronics Engineering Sciences (JCEEES); 2024, Vol. 2 Issue 2, p62-66, 5p |
| Abstrakt: |
Visually impaired individuals face significant challenges in many areas of their daily lives, such as accessing environmental information, social interaction, and safety. This article aims to enhance the independence of visually impaired individuals and facilitate their daily lives through an artificial intelligence-supported image and sound recognition system. The developed prototype consists of glasses equipped with a Raspberry Pi 4 microprocessor, integrated headphones, touch controls, and a robust battery management system. Its electronic schematic includes the integration of the camera and the microprocessor. The artificial intelligence algorithms developed successfully perform tasks such as object recognition, text reading, and face recognition. A visual question answering (VQA) model was utilized in the glasses to enable visually impaired individuals to communicate more effectively with their surroundings. Following a demonstration conducted with visually impaired individuals, the usability and effectiveness of the prototype were evaluated, and improvements were made based on user feedback. The accuracy of the algorithms was tested, resulting in a prototype with a 78% accuracy score, which was close to the 82% accuracy score of GPT-4. Additionally, tests conducted with 20 users indicated an 85% user satisfaction rate. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
