Affiliation:
1. Amrutvahini Polytechnic, Sangamner, India
Abstract
Visually impaired individuals face significant challenges when navigating and engaging with their surroundings independently. Our solution, "Smart Assistive System for visually Impaired using pi" employs a Raspberry Pi and camera for real-time image capture, precise object classification (with over 90% accuracy), and auditory feedback. The project addresses a pressing need for greater inclusion and accessibility for the visually impaired, offering a cost- effective and innovative solution that converts visual information into non-visual cues. The "Caption-Speak" system holds the potential to significantly enhance the independence, mobility, and overall quality of life for visually impaired individuals.
Reference6 articles.
1. H. Sharma, M. Tripathi, A. Kumar, and M. S. Gaur, ‘‘Embedded assistive stick for visually impaired persons,’’ in Proc. 9th Int. Conf. Comput., Commun. Netw. Technol. (ICCCNT), Jul. 2018, pp. 1–6, doi: 10.1109/ICCCNT.2018.8493707.
2. S. K. Jarraya, W. S. Al-Shehri, and M. S. Ali, ‘‘Deep multilayer perceptron- based obstacle classification method from partial visual information: Application to the assistance of visually impaired people,’’ IEEE Access, vol. 8, pp. 26612–26622, 2020, doi: 10.1109/ACCESS.2020.2970979.
3. Getting the Model From TensorFlow Hub. [Online]. Available: https://www.coursera.org/lecture/advanced-computer-vision-withtensorflow/getting-the-model-from-tensorflow-hub-tZDzl.
4. P. Viola and M. Jones, ‘‘Rapid object detection using a boosted cascade of simple features,’’ in Proc. Comput. Soc. Conf. Comput. Vis. Pattern Recognit., 2001, pp. 1–5.
5. M. Aghagolzadeh, H. Soltanian-Zadeh, and B. N. Araabi, ‘‘Multiscale face detection: A new approach to robust face detection,’’ in Proc. Int. Conf. Fuzzy Syst., Jul. 2006, pp. 1229–1234.