Open Access Open Access  Restricted Access Subscription or Fee Access

An Insight for Visually Impaired using AI Techniques

Amal K Santhosh, Natheera Beevi M, Lekshmi S S


We know that the life of blind people is very risky. They always need an assistance or another person for help her. In this project we introduce an AI Spectacles for blinds, which help them to find what happening in front of them and they can find their own things without any help. In this proposed system we using a real time object detection using YOLO v3 model: You only look once, or YOLO, is one of the faster object detection algorithms out there. Despite the fact that it is not, at this point the most exact item identification calculation, it is a generally excellent decision when you need continuous location, without loss of a lot of exactness. A route collaborator for the outwardly debilitated individuals utilizing object identification and text to discourse. “An Insight for visually impaired using AI techniques” the spectacles consist of an input camera and a part of headset. The working of the AI specs is the objects and obstacles in front of blind person are captured by the camera was detected then recognize and convert into voice and then passed through headsets.


YOLOV3, Google text to speech, Raspberry pi, Darknet NN-Pack

Full Text:



Reagan L. Galvez, Argel A. Bandala, et al. Object Detection Using Convolutional Neural Networks. TENCON 2018 - 2018 IEEE Region 10 Conference. 28-31 Oct. 2018; Jeju, Korea (South), New York: IEEE; 2019.

Ross Girshick, Jeff Donahue, et al. Rich feature hierarchies for accurate object detection and semantic segmentation. IEEE Conference on Computer Vision and Pattern Recognition. 23-28 June 2014; Columbus, OH, USA, New York: IEEE; 2014.

Ross Girshick. Fast R-CNN. IEEE International Conference on Computer Vision (ICCV). 7-13 Dec. 2015; Santiago, Chile, New York: IEEE; 2016.

Shaoqing Ren, Kaiming He, et al. Faster R-CNN: Towards RealTime Ob-ject Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2017;39(6):1137–1149.

Peilei Dong and Wenmin Wang. Better region proposals for pedestrian detection with R-CNN. Visual Communications and Image Processing (VCIP). 27-30 Nov. 2016; Chengdu, China, New York: IEEE; 2017.

Joseph Redmon and Ali Farhadi. YOLO9000: Better, Faster, Stronger. IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 21-26 July 2017; Honolulu, HI, USA, New York: IEEE; 2017.

Johannes Günther, Patrick M. Pilarski, et al. First Steps towards an Intelligent Laser Welding Architecture Using Deep Neural Networks and Reinforcement Learning. Procedia Technol. 2014;15:474– 483.

Rasim Caner Çalik and M. Fatih Demirci. Cifar-10 Image Classification with Convolutional Neural Networks for Embedded Systems. IEEE/ACS 15th International Conference on Computer Systems and Applications (AICCSA). 28 Oct.-1 Nov. 2018; Aqaba, Jordan, New York: IEEE; 2019.

Alex Krizhevsky, Ilya Sutskever, et al. ImageNet Classification with Deep Convolutional Neural Networks. Advances in Neural Information Processing Systems 25 (NIPS 2012). 2012. doi: 10.1201/9781420010749.

Mark Everingham, Luc Van Gool, et al. The Pascal Visual Object Classes (VOC) Challenge. International Journal of Computer Vision. 2010;88:303-338.


  • There are currently no refbacks.