Date: 21/11/2021-27/11/2021
Objective:
- Summary of the literature review
Activity:
Smart Glasses using Deep Learning and Stereo Camera
[Method proposed: Deep Learning and Stereo Camera]
Deep learning algorithm cannot be performed on low level MCU of smart glasses since this algorithm compute with a lot of data. The driver is informed that the location of the blind user through the buzzer and LED. The stereo cameras are used to calculate the distance between the blind and the obstacle. The vibration motor and the buzzer operate according to the distance from the obstacle. The YOLO v3 algorithm were used to recognize obstacles. This network is the base feature extractor of Darknet-53. YOLO v3 performs multilevel classification for objects detected in images.
Real Time Multi Object Detection for Blind Using Single Shot Multibox Detector
[Method proposed: Single Shot Multibox Detector]
This assistant is an alert system that captures the surrounding view of the blind person and processes it in real time with frame rate of 60 FPS (frames per second) to detect the objects and guide the subject accordingly. Then the information is sent through a text to speech conversion and the output is fed through speech signals to the earphones connected. The idea is that blind person focuses on objects that are in front and close to him/her. If manufactured as a product, this technology can assist blind people to aid their mobility at an affordable price and can match with future technology as this field is still in view of research and evolving.
Real Time Object Detection Using YOLO v3 Tiny with Voice Feedback for Visually Impaired
[Method proposed: You Only Look Once (YOLO) v3 with Google Voice Feedback]
There are three object detectors namely: Shot Detector (SSD), R-CNN, Fast R- CNN, and Faster R-CNN, Single YOLO. This project uses the sense of hearing to visualize the objects in the surrounding using the “You Only Look Once: Unified, Real-time Object Detection” algorithm trained on the COCO dataset to identify the object present before the person thereafter the label of the detected object is translated to audio by the aid of Google Text to Speech which will be the expected output. This project made with the help of Deep Leaning and Raspberry pi will greatly help visually impaired individuals to the great extent by acting as a tool connecting them to the world and surpassing their disability of vision.
Achievement:
- Gain lot more information and help from readings.