Object Detection Deep Learning Model for Smart Blind Stick
Description :
Our Smart Blind stick project utilizes Deep Learning to detect objects in the input frame captured by a camera and then giving voice feedback to the blind person of detected objects plus providing vibration based on readings of ultrasonic sensors to notify him of obstacles faced. We also provided a mobile application to the guardian to monitor the location and heartbeat rate of the holder of the stick.
So you can find the model which is :
SSD MobilNet trained on COCO dataset and then converted into TensorFlow lite version after quantization ( for implementation on Raspberry pi 4 model B (4GB)