Camera & Stabilization System to Provide Object Detection/Tracking in Real Time for Assistance of the Visually Impaired

CSEF · 2019 Electronics & Electromagnetics Fourth Award

Overview

Objectives The visually impaired often have trouble navigating their environments due to a lack of knowledge about the environment around them. My project hopes to solve this issue by creating a hand-held camera stabilization system that can assist the visually impaired by using multiple sensors, motors, and audio feedback, providing object recognition tracking, at a lower cost than guide dogs and more effective than canes. Methods To complete the object detection and tracking part of this project I tested the different tracking systems on a laptop and a Raspberry Pi to evaluate the best one, measuring them in terms of speed and ability to track effectively, recover from occlusion, changes in light, and other factors. I then designed a camera mounting system that contained 3 motors and multiple sensors including a camera, gyroscope, and rangefinders, and I collected data on the system. The initial goal was to tune PID constants quickly, less heuristically and more algorithmically. I was able to notice correlations in PID data that allowed me to complete the project. Results The final system was able to adequately stabilize a system (< 5-10 degree oscillation when encountering rotations (max angular velocity of 1 rad/sec). In addition, with a laptop, the system was able to track people quickly (>100 fps), perform object detection adequately quickly (>15 fps), and read information. Conclusions PID Constants can be tuned faster and more effectively with careful analysis of data. I also analyzed multiple tracking algorithms and object detection algorithms on constrained, low cost systems like the Raspberry Pi. The final system was able to adequately stabilize a system when encountering rotations of different speeds and in different directions. In addition, along with the assistance of the computing power of a laptop the system was able to track people and objects and perform object detection adequately quickly. It can then read out detections, helping the visually impaired navigate their environment.

Summary statement

I built a camera stabilization system to provide object detection/tracking in real time and a voice readout to help the visually impaired navigate their environment.

Help received

None. I designed and programmed a gimbal and wrote code to detect objects myself.

Awards (1)

Competition history

  • CSEF 2019 Electronics & Electromagnetics · Entry S1018

Resources

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

Source: California Science & Engineering Fair public projects

Save projects to your library

Sign in with Google to keep track of projects you find interesting, organized into folders. An account also raises your daily allowance for “Has this been done?”, and lets you create a key for the MCP server with a much higher limit than anonymous use. Browsing stays public.

Continue with Google