An Image-Processing-Based System for Object Detection

Abstract

Understanding how to recognize and track a moving object in real time is essential in the field of computer vision. Underwater computer vision can collect important data that might be used in a wide variety of practical applications. This concept is employed for surveillance, allowing us to keep tabs on the military installation, manage traffic, and coordinate with submerged devices to save lives. The robot’s position will be adjusted to the left, right, front, and rear depending on where the item is detected and identified to be moving in this project. In this way, the robot’s safe distance from its intended victim is never compromised. Hardware-wise, we're using an Arm11 Raspberry Pi, a picamera for Arduino mounting, and an Android device for tracking the robot’s movements. Python code running on Linux controls the pan and tilt camera’s ability to provide an object description through open cv.