Development of Underwater Object Detection Method Base on Color Feature

• Published in: 2018 International Conference on Computer Engineering, Network and Intelligent Multimedia (CENIM) pp. 254 - 259
• Main Authors: Susanto, Tri, Mardiyanto, Ronny, Purwanto, Djoko
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2018
• Subjects: Accidents; Buoyancy; DC motors; Image color analysis; Mathematical model; navigation; Object detection; ROV; Tracking
• DOI: 10.1109/CENIM.2018.8711290

The aim of this research is to overcome human limitation of ability when they are underwater. Thus, they need certain device to see objects in the water. This study aims to develop detection object system using color detection. The objects will be distinguished based on the selected color. Underwater robot, or commonly known as ROV(remotely operated underwater vehicle), is a kind of robot that is able to perform many activities underwater. The design of this object used ROV as a robot that is able to detect and track objects. The control system in this ROV is based on Arduino Mega, and Raspberry was used as the microprocessor in image processing system, and were processed using Open CV. RGB color detection and HSV color detection were the references in this research. Several samples that were drawn showed the detection error of HSV was 44.83% and RGB was 62.41%. This detection error has little effect on detection, only that the detected range is reduced from the desired result.