Perancangan Object Tracking Robot Berbasis Image Processing Menggunakan Raspberry Pi

• Main Authors: Prianggodo, Laksono Budi (Author), , Dr. Ratnasari Nur Rohmah, ST. MT (Author)
• Format: Book
• Published: 2016.
• Subjects: Thesis; NonPeerReviewed
• Online Access: Connect to this object online
• Notes: https://eprints.ums.ac.id/44456/1/NASKAH%20PUBLIKASI.pdf; https://eprints.ums.ac.id/44456/2/SURAT%20PERNYATAAN%20PUBLIKASI%20ILMIAH.pdf
• Keywords: TK Electrical engineering. Electronics Nuclear engineering

Object tracking is a technique in computer vision field that used to tracking on a moving object. Object tracking is done by image processing techniques through a combination of complex algorithm. This process gives the computer ability to know the movement of particular object. This study designs a robot that can follow the movement of a ball with diameter is 6.5 cm with particular color. Image processing in this study is convertion of RGB to HSV color space, color filtering, edge detector, and circle hough transform. Image processing and motion control of robot using Raspberry Pi 2 Model B mini computer and Raspberry Pi camera. Language programming that use is python with OpenCV library. OpenCV library is used to do all things related for image processing. The mechanical system using a pan tilt as camera driver to get flexibility in following the movement of ball. Robot's body move using a pair of DC motors. The test that have been done is influence the resolution of picture on the camera motion system, robot visibility, color recognition, ball's shape recognition, the speed when tracking, and minimum light intensity. The result show that robot can tracking a ball using the best image resolution in 320x240 pixel, it has a maximum visibility is 113 cm, maximum speed of object follow the movement is 22.6 cm/sec and it can recognize ball in minimum light intensity is 21.0 lux.