| dc.identifier.citation | Sural, S., Qian, G., Pramanik, S. 2002. Segmentation and histogram generation using the HSV color space for image retrieval.Proceedings of IEEE International Conference on Image Processing. 589-592. Marchand, E. 2007. Control Camera and Light Source Positions using Image Gradient Information. IEEE Int. Conf. on Robotics and Automation. 417 – 422. Evan, Y. 2010. Thresholding citra, kuliahinformatika.wordpress.com Sigit, H., & Agung, T. 2010. Image processing dasar, kliktedy.wordpress.com Kragic, D., Christensen, H.I. 2011. Survey on Visual Servoing for Manipulation: Centre forAutonomous Systems. Numerical Analysis and ComputerScience. Ikwuagu, E. 2011. Design Of An Image Processing Algorithm ForBall Detection. Computing Research Association. Brigida, A.2012. Transformasi hough, informatika.web.id. Yustinus, P. 2012. Rancang Bangun Aplikasi Pendeteksi Bentuk Dan Warna Benda Pada Mobile Robot Berbasis Webcam. Academia. Agus, K. 2015. Convert RGB To HSV for color tracking in Raspberry and openCV, ilmu-otomasi.blogspot.com. | in_ID |
| dc.description.abstract | This paper presents the research results on using the object tracking techniques for object recognition and robot's motion control based on object movement. In this study, object tracking used by the robot to recognize and follow the movement of an object. Pattern recognition and robot motion in object tracking controlare done in a real time. Robot developed in this study is a wheeled robot that has the ability to follow the motion of the certain color ball. Pattern recognition algorithms in this study includes: the color space conversion from RGB to HSV, color detection using color filtering, and shape detection using the edge detection technique and the circle hough transform. Image processing and robot‟s motion control is done by using a mini computer Raspberry Pi 2 Model B Raspberry Pi and camera. This study use python programming language, which has been embedded in OpenCV library. The performance tests in this research was conducted by analyzing the effect of the image resolution on the camera motion, the robot maximum visibility, the color recognition process, the shape recognition process, and the minimum light intensity requirement. The results show that the best image resolution for the robot to perform ball tracking is 320x240 pixels, the robot maximum visibility is 113 cm, and the minimum of light intensity for robot to recognize the ball is 21.0 lux. | in_ID |
