Drones are vehicles that can be controlled by long-distance pilots. Drones are widely used by many companies to facilitate jobs that are difficult to achieve by humans. This research will apply image processing segmentation on quadcopter to avoid obstacles without manual control (automatic). Digital image processing will use two methods, namely canny edge detection and hough transform. This research will be use a quadcopter of type Parrot Ar.Drone 2.0. The results of the tests conducted in this study will get the accuracy of the quadcopter by using the distance from the width of the obstacle. Testing based on accuracy (errors) obtained from the purpose of this system is 0.07125%. The maximum detection distance on a quadcopter that can be detected is when the width of the obstacle has reached 294 pixels and 142 pixels. And for research time estimation will get an average quadcopter travel time of 7.464857 seconds.
Copyrights © 2019