Object Detection Based On Plane Segmentation And Features Matching For A Service Robot

Abstract

With the aging of the world population and the<br> continuous growth in technology, service robots are more and more<br> explored nowadays as alternatives to healthcare givers or personal<br> assistants for the elderly or disabled people. Any service robot<br> should be capable of interacting with the human companion, receive<br> commands, navigate through the environment, either known or<br> unknown, and recognize objects. This paper proposes an approach<br> for object recognition based on the use of depth information and<br> color images for a service robot. We present a study on two of the<br> most used methods for object detection, where 3D data is used to<br> detect the position of objects to classify that are found on horizontal<br> surfaces. Since most of the objects of interest accessible for service<br> robots are on these surfaces, the proposed 3D segmentation reduces<br> the processing time and simplifies the scene for object recognition.<br> The first approach for object recognition is based on color histograms,<br> while the second is based on the use of the SIFT and SURF feature<br> descriptors. We present comparative experimental results obtained<br> with a real service robot.