Detection of Object by Robot using Image Processing

Abstract

In this paper demonstration is carried out on an anthropomorphic assist- stance robot as part of a multi- modal man-machine interaction. In this paper, by detecting hand position the robot recognizes the object and work will be carried out as per the human being. When there are many applications are used for optimized image processing methods for detection of human hand due to the camera which have been used or developed. In this hand can be detected by due to its movement in the image, where remaining image is static. Gives fast tracking. In this also Bird’s eye view the Bird’s eye view is correlated with the blurred copy of scene. So that verification takes place by inverse correlation indicate position of hand in the image. In this rotation steps used are having degrees in between 15 and 30.so that arm of robot will rotate in this degrees and it can detect the object which come under the cameras which are used at the top of robot. Mainly in this hand detection takes place, after that skin color segmentation takes place. position of hand given by stereo based localization and then tracking of object and then object recognition takes place. So once object is selected that object is used for further processing.