Vision Based Extraction of Grasping Region of Objects
Many applications of robotics include the grasping and manipulation of objects. Working in assembly robotic
environments, the robot has to accurately not only locate the part but also to recognize it in readiness for grasping. This
paper is mainly focused on the extraction of grasping region that is robust to changes in appearance of objects that have
different shapes, sizes by using predefined rules. The classification performance of Backpropagation Neural Network
(BPNN) classifier based on PCA features for grasping objects are evaluated. The experiments are carried out by using
MATLAB programming language. The proposed system has been tested successfully to a dataset of 200 images for seven
type and achieves good classification accuracy for 2D images. This system can examine types of the hand tools. It can also
be evaluated the size of tool and the locations of the tool at which to grasp the object. This system can be easily adopted to
grasp a number of common hand tools with large, medium and small size such as wrenches, screw drivers, brushes, washers
and nails and hex keys, etc.
Keywords: Appearance, Backpropagation Neural Network, classification, grasping region.