Hitoshi Maekawa

Partial contour processing using curvature function-assembly of jigsaw puzzle and recognition of moving figures

This paper proposes a method of partial contour processing by a curvature function and an adaptive integral filter. The filter parameter is determined by the length of the zero-crossing segment of the curvature function both to enhance the concavo-convex parts of the contour and to eliminate the quantization noise of the raw curvature function. Partial matching of contours is performed precisely and efficiently using the filtered curvature function. Three tasks were accomplished by the abovementioned method: assembly of a jigsaw puzzle, discrimination of overlapping and rotating figures, and tracking of a moving object which overlaps the background figures.

Stereo Method Using Local Direction Histogram for Reference Points Teaching.

In off-line teaching of a robot, motion data is generated on model space in a computer. The data is then to be converted to the coordinate system in real working space. This paper presents a simple stereo method for this coordinate transformation.The stereo method runs as follows. Each stereo image is divided into small rectangle sub-areas for fast feature extraction. In each area, a local direction histogram (LDH) is calculated from intensity gradient. By the shape of LDH, sub-areas which include vertices of the object are selected as feature areas. Then 3-D location of vertices are calculated by stereo-corresponding of these feature areas. The transformation matrix between model space and real working space is obtained by selecting corresponding vertices as reference points from stereo data and from the object model.The method was implemented on a personal computer. A result of simulation experiment of reference points teaching is shown.

Torch trajectory generation along weld line in off-line programming system of arc welding robot.

The authors proposed an off-line programming system which consisted of four levels of software modules hierarchically; i) task level, ii) tool level, iii) robot level, and iv) machine code level. This paper describes the generation of the trajectory for the weld torch on the work model at the tool level.The torch angles specified by the system user along each of weld lines is checked automatically by calculating intersections between planes of the work model and the check plane, which is an imaginary plane defined by the weld line and the axis of torch head. If an intersection beyond the weld line is detected, it is recognized that a corner obstacle is there, and the obstacle is avoided by tilting the torch head along the weld line. If the intersection can be avoided by rotating the check plane, it is recognized as a projection obstacle, and the torch angle is modified with the torch head in the rotated plane. If the intersection still exists, it is recognized as a bridge obstacle above the weld line. When the obstacle is near either end of the weld line, the bridge obstacle is avoided by inserting the torch head with tilting. Otherwise the weld line is divided into two segments at the center of the bridge area.The validity of the algorithm is confirmed using some sample objects with the motion simulation on display monitor.