Implementation of computer vision algorithms for position correction of chip-mounter machine

Abstract

Chip mounter is a machine in SMT Line that has the task of picking and placing SMD components onto a PCB that has been coated with solder paste. To be able to place SMD components accurately and rapidly onto PCB, it is necessary to have a desciption on the location the chip to be placed on the PCB, the current position of the chip, and the current position of the chip footprint on the board. This paper describes the steps that have been designed and implemented to meet the 3 requirements mentioned by using a digital microscope camera as a downward vision and the use of image processing algorithms as the computer vision feature of the chip-mounter. In the image processing algorithms used here, color thresholding in HSV color model for object selection is incorporated. In order to sharpen the result of object selection, morphological opening and closing are invoked. Canny operator is used to get the edge of the object. Upon contouring the object, the centroid location of the selected object can subsequently be calculated. As shown during the test, the computer vision method implemented in this work is capable of producing data required by the chip-mounter to do position correction if there is non-uniformity errors of the pattern on the PCB panel.