Algorithms to detect in real time size, position and orientation of rectangular-type electronic components are developed for industrial vision.
The size detection is accomplished via the projection of binary image, its difference and dynamic thresholding method of projection in order to distinguish between the boundary of a component and any marking on the component.
The position detection algorithm is implemented with the use of an image processing window, and is found to finish the task within 400 msec.
The orientation is determined also within 400 msec by Hough transform using simple edge detection method to cover ±25 degree misalignment.
In order to conduct detection of both position and orientation in real time, parallel processing of image data is used by employing two cup's.