Abstract:
In the hard disk drive (HDD), the orientation of the head gimbal assembly (HGA) prior attach to the actuator arm is considerably. The orientation is subjected to achieve either detect in pick and place activity. This paper deals with machine vision technology that apply the invariant moment technique that using the geometric invariant feature extraction to produce the simulation to detect whether HGA is in a
good orientation. Pre-processing of image acquisition perform on the HGA image from CCD camera. Intensity thresholding corresponding perform with Gaussian filtering in order to eliminate noise reduction, then, morphological operation perform the feature extraction of the HGA image. After image acquisition, invariant moment that has system for a scale, translation and orientation are calculated for each significant region in the input images. The calculated output will compare against the good orientation that are considered. Results show that, however, the invariant moment may not vary to several angle rotations with no significant different. In additional, the position can be obtained with good orientation. Finally, this novel method can improve the hard disk drive assembly process productivity and yield gain.