Detection of micro contamination in hard disk drives using angle measurements and Bayesian classification

Abstract

Micro Contamination has been a major critical defects peculiar to the hard disk drive assembly particularly to head gimbal assembly compartment and it occurs mainly on the airbearing surface. Consequently, reliable measures to improve the read/write process in a defect of this nature requires an inspection process to check and verify the hard disk components. Therefore, we proposed an image-based automatic inspection method for micro-contamination detection. This proposed inspection approach leverage on angle measurements and Bayesian classification for identifying detects. However, a comparative analysis between this technique and previous angle measurement and intensity threshold methods was carried out to ascertain the performance improvement. The experimental results showed that this method is robust and outperformed the existing methods in literature. Meanwhile, the improvement recorded showed that this method offers a minimal false detection rate for real-time practical applications.