Reducing False Detection during Inspection of HDD using Super Resolution Image Processing and Deep Learning

Abstract

High false detection rates are a key reliability challenge in the Hard Disk Drive (HDD) industry. Therefore, automatic visual inspection is increasingly employed for HDD inspection. In order to improve the quality and reliability of HDD products, the false detection rate must be reduced. This paper presents a super - resolution image - based method for improving the performance of Head Gimbals Assembly (HGA) inspectio n. The experimental results confirm the efficiency of the super - resolution image processing for improving automatic inspection of defects such as pad burning and micro contaminations. Moreover, combining super resolution image processing with deep learning reduces the false detection rate and improves the accuracy of HGA inspection.