PCA-based hard disk media defect classification

• المؤلف الرئيسي: Zhang, Jian Liang
• مؤلفون آخرون: Mao Kezhi
• التنسيق: Theses and Dissertations
• اللغة: English
• منشور في: 2009
• الموضوعات: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• الوصول للمادة أونلاين: http://hdl.handle.net/10356/18760
• المؤسسة: Nanyang Technological University

Nowadays people are more familiar with hard disks. We use them everyday to save our photos, videos, writings, etc. Hard disk media defect classification is very important for hard disk failure analysis. Through failure analysis we can get the rood of failure, and furthermore improve the quality of hard disks. This dissertation will show the design of a classification system, which automatically classifies images of hard disk media defects. The design is based on principal component analysis (PCA). PCA is a common statistical technique for finding patterns in data of high dimension, and has found application in field such as face recognition and image compression. The design system is evaluated based on 640 defect images. An acceptable result is achieved. Comparisons for different feature selection and different classifiers are also showed in the dissertation.