Graduation Semester and Year

2012

Language

English

Document Type

Thesis

Degree Name

Master of Science in Electrical Engineering

Department

Electrical Engineering

First Advisor

Michael T Manry

Abstract

An algorithm is presented for the recognition of four types of defects present in silicon wafer images. Defect recognition is achieved by following a 3-step process: segmentation, feature extraction and classification.Multiple image segmentation algorithms are tried for locating and isolating the defects present in the silicon wafer images. The proposed image segmentation technique is based on simple concept of threshold based segmentation and edge detection based segmentation. Combination of four segmentation algorithms based on above mentioned techniques are used such that each segmentation algorithm specializes in segmenting a certain type of defect, thereby ensuring high chances of correct segmentation. Out of these segmented images, the most relevant and distinctive features are extracted and used to train an efficient neural network based classifier. For the standard sized images 2D DFT features are calculated and fed into HWO-MOLF classifier that can determine the type of defect present.Results are presented for all four types of defects.

Disciplines

Electrical and Computer Engineering | Engineering

Comments

Degree granted by The University of Texas at Arlington

Download

Share

COinS