Feature Extraction Without Edge Detection

Title: en_US Feature Extraction Without Edge Detection
Creator: en_US Chaney, Ronald D.
Date: 2004-10-20T19:55:16Z
Date Available: 2004-10-20T19:55:16Z
Date Issued: en_US 1993-09-01
Identifier: en_US AITR-1434
uri: http://hdl.handle.net/1721.1/6794
Abstract: en_US Information representation is a critical issue in machine vision. The representation strategy in the primitive stages of a vision system has enormous implications for the performance in subsequent stages. Existing feature extraction paradigms, like edge detection, provide sparse and unreliable representations of the image information. In this thesis, we propose a novel feature extraction paradigm. The features consist of salient, simple parts of regions bounded by zero-crossings. The features are dense, stable, and robust. The primary advantage of the features is that they have abstract geometric attributes pertaining to their size and shape. To demonstrate the utility of the feature extraction paradigm, we apply it to passive navigation. We argue that the paradigm is applicable to other early vision problems.
Extent: en_US 159 p.; 1640697 bytes; 2318330 bytes
Format: application/octet-stream; application/pdf
Language: en_US
Relation: en_US AITR-1434
Subject: en_US feature extraction; en_US structure from motion; en_US edge detection; en_US spassive navigation