Title page for 83322067


[Back to Results | New Search]

Student Number 83322067
Author Chen, Ching Ming(³¯¼yÄP)
Author's Email Address No Public.
Statistics This thesis had been viewed 293 times. Download 10 times.
Department Civil Engineering
Year 1995
Semester 2
Degree Master
Type of Document Master's Thesis
Language zh-TW.Big5 Chinese
Title Edge Extraction and Matching for Digitized Aerial Stereopairs
Date of Defense
Page Count 61
Keyword
  • digital terrain model
  • edge
  • feature matching
  • photogrammetry
  • Abstract Image matching is one of the most important tasks
    in automated        extraction of digital terrain models
    in digital photogrammetry        Area-based matching
    usually works well for the smooth terrain         with
    low relief, especially when low resolution image pairs are
    applied. However, area-based matching is not feasible for high
    resolution image pairs where artificial buildings exist. On the
    other hand, feature matching has better chance to cope with the
    problem. Hence, we propose a scheme which extracts image
    features       first then organizes a feature matching
    strategy to achieve          better matching precision
    in 3-D information extraction. The most       obvious
    objects in high resolution aerial photos are artificial
    buildings where the canopy surface changes drastically. In
    general, artificial structures have sharp edges with distinct
    geometric characteristics. Accordingly, we may use edge
    detection       technique to extract feature lines where
    gray values change          obviously. However, in the
    real nature, different imaging geo-        metry,
    illumination, and texture variations cause the boundary
    extraction very complex. Thus, a preprocessing for selecting
    wanted information is need.The preprocessing procedure includes
    edge thinning, line extraction, edge linking, and polygon extra-
    tion. In this investigation, we develop a new feature-based
    matching scheme, i,e., Azimuth-Moment Matching method, to
    correspond the potential conjugate features. Then using the
    match       ed feature as an initial position, a templet
    matching is applied       Experimental results indicate
    that for a pair of high resolution       aerial
    photographs the proposed scheme may accomplish the task
    that conventional templet matching dose not.
    Table of Content
    Reference
    Advisor
  • L. C. Chen(³¯¨}°·)
  • Files No Any Full Text File.
    Date of Submission

    [Back to Results | New Search]


    Browse | Search All Available ETDs

    If you have dissertation-related questions, please contact with the NCU library extension service section.
    Our service phone is (03)422-7151 Ext. 57407,E-mail is also welcomed.