Files in this item
|(no description provided)|
|Title:||Video metrology for documentation of engineering construction|
|Author(s):||Obaidat, Mohammed Taleb Mefleh|
|Doctoral Committee Chair(s):||Wong, Kam W.|
|Department / Program:||Civil and Environmental Engineering|
|Discipline:||Civil and Environmental Engineering|
|Degree Granting Institution:||University of Illinois at Urbana-Champaign|
|Abstract:||A practical method was developed to calibrate the interior geometry of CCD cameras. The method employs a flat brick wall as a control field. A brick wall is an excellent control field, because it provides a large number of well-defined points at almost any focal setting. When compared with laboratory calibration, which requires a three-dimensional control field, the method of the planar constraint was capable of providing results of comparable accuracy. A RMS error of better than 0.2 pixels in the image residuals was achieved consistently for various focal settings. Using about 50 stereo image points, the interior geometry of a CCD camera was effectively modeled. Increasing the number of image points could increase the accuracy of the calibration parameters.
A knowledge-based system was developed to assist inexperienced users in data processing and analysis. It could perform the following tasks: (1) check the validity of the input data; (2) provide guidance during failure modes; (3) perform robust blunder detection; and (4) assist in interpretation of results and decision making. Therefore, the result was a vision system that could be used productively without any in-depth knowledge about photogrammetry.
A rigorous statistical analysis scheme based on error propagation of the image coordinates was implemented to automatically evaluate the accuracy of the extracted 3-D measurements. Test results showed that the calculated errors consistently fell within 3 times the estimated standard errors.
Experimental results showed that an accuracy of $\pm$1 pixel on the image plane was achieved for object space coordinates. Lower measurement accuracy in the range of 4-5 pixels was obtained for the depth direction because of the intersection geometry and accuracy limitation in manual image matching. Comparison with actual survey measurements showed that distances could be measured with an accuracy of 2 pixels (at the image scale), and volume and surface area were measured to within 3%. Image scale, base/object distance ratio, number and distribution of control points, and accuracy limitation in manual matching had significant impact on the measurement accuracy.
Although this research was intended for documentation of engineering construction, its basic methodology can find numerous applications in many other fields, including: manufacturing, gaging, bioengineering, archeology, crime and accident investigation, robot control positioning, and IVHS (Intelligent Vehicle Highway Systems).
|Rights Information:||Copyright 1994 Obaidat, Mohammed Taleb Mefleh|
|Date Available in IDEALS:||2011-05-07|
|Identifier in Online Catalog:||AAI9416418|
This item appears in the following Collection(s)
Graduate Dissertations and Theses at Illinois
Graduate Theses and Dissertations at Illinois
Dissertations and Theses - Civil and Environmental Engineering