Evaluation on the Usefulness of Digital Microscope Photogrammetry for the Measurement of small Object
3. Data processing
about data processing. I used VirtuoZo system to be performed at silicon Graphics Indy workstation. About image, I transformed it from BMP format of scanner suitably VirtuoZo Obtaining the 3-dimensional image by moving object may be referred to as inefficient method. Because this is not the desirable method, through relatively oriented models are obtained. All base – line direction movement are related to which is rotation parameter of axis direction generally. Experiment about metal stick, tooth model and artificial skin, elative orientation and absolute orientation and absolute orientation were processed well because of the surrounding control object, but the object, itself had the unsuitable texture. While I could process the data more or less successfully in metal stick and artificial
skin, there was not the effectiveness in the processing of tooth model of people.
out of 4 kinds of objects, concrete specimen extracted the best result. So
in this experiment, I will discuss the concrete specimen mainly.The processing course of data as figure2.
If we get to read the related to the general data processing course, the system of VirtuoZo is processed in order according to the establishment of program menu as follows:
Create Ortho image
Create contour image
Ortho image +contour
Orientation menus are as follows:
To explain the data processing course in the order to designate the menu of VirtuoZo is as follows:
Figure 2. Flow Chart of Data Processing
In the interior orientation course of general air- photo, fiducial mark is used. but in the non metric imagery, the course to fit the index was not necessary.
Relative Orientation :
In the completely automat zed relative orientation solution, many points of about 150points are chosen, and the root mean square error of matching was 0.12mm.
Absolute Orientation :
absolute orientation was made very difficultly. As we should use the control point surely.
Epipolar Image :
I could generated this very easily, as it was do for me to set only the range of zone.
Image Matching :
In this stage, points are matched on whole surface automatically. The result of matching is examined at display /Edit step.And The contour showed even the fine particulars of elaborated refined surface very well.
DTM creation :
I creation DTM at object space at the request set by the parameters defined by work place. Before DTM is the obtained,the difficult
problems of parameter setting showed. As the size of object is nothing but tens of millimeter, I could create DTM by inputting the size of pixel by 10 times than original size after the trial and error over several times.
I overcome most other problems experientially by performing some kinds of experiments. This result will be utilized very efficiently in the similar project to include microscope image. the drape image of DTM created about the destruction side of concrete is as in fig.3
Figure 3. VirtuoZo Drape Image
About the coordinate of 10 DTM points that the position may be decided in image, I decided it with average 0.01mm and root mean square 0.15mmby using MPS-2 analytical plotter, In addition, for the independent evaluation of these coordinates. I showed the coordinate of 10points observed at VirtuoZo by comparing as is in table 1.
Table 1 Comparison of Measurement Value of Coordinate
As the result of measurement, the coordinate measured at VirtuoZo and the coordinate of DTM point observed in MPS-2 analytical plotter showing almost consistent tendency. So I could prove the reliability of this experiment. In the combination of film and scaner that I tried like this, I could obtain the proper resolution with low expense. So it will be able to be utilized usefully for the interpretation of small object.
As the result that I interpreted the magnifying microscope photo about small object with digital photogrammetry method by acquiring it, I obtained the following conclusion
I could reduce the function dependency to be based on the skill of workers which was caused at the time of existing analytical photogrammetry greatly by processing a series of courses such as the generation of interior, relative, absolute, contour map, close inspection photo and DTM etc. promptly and consistently.
As for the magnifying photographing is very inferior in comparison with existing photogrammetry, as it is difficult to maintain the proper base to object distance. Thus, I could improve the geometrical condition of photographing after tiling object.
As for the result that I measured the coordinates about optional 10 points with Virtuozo and MPS-2 analytical plotter is showing almost consistent tendency, I could confirm the reliability of this findings.
As the combination of film camera and scanner was simple and efficient and it can obtain the proper resolution with low expense, the application is expected in the future also.
This study is a part of subject performed under the support of oversea research study for Post -Doc. During the first half period of 1996 by Korea Science & Engineering Foundation. And I wish to express my deep gratitude for the said foundation which supported this research study. In addition, I give thanks to professor G. Fryer and Kerry L. Mcintosh at the university of Newcastle in Australia who didn’t spare the devoted care, so that I might not feel the inconvenience about research during the period for research study.
Won-Jin Oh, Harvey L. Metchell & Eric Kniest, 1998, Experience with microscope digital photogrammetry, International Archives of Photogrammetry and remote Sensing, Vol., 32, part5, Hakodate, pp, 128-131.
Chadwick, R.G., 1992, Close range photogrammetry-ac clinical dental research tool, J. of Dentistry, 20:235-239.
Diamond, S. & Mindness, S., 1992, SEM investigation of fracture surfaces using stereo pairs :I:Fracture surfaces of rock and cement paste,Cement and Concrete Research, 22:67-78.
Jacobsen, K., 1992, Surfaces determination with an accuracy of a few microns, Int. Arch. Photogrammetry & Remote Sensing, XXIX, B5:628:632.
Layton, E. & Cox, P.H. 1982, The application of photogrammetry to quantitative microscopy, Photogrammetria, 37(6):203-218.
Mitchell, H.L., Chadwick, R.G. & McCabe, J.F.,1989, Stereomicroscopic photogrammetry for the measurement of small object, Photogrammetric Record, 13(74):289-299.1