DEVELOPMENT OF INSTRUMENTATION SYSTEM FOR NON-CONTACT SURFACE PROFILOMETRY USING DIGITAL FRINGE PROJECTION TECHNIQUES
Measuring the 3-dimensional shapes of objects is important to ensure quality of the objects. The measurement can be done by doing surface profilometry. The methods for surface profilometry are classified as contact and noncontact methods. An example of noncontact method is digital fringe profilometr...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/43137 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Measuring the 3-dimensional shapes of objects is important to ensure quality of the objects. The measurement can be done by doing surface profilometry. The methods for surface profilometry are classified as contact and noncontact methods. An example of noncontact method is digital fringe profilometry.
Digital fringe profilometry utilizes light structured patterns projected to an object. This research develops instrumentation system using digital fringe profilometry. Fringe patterns is generated and projected onto the object’s surface and a set of images is captured by camera obtaining distorted fringe patterns. The images are analysed to extract phase information using Phase Shifting Interferometry. Absolute phase is obtained by phase unwrapping and baseline offset removal. However, the reconstruction is not ideal due to noise. The noise can be suppressed using Butterworth Band-Pass Filter and Mean Curvature Diffusion filter. The developed system is applied to reconstruct and obtain depth of flat and curved objects. System is also implemented to reconstruct surface of animal’s skin.
The result of this research is Graphic User Interface that helps users to set any inputs and get results without doing any calculation. Therefore, the system will be more user-friendly. The system is already calibrated for metric units. The system successfully reconstructs flat and curved objects with error ranged from 0.05 to 0.07. The system is also able to reconstruct four variants of surface of animal’s skin. The reconstruction is analysed by entropy analysis, histogram plot, and variance calculation. |
|---|