| المستخلص: |
In recent decades, several techniques have been discovered for implementing 3D profilometry. One of these techniques is the use fringe patterns; a technique that is based on projecting fringe patterns on the surface of the object under measurements then capturing the images. Images are analyzed using different approaches that mostly rely on phase calculation techniques, where the captured images are processed in the frequency domain and the phase of the processed images is computed and converted to height distribution. In this research, a new intensity-based technique, which relies on the projection of saw-tooth fringe patterns, has been studied and tested. Simple surface test objects were used for projecting and capturing images taking the surrounding conditions into consideration. In general, different techniques for image enhancement and noise filtering were investigated. The images are then subjected to the proposed intensity based technique to compute the height distribution of the object under test. The study started with a cross-sectional profilometry and then expanded to the whole surface. The obtained results are compared with other existing methods in terms of performance and speed. The proposed technique gives a better root mean square error (RMSE) value when compared with Talebi, Yang and Cao algorithms. However, it gives a higher RMSE value when compared with Furqan algorithm that uses a four step phase shift profilometry (4 PSP) but this method need a high computation. With respect to speed, implementing one line of profilometry using the proposed algorithm gives better speed values than the values obtained using Fourier method. On overall, the proposed technique achieves a good improvement in RMSE values, fewer calculations and a high speed implementation.
|
