In many vision tasks one of the major obstacles is the specular highlight of smooth objects, which causes a misinterpretation of objects.
The thesis presents an efficient algorithm for highlight detection and object reconstruction, based on the theo...
In many vision tasks one of the major obstacles is the specular highlight of smooth objects, which causes a misinterpretation of objects.
The thesis presents an efficient algorithm for highlight detection and object reconstruction, based on the theory of photometric stereo in which th location of highlight changes as the position of illumination source changes. Two images, referred to as base image and reference image, are sequentially taken with two different position of the illumination source but from the same viewing detection. The reference image is normalized by the average of the two images. The difference image is thresholded to detect the specular spike is detected to reconstruct the object.
The proposed algorithm can be applied to metals and dielectrics, regardless of the characteristics. This method can also be applied to the case when the background is brighter than the object.