Rough surfaces can be classified by the point statistics of their derivative fields, estimated using photometric stereo. Such a scheme is proposed and found to be more accurate and robust than image-intensity-based classification. It is particularly effective when applied to directional surfaces, even under rotation. The scheme is therefore robust and economic - suitable for many applications and worthy of further investigation. © 2000 Elsevier Science B.V.
|Number of pages||12|
|Journal||Pattern Recognition Letters|
|Publication status||Published - Jun 2000|
- Photometric stereo
- Rotation invariance
- Surface classification
- Texture analysis