Detecting orientation discontinuities in range images by use of directional derivatives

The author exposes a step of the segmentation process of range images. Based upon an analogy between step edges and orientation discontinuities, it is proposed to detect surface orientation discontinuities by extrema of second directional derivatives or by zero crossings of third directional derivatives. A method is presented to compute derivatives up to the fourth order. Results obtained with these methods are shown for real range images.<<ETX>>

[1]  Rachid Deriche,et al.  Using Canny's criteria to derive a recursively implemented optimal edge detector , 1987, International Journal of Computer Vision.

[2]  Robert M. Haralick,et al.  Digital Step Edges from Zero Crossing of Second Directional Derivatives , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Anil K. Jain,et al.  Segmentation and Classification of Range Images , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Paul J. Besl,et al.  Surfaces in Range Image Understanding , 1988, Springer Series in Perception Engineering.

[5]  Jean Ponce,et al.  Toward a surface primal sketch , 1985, Proceedings. 1985 IEEE International Conference on Robotics and Automation.

[6]  Andre Davignon.  Contribution of edges and regions to range image segmentation , 1992, Defense, Security, and Sensing.

[7]  Bahram Parvin,et al.  Adaptive multiscale feature extraction from range data , 1989, Comput. Vis. Graph. Image Process..

[8]  Rachid Deriche.  Fast Algorithms for Low-Level Vision , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Tony Kasvand,et al.  The k1k2 space in range image analysis , 1988, [1988 Proceedings] 9th International Conference on Pattern Recognition.

[10]  Jun Shen,et al.  An optimal linear operator for step edge detection , 1992, CVGIP Graph. Model. Image Process..

[11]  Ramakant Nevatia,et al.  Segmented descriptions of 3-D surfaces , 1987, IEEE Journal on Robotics and Automation.