Heuristic procedure for segmentation of 3-D range images

A heuristic procedure for the segmentation of 3-D range images is presented. The range image is coarsely segmented into convex, concave, and flat patches using curvature sign labelling at each surface point. The jump edges are then imposed on the coarsely segmented image, resulting in the initial image segmentation. Next, the boundary points between adjacent surface patches are examined for the purpose of merging compatible patches. A boundary point between patches, which is not on a jump edge, has to belong to a crease or to a ridge line. This is detected by the presence of a large local norm of the residuals for an approximating plane on a window surrounding the boundary point. The present method is expected to be faster than many existing methods. Two examples are used in illustration. Comments and conclusions are given.

[1]  Luther Pfahler Eisenhart,et al.  Introduction to Differential Geometry , 2015 .

[2]  Paul Beaudet,et al.  Rotationally invariant image operators , 1978 .

[3]  Ramesh C. Jain,et al.  Invariant surface characteristics for 3D object recognition in range images , 1985, Comput. Vis. Graph. Image Process..

[4]  Takeo Kanade,et al.  Three-Dimensional Machine Vision , 1987 .

[5]  Azriel Rosenfeld,et al.  Noise reduction in three-dimensional digital images , 1984, Pattern Recognit..

[6]  N. Abdelmalek.  A computer program for the Chebyshev solution of overdetermined systems of linear equations , 1976 .

[7]  Francois Blais,et al.  Real-time correction of three-dimensional nonlinearities for a laser rangefinder , 1986 .

[8]  Ernest L. Hall,et al.  Computer Image Processing and Recognition , 1980, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Azriel Rosenfeld,et al.  Picture Processing and Psychopictorics , 1970 .

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

[11]  W. Eric L. Grimson,et al.  Discontinuity detection for visual surface reconstruction , 1985, Comput. Vis. Graph. Image Process..

[12]  M Rioux,et al.  Laser range finder based on synchronized scanners. , 1984, Applied optics.

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