Segmentation of lines and arcs and its application for depth recovery

In this paper we describe an advanced segmentation approach for stereo images improving the computation of depth compared to the commonly used straight line segmentation. Using a straight line-circular arc approximation of chain coded lines, the number of primitives is reduced significantly. This approximation lowers the computational effort as well as the frequency of erroneous matches. Starting with matched pairs of primitives, a disparity image is computed containing the initial disparity values for a subsequent block matching algorithm. The output of this algorithm is the partially dense depth image of one aspect of the object. We describe the result of a parallel implementation using object-oriented programming techniques. In segmentation as well as in matching we evaluate color information to improve accuracy and reliability of the depth values. The algorithms are part of a system computing depth from monocular image sequences. Taking a sequence of different views by a camera mounted to a robot hand, each two consecutive images are considered as a stereo image. The depth images computed from these stereo images are fused to one complete depth map of the object surface. The results show substantial improvements in comparison to a monochrome system with respect to speed, accuracy, and completeness.

[1]  H. Hessenauer,et al.  Architecture and realization of the modular expandable multiprocessor system MEMSY , 1994, Proceedings of the First International Conference on Massively Parallel Computing Systems (MPCS) The Challenges of General-Purpose and Special-Purpose Computing.

[2]  Peter Damaschke.  The linear time recognition of digital arcs , 1995, Pattern Recognit. Lett..

[3]  Ramakant Nevatia,et al.  Segment-based stereo matching , 1985, Comput. Vis. Graph. Image Process..

[4]  Olivier D. Faugeras,et al.  Building Three-Dimensional Object Models from Image Sequences , 1995, Comput. Vis. Image Underst..

[5]  Jack Koplowitz,et al.  Corner detection for chain coded curves , 1995, Pattern Recognit..

[6]  Rudiger Bess,et al.  Registering Depth Maps from Multiple Views Recorded by Color Image Sequences , 1997 .

[7]  D. Paulus,et al.  3D recovery using calibrated active camera , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.

[8]  Stefan Posch,et al.  Automatische Tiefenbestimmung aus Grauwertstereobildern , 1990 .

[9]  Michael Harbeck.  Objektorientierte linienbasierte Segmentierung von Bildern , 1996 .

[10]  Jack Koplowitz,et al.  Corner detection for chain-coded curves , 1993, Other Conferences.

[11]  Allen R. Hanson,et al.  The image understanding environment program , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[12]  Perry S. Plexico,et al.  Data abstraction and object-oriented programming in C++ , 1990 .

[13]  A. Rosenfeld,et al.  Edge and Curve Detection for Visual Scene Analysis , 1971, IEEE Transactions on Computers.

[14]  Franz Leberl,et al.  Trade-Offs in the Reconstruction and Rendering of 3-D Objects , 1994 .

[15]  Joachim Hornegger,et al.  Pattern Recognition and Image Processing in C++ , 1995, Vieweg+Teubner Verlag.

[16]  Heinrich Niemann,et al.  Knowledge Based Image Understanding by Iterative Optimization , 1996, KI.

[17]  Dietrich Paulus,et al.  Objektorientierte und wissensbasierte Bildverarbeitung , 1992, Artificial intelligence - Künstliche Intelligenz.