Structural shape recognition in a multiresolution environment

Abstract A new structural approach for shape recognition in a multiresolution environment is illustrated. It can be divided into several steps: in the first a local classification of the instances of the detected contour is performed, considering a fixed number of paradigm patterns. Following a bottom-up procedure based on the generalized Hough technique, the local features are analyzed to look for greater aggregates of the recognized sub-shapes. The implementations of the process on an array processor SIMD-machine and on a pyramid machine are considered. In the latter case an improvement of the performances is obtained in several practical cases by adopting planning strategies on hierarchies and multiresolution facilities. Some experimental results on artificial images are finally discussed.

[1]  B. Zavidovique,et al.  Intégration d'un processeur cellulaire pour une architecture pyramidale de traitement d'image , 1985 .

[2]  Murray H. Loew,et al.  A new method for characterization of shape , 1985, Pattern Recognit. Lett..

[3]  Dana H. Ballard,et al.  Parameter Nets , 1984, Artif. Intell..

[4]  Michael J. B. Duff.  Pyramids—expected performance , 1986 .

[5]  Henry Lieberman,et al.  Design Issues in Parallel Architectures for Artificial Intelligence , 1983, COMPCON.

[6]  Azriel Rosenfeld.  The prism machine: An alternative to the pyramid , 1985, J. Parallel Distributed Comput..

[7]  D. Ballard,et al.  Experience with the Generalized Hough Transform , 1980 .

[8]  King-Sun Fu,et al.  Attributed Grammar-A Tool for Combining Syntactic and Statistical Approaches to Pattern Recognition , 1980, IEEE Transactions on Systems, Man, and Cybernetics.

[9]  Richard A. Volz,et al.  Recognizing Partially Occluded Parts , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[10]  Renato Stefanelli,et al.  PAPIA: Pyramidal architecture for parallel image analysis , 1985, 1985 IEEE 7th Symposium on Computer Arithmetic (ARITH).

[11]  Larry S. Davis,et al.  Hierarchical generalized Hough transforms and line-segment based generalized H ugh transforms , 1982, Pattern Recognit..

[12]  Dana H. Ballard,et al.  Generalizing the Hough transform to detect arbitrary shapes , 1981, Pattern Recognit..

[13]  Herbert Freeman.  Lines, Curves, and the Characterization of Shape , 1980, IFIP Congress.

[14]  King-Sun Fu,et al.  Hybrid Approaches to Pattern Recognition , 1982 .

[15]  Steven L. Tanimoto,et al.  A hierarchical cellular logic for pyramid computers , 1984, J. Parallel Distributed Comput..

[16]  Herbert Freeman,et al.  Shape description via the use of critical points , 1978, Pattern Recognit..