A critical view of pyramid segmentation algorithms

Abstract Although pyramid segmentation algorithms have been reported to be successful in a large number of fields, some of their basic properties have not been systematically investigated. They were especially used in applications involving compact objects, in which moreover the connectivity of the final segments was of no great importance. We will show that most pyramid segmentation algorithms are shift-, rotation- and scale-variant and hence that the results of a pyramid segmentation are very data-dependent and not reproducible. We will demonstrate that the fundamental reason for this shortcoming is the subsampling introduced in the higher levels of the pyramid. We will also discuss why multi-resolution algorithms in general have a fundamental and inherent difficulty in analyzing elongated objects and ensuring connectivity. Finally, we will discuss the way in which a new pyramid algorithm using irregular tessellations, reduces some of those problems but introduces some new ones.

[1]  Alexander Toet,et al.  Image Analysis -Topological Methods , 1984 .

[2]  Charles R. Dyer,et al.  An analysis of node linking in overlapped pyramids , 1984, IEEE Transactions on Systems, Man, and Cybernetics.

[3]  Azriel Rosenfeld,et al.  Segmentation and Estimation of Image Region Properties through Cooperative Hierarchial Computation , 1981, IEEE Transactions on Systems, Man, and Cybernetics.

[4]  Shimon Edelman,et al.  Line connectivity algorithms for an asynchronous pyramid computer , 1987, Comput. Vis. Graph. Image Process..

[5]  Walter G. Kropatsch,et al.  A pyramid that grows by powers of 2 , 1985, Pattern Recognit. Lett..

[6]  John M. Gauch,et al.  Interactive 2D And 3D Object Definition In Medical Images Based On Multiresolution Image Descriptions , 1988, Medical Imaging.

[7]  Azriel Rosenfeld,et al.  Hierarchical Image Analysis Using Irregular Tessellations , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  Michael Shneier,et al.  Extracting Compact Objects Using Linked Pyramids , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Theodosios Pavlidis,et al.  A hierarchical data structure for picture processing , 1975 .

[10]  Ze-Nian Li,et al.  Pyramid Vision Using Key Features to Integrate Image-Driven Bottom-Up and Model-Driven Top-Down Processes , 1987, IEEE Transactions on Systems, Man, and Cybernetics.

[11]  Azriel Rosenfeld,et al.  Boundary localization in an image pyramid , 1986, Pattern Recognit..

[12]  Michael Shneier.  Extracting linear features from images using pyramids , 1980 .

[13]  Azriel Rosenfeld,et al.  Multiresolution object detection and delineation , 1987 .

[14]  Ramesh C. Jain,et al.  A Pyramid-Based Approach to Segmentation Applied to Region Matching , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  S. Pizer,et al.  An Image Description for Object Definition, Based on Extremal Regions in the Stack , 1986 .

[16]  Azriel Rosenfeld,et al.  Frequency Domain Analysis and Synthesis of Image Pyramid Generating Kernels , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  Walter G. Kropatsch.  Curve representations in multiple resolutions , 1987, Pattern Recognit. Lett..

[18]  Charles R. Dyer,et al.  Two-dimensional object recognition using multiresolution models , 1986, Comput. Vis. Graph. Image Process..

[19]  Hendrik James Antonisse.  Image segmentation in pyramids , 1982, Comput. Graph. Image Process..

[20]  Azriel Rosenfeld,et al.  Image Smoothing and Segmentation by Multiresolution Pixel Linking: Further Experiments and Extensions , 1982, IEEE Transactions on Systems, Man, and Cybernetics.

[21]  Peter Meer,et al.  Stochastic image pyramids , 1989, Comput. Vis. Graph. Image Process..

[22]  Leonard Uhr,et al.  A pyramidal approach for the recognition of neurons using key features , 1986, Pattern Recognit..

[23]  Azriel Rosenfeld,et al.  Multiresolution image processing and analysis , 1984 .

[24]  Azriel Rosenfeld,et al.  Compact Region Extraction Using Weighted Pixel Linking in a Pyramid , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.