Multiple Subclass Pattern Recognition: A Maximin Correlation Approach

This paper addresses a correlation based nearest neighbor pattern recognition problem where each class is given as a collection of subclass templates. The recognition is performed in two stages. In the first stage the class is determined. Templates for this stage are created using the subclass templates. Assignment into subclasses occurs in the second stage. This two stage approach may be used to accelerate template matching. In particular, the second stage may be omitted when only the class needs to be determined. The authors present a method for optimal aggregation of subclass templates into class templates. For each class, the new template is optimal in that it maximizes the worst case (i.e. minimum) correlation with its subclass templates. An algorithm which solves this maximin optimization problem is presented and its correctness is proved. In addition, test results are provided, indicating that the algorithm's execution time is polynomial in the number of subclass templates. The authors show tight bounds on the maximin correlation. The bounds are functions only of the number of original subclass templates and the minimum element in their correlation matrix. The algorithm is demonstrated on a multifont optical character recognition problem. >

[1]  A. Ardeshir Goshtasby,et al.  A Two-Stage Cross Correlation Approach to Template Matching , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Azriel Rosenfeld,et al.  Two-Stage Template Matching , 1977, IEEE Transactions on Computers.

[3]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[4]  Roberto Cusani,et al.  A correlation based technique for shift, scale, and rotation independent object identification , 1987, ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[5]  S. V. Rice.  A report on the accuracy of OCR devices , 1992 .

[6]  Peter E. Hart,et al.  Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.

[7]  Moncef Mettiji,et al.  Robust multiple classification of known signals in additive noise - An asymptotic weak signal approach , 1993, IEEE Trans. Inf. Theory.

[8]  Philip E. Gill,et al.  Practical optimization , 1981 .

[9]  Hampapuram K. Ramapriyan.  A Multilevel Approach to Sequential Detection of Pictorial Features , 1976, IEEE Transactions on Computers.

[10]  Keinosuke Fukunaga,et al.  An Optimal Global Nearest Neighbor Metric , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  George B. Dantzig,et al.  Linear programming and extensions , 1965 .

[12]  Noriyuki Hiraoka,et al.  Automatic Pattern Recognition System With Self-Learning Algorithm Based On Feature Template Matching , 1986, Other Conferences.

[13]  Theodosios Pavlidis,et al.  On the Recognition of Printed Characters of Any Font and Size , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Julius T. Tou,et al.  Pattern Recognition Principles , 1974 .

[15]  Evangelos E. Milios,et al.  Matching range images of human faces , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[16]  Ching Y. Suen,et al.  Historical review of OCR research and development , 1992, Proc. IEEE.

[17]  G. Dantzig,et al.  COMPLEMENTARY PIVOT THEORY OF MATHEMATICAL PROGRAMMING , 1968 .

[18]  C. McGillem,et al.  Experimental examination of similarity measures and preprocessing methods used for image registration , 1976 .