A new segment based, scale invariant, optimized boundary representain algorithm

This paper presents a boundary representation algorithm, the algorithm is developed for Nastaliq characters, but is it general in nature and can be used for boundary representation of any object. The basic idea is that instead of storing and computing the chain code of each pixel, segments of the object are computed by using corner points and a new B-Rep code is developed for the segments not for pixels, that code is scale invariant and much shorter than the chain code. A segment is a portion between two corner points; as the ratio between segments will remain same even when we change the scale of image, so we compute the ratio of each segment to the whole image, and finally compute optimized scale invariant code. Scale invariant code is a code that will remain identical even when we change the scale of an object means when we scale in/out, the output chain code will remain same, so it can be used in object representation and recognition.

[1]  S. A. Husain.  A multi-tier holistic approach for Urdu Nastaliq recognition , 2002 .

[2]  Herbert Freeman,et al.  On the Encoding of Arbitrary Geometric Configurations , 1961, IRE Trans. Electron. Comput..

[3]  Venu Govindaraju,et al.  The Role of Holistic Paradigms in Handwritten Word Recognition , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Kamran Zamanifar,et al.  Multi-Font Farsi/Arabic Isolated Character Recognition Using Chain Codes , 2008 .

[5]  Lele Zhou,et al.  A New Efficient Context-Based Relative-Directional Chain Coding , 2006, 2006 IEEE International Symposium on Signal Processing and Information Technology.

[6]  Michael C. Fairhurst,et al.  A new chain-code quantization approach enabling high performance handwriting recognition based on multi-classi .er schemes , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[7]  Ernesto Bribiesca,et al.  A new chain code , 1999, Pattern Recognit..

[8]  Quintin Gee,et al.  Implementation Challenges for Nastaliq Character Recognition , 2008, IMTIC.

[9]  Vijay Laxmi,et al.  Corner Detection Using Difference Chain Code as Curvature , 2007, 2007 Third International IEEE Conference on Signal-Image Technologies and Internet-Based System.

[10]  Venu Govindaraju,et al.  Offline Arabic handwriting recognition: a survey , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Sarmad Hussain,et al.  Context Sensitive Shape-Substitution in Nastaliq Writing System: Analysis and Formulation , 2007 .