Coding efficiency of multiring differential chain coding (line drawing transmission)

A new coding scheme for line drawings, multiring differential chain coding (MRDCC) is introduced, based on relative vectors drawn from multiple coding rings. The performance of MRDCC depends on the probability distribution of the different rings included in the code. A model is presented for the evaluation of ring probabilities; these have been used, together with the corresponding relative vector probabilities, to obtain the coding rate. Compared to the standard chain code (SCC), generalised chain code (GCC) and differential chain code (DCC), MRDCC is expected to reduce the coding rate by 40%, 53% and 13%, respectively. Experiments performed show an average reduction in coding rate by 50%, 23% and 5% compared to SCC, GCC and DCC, respectively.

[1]  John A. Saghri,et al.  Analysis of the Precision of Generalized Chain Codes for the Representation of Planar Curves , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  H. Freeman,et al.  Comparative analysis of line-drawing modeling schemes , 1980 .

[3]  Ramjee Prasad,et al.  Multiring differential chain codes for line drawings , 1990 .

[4]  David L. Neuhoff,et al.  A rate and distortion analysis of chain codes for line drawings , 1985, IEEE Trans. Inf. Theory.

[5]  R. Prasad,et al.  Relative vector probabilities in differential chain-coded line-drawings , 1989, Conference Proceeding IEEE Pacific Rim Conference on Communications, Computers and Signal Processing.

[6]  F. Groen,et al.  Freeman-code probabilities of object boundary quantized contours , 1978 .