53IV. CONCLUDING REMARKS The objective of this brief was to extend the CNN paradigm to multilevel halftoning of digital images. The primary contributions of the brief include using the CNN transient mode, one of the general working modes of the CNN [6], for multilevel halftoning and developing a stopping criterion for selection of the network output. Potential applications of CNN based multilevel halftoning include image preprocessing, e.g., enhancement or segmentation, and image compression.The proposed system lends itself to real-time processing and simple VLSI realization. Moreover, an analog programmable computing machine [7] could accept CNN based multilevel halftoning as the algorithmic step while perform more complex image processing tasks. Real-time realization of VMSE measure would allow utilization of the dynamic stopping criterion in applications that require fast processing. We are presently developing methods of realization of the CNN for multilevel halftoning using the multilayer [1], modularized, or programmable analogic [7] CNN.Abstract-Residue number systems have computational advantages for addition and multiplication since operations on residue digits are performed independently and so these processes can be performed in parallel. However other operations such as input/output conversions are significantly more difficult. A method for conversion from a specific residue number system with moduli of the form (2 k 0 1; 2 k ; 2 k + 1) to a weighted number system is presented here. The digit parallel method is significant, in that the largest number which must be handled is of the same order as the moduli, the digits of the result are calculated in parallel, and the required moduli operations are accomplished with addition or subtraction of a constant.