No abstract

An index is presented for quantifying the geometric complexity of a three-dimensional solid model. This provides a measure by which components may be compared one with another in relation to their relative complexity. The index is alphanumeric and readily computable. Such an index can be of use in the field of feature recognition as a means to determine how efficiently one algorithm handles components of varying complexity compared to another such algorithm.The performance of the authors' own feature recognition algorithm is tested against components of differing complexity as determined by the index.

No abstract

The backpropagation neural network learning algorithm is generalized to include complex-valued interconnections for possible optical implementations.

No abstract