Many CO-processors which are designed for learning of selforganizing maps (SOM) have been proposed in order to re--. . duce the processing time [9]:[14]. However, hardware in which-all p6cesses of the learning of the SOM are achieved is not realized, because it needs many complex cdcufations. In this shdy, a new le&ng algorithm of the SOM in which input vecton_&d.weight vectors are represented as binary data is proposed. The effectiveness of the proposed algorithm is verified by designing the digital hardware of the proposed algorithm using HDL: ---~ .~

No abstract

No abstract

We propose a new measurement method for a degree of roughness of a given object surface. This method is not to measure the degree of roughness of the object surface directly, but to estimate the roughness from surface images. Named as circulating light sources (CLS), its multiple light sources aligned in a circle illuminate sequentially, and produce images including shadow of the object surface. As the shadows on the images reflect a shape of the object surface, the shapes of the surface, concavo-convex shape, can be estimated by these shadows. In this paper, features of surface roughness are extracted by a Wavelet Multiresolution Analysis (MRA) from the shadow images produced by the CLS, and are classified by a Self-Organizing Map (SOM). A roughness of an unknown surface can be estimated by the SOM after learning.

Genetic algorithm (GA) are search techniques used in computing to find true or approximate solutions to optimization and search problems. GA are a particular class of evolutionary algorithms that use techniques inspired by evolutionary biology such as inheritance, selection, crossover and mutation. GA are categorized as global search heuristics. Hardware accelerators for GA are required to reduce the execution time. Therefore, many research results for hardware implementation of GA have been reported. In the hardware implementation of GA, a circuit design of roulette wheel selection influences the performance of the GA hardware. This paper propose a new selection circuit based on Rough Comparison Method (RCM). The RCM selects the larger data included within a certain definite range as selection candidates (SC). Figure 1 illustrates an example of the RCM. In the data (D 0 ~ D 5), the high-order several bits which are defined Range are used for data comparison. The other bits are considered as a rounding error. When the high-order four bits are compared (i.e. Range = 4), the values of D 0 , D 4 and D 5 are 112. Thus, D 1 , D 2