Vector quantization has been shown to be an effective technique for image compression. In this
paper, an unsupervised parallel approach called the Fuzzy Competitive Learning Network (FCLN) for
vector quatization in image compression is proposed. The goal is to apply an unsupervised scheme
based on a neural network using the fuzzy clustering technique so that on-line learning and parallel
implementation for codebook design are feasible. In FCLN, the codebook design is conceptually
considered as a clustering problem. Here, it is a kind of neural network model imposed by the fuzzy
clustering strategy working toward minimizing an objective function defined as the average distortion
measure between any two training vectors within the same class. For an image of n training vectors and
c interesting objects, the proposed FCLN would consist of n input and c output neurons. The
experimental results show that a promising codebook can be obtained using the fuzzy competitive
learning neural network based on least squares criteria in comparison with the generalized Lloyd
algorithm.
