A new initialization technique for VQ codebook design

The generalized Lloyd algorithm plays an important role in the design of vector quantizers (VQ) and in feature clustering for pattern recognition. In the VQ context, this algorithm provides a procedure to iteratively improve a codebook and results in a local minimum which minimizes the average distortion function. We propose an efficient method to obtain a good initial codebook that can accelerate the convergence of the generalized Lloyd algorithm, achieve a better local minimum and also result into lower entropy. Thus, it can achieve an overall better performance since the resulting rate-distortion pair is closer to the rate-distortion function of the input source