A filter banks design using a multiobjecive genetic algorithm for an image coding scheme
ICIP'09 Proceedings of the 16th IEEE international conference on Image processing
| Hi-index | 0.00 |
In this paper we present a new optimization-based method for designing filter banks in subband image coding. We formulate the design problem as a nonlinear optimization problem whose objective consists of both the performance metrics of the image coder, such as the peak signal to noise ratio (PSNR), and those of individual filters. In contrast to previous methods that design filter banks separately from the other operations in image coding, our formulation allows us to search for the filters in the context of an image coder to maximize coding quality. Due to the nonlinear nature of the performance metrics, the optimization problem is solved by using simulating annealing. In our method, we first apply the optimization method to find good filter banks for individual training image and then select the one that performs best across all training images to be the final solution. In experimental results, we show that the filter bank designed by our method improves the coding quality of the best existing filter bank in terms of PSNR %and visual quality on nine benchmark images.