Optimized Two Dimensional Wavelet Filter from BAT Algorithm

Abstract

Due to the effect of a quantization error, it is not possible to fully restore the original image in the lossy wavelet-based image compression. However, the quantization error can be minimized by optimizing or evolving the filter bank. In this work, the coefficients of the standard wavelet filter and its inverse filter were optimized by evolution of Bat algorithm. The optimized wavelet filters were used with the SPIHT encoder/decoder for image compression. The performance of the optimized filters in reconstruction during the decompression process was investigated using the metrics PSNR, MSE and SSIM. The results obtained show that the proposed filter outperforms standard wavelet filters by minimizing the error between the original and the decompressed image.