Adaptive lifting integer wavelet transform for lossless image compression

This paper presents an adaptive lifting scheme for integer-to-integer wavelet transform, and its performance on lossless compression of digital images. We optimize the coefficients of the predict filter to minimize the predictor error variance for every image. The optimized coefficient depends on the variance-normalized autocorrelation function of the image. The proposed lifting scheme adapts not only to every image but also to its horizontal and vertical directions. Experimental results are obtained using different types of images. These results show that the proposed method is competitive to few well-known methods for lossless image compression, in terms of compression ratio and computational efficiency.