| Summary: | 碩士 === 國立成功大學 === 資訊工程學系碩博士班 === 98 === Novel 5/3 and 9/7 wavelet-based image interpolations in lifting structures with/without involving the evolutionary programming (EP) for image compression are proposed in this paper. Based on the low-band wavelet coefficients, the proposed prediction algorithms estimate high-frequency wavelet coefficients of the original image, so the original image can be reconstructed from the low-band image by using the proposed prediction algorithms. Due to the characteristics of the odd-length wavelet filter which is dominated by the intermediate coefficient, taking even or odd downsampling of the filtered coefficients significantly dominants the image reconstruction performance. Therefore, constructing the low-band wavelet coefficients through the odd-length wavelet filter in lifting structure based on even and odd pixels of the original image are respectively performed in this paper. Then, select the better one to fit the characteristics of the odd-length wavelet filters such as 5/3 and 9/7 filters and the pre-specified image for high performance image compression. To further improve the reconstruction performance and show the optimality of the above-mentioned proposed algorithms, the EP-based image interpolations are also presented in this paper. The proposed prediction algorithms with/without EP outperform other standard methods, and the computational complexity without EP is also acceptable in the experimental results.
|
|---|
