【摘要】：為了改善小波變換的圖像稀疏表示性能,提出了一種小波域的灰色關(guān)聯(lián)度圖像壓縮算法.首先,利用小波變換對測試圖像進行分解,獲得不同區(qū)域的小波系數(shù);然后,利用小波系數(shù)特點,將灰色關(guān)聯(lián)度用于系數(shù)關(guān)聯(lián)度的刻畫中,并計算不同尺度間系數(shù)的灰色關(guān)聯(lián)度;根據(jù)小波系數(shù)區(qū)域特征,將小波系數(shù)進行分類,構(gòu)造出不同系數(shù)類型下的稀疏表示方法;最后,將該算法應(yīng)用于圖像壓縮.實驗結(jié)果表明,在相同壓縮率下,所提算法的客觀評價指標(biāo)峰值信噪比較現(xiàn)有同類算法提高了1.04~3.65 d B,圖像主觀視覺質(zhì)量明顯提高.所提算法能夠結(jié)合系數(shù)特征和視覺特性自適應(yīng)地構(gòu)造字典,提高了圖像稀疏表示能力,進一步提高了圖像壓縮性能.
[Abstract]:In order to improve the sparse representation performance of wavelet transform, a gray relational image compression algorithm in wavelet domain is proposed. Firstly, the wavelet transform is used to decompose the test image to obtain the wavelet coefficients in different regions. Then, by using the characteristics of wavelet coefficients, the grey correlation degree is applied to the description of coefficient correlation degree, and the grey correlation degree of coefficients between different scales is calculated. According to the regional characteristics of wavelet coefficients, the wavelet coefficients are classified and sparse representation methods under different coefficient types are constructed. finally, the algorithm is applied to image compression. The experimental results show that under the same compression ratio, the objective evaluation index of the proposed algorithm is 1.04 鈮，

本文編號：2475641