一種新的基于參數(shù)估計的自適應雙邊濾波算法
發(fā)布時間:2018-10-15 14:20
【摘要】:傳統(tǒng)雙邊濾波算法需要根據(jù)經(jīng)驗預先設置空間標準差和灰度標準差,參數(shù)固定且不具有通用性。針對此問題,提出了一種新的基于參數(shù)估計的自適應雙邊濾波算法。通過圖像灰度共生矩陣實現(xiàn)空間標準差的自適應,利用統(tǒng)計方法估計光滑區(qū)域噪聲標準差,根據(jù)噪聲標準差設置灰度標準差,從而實現(xiàn)自適應雙邊濾波。仿真結果表明,提出的算法無論在主觀還是客觀評價上都取得了較好的效果。
[Abstract]:The traditional two-sided filtering algorithm needs to set the spatial standard deviation and the gray standard deviation in advance according to the experience, and the parameters are fixed and not universal. To solve this problem, a new adaptive bilateral filtering algorithm based on parameter estimation is proposed. The spatial standard deviation is adaptively realized by image gray level co-occurrence matrix, the noise standard deviation of smooth region is estimated by statistical method, and the gray level standard deviation is set according to the standard deviation of noise, so that adaptive bilateral filtering is realized. Simulation results show that the proposed algorithm has achieved good results in both subjective and objective evaluation.
【作者單位】: 四川大學電子信息學院;
【基金】:國家自然科學基金資助項目(61403265)
【分類號】:TN713;TP391.41
,
本文編號:2272818
[Abstract]:The traditional two-sided filtering algorithm needs to set the spatial standard deviation and the gray standard deviation in advance according to the experience, and the parameters are fixed and not universal. To solve this problem, a new adaptive bilateral filtering algorithm based on parameter estimation is proposed. The spatial standard deviation is adaptively realized by image gray level co-occurrence matrix, the noise standard deviation of smooth region is estimated by statistical method, and the gray level standard deviation is set according to the standard deviation of noise, so that adaptive bilateral filtering is realized. Simulation results show that the proposed algorithm has achieved good results in both subjective and objective evaluation.
【作者單位】: 四川大學電子信息學院;
【基金】:國家自然科學基金資助項目(61403265)
【分類號】:TN713;TP391.41
,
本文編號:2272818
本文鏈接:http://sikaile.net/kejilunwen/ruanjiangongchenglunwen/2272818.html
最近更新
教材專著