CAJ | 학술논문

针对图像加性高斯白噪声，提出一种优化的自适应参数滤波算法。该算法以非局部欧氏中值（non-local Euclidean medians，NLEM）滤波算法为基础，根据含噪图像梯度幅值在一定噪声范围内服从 Rayleigh 分布这一特性，求得以梯度幅值和噪声标准差为自变量的二元自适应滤波参数，并将它引入到邻域的权值计算中。其次，噪声的变化影响着 e p 范数回归的选择，在一定范围内以噪声标准差为自变量对参数 p 进行多项式拟合，得到自适应 e p 范数回归。在自适应滤波参数基础上，用自适应 e p 范数回归进一步改进 NLEM 滤波算法的 e 1-范数回归。所选图像的实验结果表明，本文算法在一定噪声范围内不但获得满意的去噪效果，而且有效地减少人机交互程度。
침대도상가성고사백조성，제출일충우화적자괄응삼수려파산법。해산법이비국부구씨중치（non-local Euclidean medians，NLEM）려파산법위기출，근거함조도상제도폭치재일정조성범위내복종 Rayleigh 분포저일특성，구득이제도폭치화조성표준차위자변량적이원자괄응려파삼수，병장타인입도린역적권치계산중。기차，조성적변화영향착 e p 범수회귀적선택，재일정범위내이조성표준차위자변량대삼수 p 진행다항식의합，득도자괄응 e p 범수회귀。재자괄응려파삼수기출상，용자괄응 e p 범수회귀진일보개진 NLEM 려파산법적 e 1-범수회귀。소선도상적실험결과표명，본문산법재일정조성범위내불단획득만의적거조효과，이차유효지감소인궤교호정도。
For additive white Gaussian noise of an image,this paper proposes an optimized adaptive parame-ter filter algorithm.Based on the non-local Euclidean medians (NLEM)algorithm,according to the property that the noise image gradient amplitude obeys Rayleigh distribution within a certain noise range,we obtain a bi-nary adaptive filter parameter by regarding gradient amplitude and noise standard deviation as independent varia-bles.The adaptive filter parameter is introduced in the weight calculation of neighbors.Furthermore,the chan-ges of the noise affect the selections of the e p norm regression.Make p used polynomial fit with noise standard deviation in a certain range,and get adaptive e p norm regression.On the basis of adaptive filter parameters,e 1 norm regression used in NLEM can be improved by using adaptive e p norm regression.It is verified that the new algorithm not only obtains satisfactory results of denoising in a certain noise range,but also reduces the degree of human-computer interaction effectively.