基于BEMD二次分解与维纳滤波相结合的图像去噪算法

目的提出基于二维经验模态分解(BEMD)与维纳滤波相结合的图像去噪算法并验证其去噪效果。方法通过对含噪的图像进行BEMD尺度分解,提取前两阶含噪的尺度分量,使用维纳滤波进行一次去噪,然后再次进行BEMD分解和滤波去噪,从而达到去噪的目的。将本文提出的算法与经典算法进行比较,验证该算法的可行性和鲁棒性。结果实验表明,该算法不仅能够有效降低图像的噪声,而且能很好地保持图像信息。结论本文提出的图像去噪算法思路新颖,效果良好,在法庭科学图像处理领域有很好的参考和应用价值。

Image De-noising Manipulation through Twice BEMD Coalesced with Wiener Filtering

Objective To put forth an idea about image de-noising through the coalescence of Bidimensional Empirical Mode Decomposition(BEMD)and Wiener filtering manipulation,with the resultant effect being verified.Methods One noise-bearing image was disintegrated with the BEMD handling so that the resultant fi rst two-order noise-bearing componential portions of the image were extracted to de-noise with Wiener filtering.Both the BEMD disintegration and Wiener filtering were performed once again into the de-noised two-order componential portions that had been dismantled out from the preceding manipulation,thereby making the noise borne into the image removed one more time.Finally,all the resultant de-noised componential portions were retroactively combined with the intact remainders of the image,hence coming into being the new image of effectual de-noising.With comparison against the conventional disposal choices,the manipulation adopted here was verified of its feasibility and robustness.Results The post-manipulation image demonstrated that such the processing here can not only effectively reduce the noise from image but also maintain the image information well.Conclusion The image de-noising manipulation introduced here is novel and effective,having good reference and applicability in forensic image processing.