Abstract: In order to restrain the noise and obtain better visual effect, a novel image denoising method is proposed based on hybrid variation model. The harmonic model and the total variation model are integrated to enhance the filtering performance, and then the parameters are adaptively selected. When combinatorial coefficient is large, hybrid model is close to total variation model, and when combinatorial coefficient is small, hybrid model is close to harmonic model to effectively remove the staircase effect to protect the edge detail. The simulation experiment is carried out to test performance of model. The results show that, compared with other denoising models, the proposed model has better denoising effect and can better protect edge information under the same conditions.

Key words: image denoising, harmonic model, total variational model, adaptive denoising

摘要： 为了有效抑制噪声，获得更好的视觉效果，提出了一种基于混合变分模型的图像去噪方法。将调和模型和全变分模型进行融合，增强模型的去噪性能，根据自适应选取组合系数，组合系数较大时偏向于全变分模型，较小时偏向于调和模型，这样不仅可以有效去除阶梯效应，同时保护边缘细节，采用仿真对比实验以测试模型性能。结果表明，相对其他去噪模型，相同条件下，该模型取得更优的去噪声效果，提高了图像的质量。

关键词: 图像去噪, 调和模型, 全变分模型, 自适应去噪