摘要
在对图像变分描述的前提下,为有效地利用条带噪声之间的相似性,本文将条带噪声的群稀疏表示引入到单向变分模型中,提出群稀疏技术限制的单向变分模型,并采用交替方向乘子法求解该模型。对比实验证明,本文所提出的群稀疏限制的单向变分模型能有效地利用条带噪声的相似性实现条带噪声的消除,更好地重构图像的细节信息,峰值信噪比与结构相似性比其他模型分别提高6.76 dB和0.25,图像去噪性能更优。
By introducing the group sparse representation of strip noise into the single variation model,a single variational model restricted by group sparse technique is proposed to remove noise.The proposed model is solved by alternating direction method of multipliers.The experiments show that the proposed model can more effectively use the similarities between strip noise to remove noise compared with other existing models,and it can reconstruct the image information better.The values of peak signal-to-noise ratio and structural similarity are increased by 6.76 dB and 0.25,respectively,which shows good image denoising performance of our model.
作者
陈柳
陈明举
吴浩
薛智爽
CHEN Liu;CHEN Ming-ju;WU Hao;XUE Zhi-shuang(School of Information Engineering, Sichuan University of Science & Engineering, Zigong 643000, China;Robot Technology Used for Special Environment Key Laboratory of Sichuan Province, Southwest University of Science and Technology, Mianyang 621010, China)
出处
《液晶与显示》
CAS
CSCD
北大核心
2020年第6期604-611,共8页
Chinese Journal of Liquid Crystals and Displays
基金
四川省科技厅项目(No.2018GZDZX0043,No.2019YJ0476,No.2019YJ0477)
企业信息化与物联网测控技术四川省高校重点实验室项目(No.2018WZY01)。
关键词
条带噪声
变分
群稀疏
交替方向乘子法
strip noise
variation
group sparse
alternating direction method of multipliers
