摘要
针对大规模平面设计中对图像去噪与失真还原的低复杂度要求,文中提出基于回溯分段正则化最优匹配的压缩感知图像处理方法。在构建压缩感知信号模型的基础上,推导图像重构的等价最优化问题。此外,为了降低最优化问题求解的复杂度,通过正则化二次处理与分段筛选,在保证图像重构质量的同时,可以有效地减少图像处理时间。仿真实验结果表明,所提算法对于不同的图像及采样率均具有良好的重构质量,且相对于正交匹配算法,其复杂度显著降低。
In allusion to the requirements of low complexity for the image de⁃noising and distortion restoration in the large⁃scale graphic design,a method of compressive sensing image processing based on backtracking piecewise regularization optimal matching is proposed.The equivalent optimization of the image reconstruction is deduced on the basis of construction of compressive sensing signal model.In order to reduce the complexity of optimization solution,the regularization of secondary processing and segmentation screening are performed,by which the image processing time can be effectively reduced while the quality of image reconstruction is guaranteed.The simulation experimental results show that the proposed algorithm has good reconstruction quality for different images and sample rates,and its complexity is lower than that of the orthogonal matching algorithm.
作者
李辉
LI Hui(School of Information Technology,Guilin University of Electronic Technology,Guilin 541004,China)
出处
《现代电子技术》
北大核心
2020年第18期19-21,25,共4页
Modern Electronics Technique
基金
广西创新驱动发展专项(科技重大专项)项目(桂科AA17204006)
国家社科基金艺术学项目(NS140042)。
关键词
平面设计
图像处理
压缩感知
正则化处理
分段筛选
仿真实验
graphic design
image processing
compressive sensing
regularization processing
segmentation filtering
simulation experiment
