摘要
对压缩图像传输提取重构优化,可以提高压缩图像传输的速率。进行图像传输提取重构时,应对压缩图像信号函数进行稀疏表示,利用小波信号稀疏特征选择信号的重构特征,传统的重构方法主要通过提取压缩图像传输特征进行重构,但是无法稀疏表示压缩图像信号函数,不能利用小波信号稀疏特征对信号的重构特征进行选择,导致存在重构不准确、效率低的问题。提出基于小波变换的压缩图像传输提取重构方法。首先,运用小波变换将原始信号由时域转换至频域,将信号函数利用伸缩及平移等方式进行求解及分析,使高、低频信号分别得到多尺度细化处理,最终该信号函数采用稀疏系数表示,获得信号细节特征;然后,利用多空间稀疏特征求解重构方法,分别将PAR残差系数、图像梯度进行最小化处理后,使二者作为优化计算对象,再依据变换后的小波信号稀疏特征,对其运用加权方式选择信号的重构特征,利用该特征对压缩图像进行重构。仿真结果表明,利用改进的重构方法,能够较好对图像边缘及纹理进行重构,提高了图像重构后的高分辨率。
This article proposes a reconstruction method for extraction transmission of compression image based on wavelet transform. The research converted original signal from time domain to frequency domain using the wavelet transform and solved and analyzed signal function using way of stretch and translation, which made signal with high and low frequency obtain multi-scale refining process respectively. Then sparse coefficient was used to express the signal function and obtained detail feature of signal. Moreover, the research carried out minimum process for PAR re- sidual coefficient and image gradient respectively using reconstruction method of solving sparse feature with muhi- space and used both of them as object of optimal computation. Once more, the research carried out weighting method to select reconstruction feature of signal for sparse feature of wavelet signal according to the sparse feature after trans- formation. Simulation results show that the modified reconstruction method can reconstruct border and texture of im- age. It improves the resolution of image after reconstruction.
出处
《计算机仿真》
北大核心
2017年第7期316-319,共4页
Computer Simulation
关键词
压缩图像
传输
提取
重构
Compression image
Transmission
Extraction
Reconstruction
