發(fā)布時間：2018-01-08 12:28

  本文關(guān)鍵詞：基于小波域稀疏最優(yōu)的圖像修復(fù)方法　出處：《電子學(xué)報(bào)》2016年03期 　論文類型：期刊論文

  更多相關(guān)文章： 模糊 小波 稀疏性 凸函數(shù) 加速

【摘要】：由模糊和噪聲引起的圖像退化屬于非線性病態(tài)逆問題,修復(fù)比較困難.由于小波的稀疏表示能力較強(qiáng),為提高修復(fù)質(zhì)量,提出利用正交小波作為稀疏基,以小波系數(shù)的稀疏性為先驗(yàn)構(gòu)造凸函數(shù),最小化后得到修復(fù)圖像;并提出將優(yōu)化問題轉(zhuǎn)化為逼近算子形式,利用不動點(diǎn)理論求解;證明了只需對構(gòu)造出來的迭代形式的解析解反復(fù)迭代就可以得到最優(yōu)解.對方法的構(gòu)造過程、收斂性和復(fù)雜度進(jìn)行了細(xì)致的分析,給出了迭代解,并結(jié)合加速方法提高了算法速度.仿真表明,本文方法具有較強(qiáng)的修復(fù)能力,收斂速度較快,能夠有效去除模糊和噪聲,保留圖像的邊緣和細(xì)節(jié)信息.
[Abstract]:Image degradation caused by blur and noise is a nonlinear ill-conditioned inverse problem, which is difficult to repair. Because of the strong sparse representation ability of wavelet, in order to improve the repair quality, orthogonal wavelet is used as sparse basis. A convex function is constructed with the sparsity of wavelet coefficients as a priori, and the reconstructed image is obtained after minimization. The optimization problem is transformed into approximation operator and solved by fixed point theory. It is proved that the optimal solution can be obtained by iterating the analytic solution of the constructed iterative form. The convergence and complexity of the method are analyzed in detail, and the iterative solution is given. Simulation results show that the proposed method has a strong ability to repair and converge quickly, which can effectively remove blur and noise and retain the edge and detail information of the image.
【作者單位】： 西安理工大學(xué)機(jī)械與精密儀器工程學(xué)院;
【基金】：陜西省自然科學(xué)基金(No.2014JM7273)
【分類號】：TP391.41
【正文快照】： 1引言圖像在拍攝、傳輸和儲存中會降低質(zhì)量,圖像修復(fù)很重要.圖像降質(zhì)的主要原因有模糊、噪聲、像素丟失、有損壓縮等.造成模糊的主要原因包括相機(jī)振動(運(yùn)動模糊)、對焦不準(zhǔn)(高斯模糊)、目標(biāo)高速運(yùn)動(運(yùn)動模糊)、大氣湍流(湍流模糊)、人為因素(均值模糊).目前,去模糊方法多種多

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 劉三陽;;凸函數(shù)的新發(fā)展[J];西安電子科技大學(xué)學(xué)報(bào);1990年01期

2 楊建芳;;具有負(fù)系數(shù)的近于凸函數(shù)[J];杭州電子科技大學(xué)學(xué)報(bào);2007年01期

3 吳從p，

本文編號：1397131