系统工程与电子技术
繫統工程與電子技術
계통공정여전자기술
SYSTEMS ENGINEERING AND ELECTRONICS
2010年
1期
188-191,194
,共5页
图像恢复%扩散张量%自适应正则化%变分模型%冲击滤波器
圖像恢複%擴散張量%自適應正則化%變分模型%遲擊濾波器
도상회복%확산장량%자괄응정칙화%변분모형%충격려파기
image restoration%diffusion tensor%adaptive regularization%variation model%shock filter
结构张量是描述图像的有效工具.利用结构张量对图像灰度变化的方向和大小进行判断,提出基于扩散张量的自适应正则化变分模型.该模型将冲击滤波器耦合在其中,使其在恢复图像的同时能有效地增强边缘.同时,给出一种构造正则化参数的方法.仿真实验表明,该模型在对带噪图像进行自适应恢复时,能较好地保护边缘信息,增强纹理特征,得到了较为满意的结果.
結構張量是描述圖像的有效工具.利用結構張量對圖像灰度變化的方嚮和大小進行判斷,提齣基于擴散張量的自適應正則化變分模型.該模型將遲擊濾波器耦閤在其中,使其在恢複圖像的同時能有效地增彊邊緣.同時,給齣一種構造正則化參數的方法.倣真實驗錶明,該模型在對帶譟圖像進行自適應恢複時,能較好地保護邊緣信息,增彊紋理特徵,得到瞭較為滿意的結果.
결구장량시묘술도상적유효공구.이용결구장량대도상회도변화적방향화대소진행판단,제출기우확산장량적자괄응정칙화변분모형.해모형장충격려파기우합재기중,사기재회복도상적동시능유효지증강변연.동시,급출일충구조정칙화삼수적방법.방진실험표명,해모형재대대조도상진행자괄응회복시,능교호지보호변연신식,증강문리특정,득도료교위만의적결과.
Structure tensor is good for describing images, this paper puts forward an adaptive regularized variation model based on diffusion tensor. In this model, the direction of diffusion and the characters of different kinds of pixel in noisy images are characterized by the eigenvector and eigenvalues of structure tensor. In order to enhance edges, the shock filter is coupled to it. And the principle of selecting the parameters is presented. Simulation experiments indicate that the proposed model can not only denoise efficiently but also preserve detail information well, thus obtainining the better results.
