基于纹理特征与最优稀疏表示的图像修复算法

发布时间：2019-04-15 22:44

【摘要】：目的解决当前图像修复算法忽略了对修复块后续的优化处理,导致修复图像易出现不连贯效应以及块效应等的不足。方法提出基于纹理特征与稀疏表示的图像修复算法,首先利用像素点对应的数据项,构造了优先权模型。然后,利用像素点在R,G,B分量上对应的像素值来构造纹理特征度量模型,对待修复块中像素点对应的纹理特征进行度量,并根据度量结果,选择其对应样本集的大小。引入SSD型,从样本集中搜索与待修复块最相似的最优样本块,对待修复块进行填充。最后,利用最优样本块函数,构造最优稀疏表示模型,从而实现图像修复。结果仿真结果显示,与当前图像修复算法相比,所提图像修复算法具备更高的复原质量,能有效克服修复图像中出现的不连贯效应以及块效应。结论所提算法具有较高的修复视觉质量,在数字图像处理领域具有较好的应用价值。
[Abstract]:Aim to solve the problem that the current image restoration algorithm neglects the subsequent optimization of the restoration block, which leads to the incoherence effect and block effect in the image restoration. Methods an image restoration algorithm based on texture features and sparse representation was proposed. Firstly, the priority model was constructed by using the corresponding data items of pixel points. Then, the texture feature measurement model is constructed by using the corresponding pixel values in the R, G and B components. The texture features corresponding to the pixel points in the repair block are measured, and the size of the corresponding sample set is selected according to the measurement results. The SSD type is introduced to search the optimal sample block from the sample set which is most similar to the block to be repaired and to fill the repair block. Finally, the optimal sparse representation model is constructed by using the optimal sample block function to realize image restoration. Results the simulation results show that compared with the current image restoration algorithm, the proposed image restoration algorithm has higher restoration quality, and can effectively overcome the incoherent effect and block effect in the restoration image. Conclusion the proposed algorithm has high restoration visual quality and has good application value in the field of digital image processing.
【作者单位】：郑州铁路职业技术学院;郑州大学;
【基金】：国家自然科学基金(61103202) 河南省科技计划重点项目(102102210416)
【分类号】：TP391.41

【相似文献】