节点文献
基于改进PCNN的昆虫羽翅图像修复
Image inpainting for insect feathered wing based on improved PCNN
【摘要】 昆虫羽翅是一种结构比较复杂的纹理图像,目前的研究算法都是沿等照度线方向扩散信息,因而在修复线性结构的图像时效果较好,但在修复具有复杂结构的图像时效果不理想。提出了一种新的可以同时修复复杂结构和纹理细节的修复方法。结果表明,该方法对于昆虫羽翅的修复是有效的。
【Abstract】 The image of insect feathered wing is a kind of texture image with complex restructure.In the posed years,researchers proposed many algorithms which diffused the pixels,which came from the surrounding neighborhood or selected example,into damaged region along isophate direction.The paper advanced a novel method which simultaneously inpaints structures and textures of damaged images.The experimental result showed that it was effective to inpaint the insect feathered wing.
【关键词】 昆虫羽翅;
图像修复;
纹理分割;
脉冲耦合神经网络;
【Key words】 insect feathered wing; image inpainting; texture segmentation; Pulse-Coupled-Neural-Network;
【Key words】 insect feathered wing; image inpainting; texture segmentation; Pulse-Coupled-Neural-Network;
【基金】 国家自然科学广东联合基金项目(U1035004);国家自然科学基金面上项目(61070090);广东省工业科技计划项目(2009B030803004);广东省科技计划项目(2010A080402005)
- 【文献出处】 广东农业科学 ,Guangdong Agricultural Sciences , 编辑部邮箱 ,2011年13期
- 【分类号】TP391.41
- 【下载频次】69