基于边缘最大相关性的快速图像插值算法

处理视频、网络信号要求有较好的视觉质量和较低的算法复杂度、运算量。传统的图像插值算法会产生细节模糊和边缘锯齿化,为此，提出了一种基于图像边缘多方向最大相关性的快速数字图像插值算法。为了得到高分辨率图像相关信息，根据图像空间域内多个方向邻近像素点的相关特性，利用与待内插值点相邻6个降采样像素值。实验证明，该算法降低了运算复杂度，有效地保持了边缘信息，得到了视觉质量较好的插值图像。

Fast image interpolation algorithm based on edge-directed max-relativity

Better subjective quality and lower computational complexity of the algorithm are important in video and network signal processing. And the smooth effect and staircase of image details or edges are inevitable during conventional image interpolation. A new rapid image interpolation algorithm based on edge-directed max-relativity was proposed. The algorithm took account of the multiple directions relativity of local neighbors and made full use of six down-sampled neighbors involved in the low-resolution image to estimate relative information at high resolution. Experimental resuhs demonstrate that the algorithm not only decreases the computational complexity but also substantially improves the visual quality of the interpolated images.