中文摘要：

该文提出一种基于结构成分双向扩散的插值方法，有效地减小了插值图像的边缘扩散，从而获得更为清晰的边缘。该方法采用改进的耦合双向扩散滤波器对轮廓模板插值图像进行边缘增强。其中，为了使滤波器更精确地作用于边缘轮廓，利用形态成分分析（MCA）分离出初始插值图像中的结构分量再实行滤波；同时，改进双向扩散模型，使其能够根据边缘梯度自适应地调整边缘扩散程度，且更加柔和地控制梯度方向的像素值变化。实验结果表明，对比传统的插值方法、相关的边缘自适应插值方法以及几种应用普遍的商用软件，该方法获得的插值图像主、客观质量均有明显提升，不仅有效提高图像锐度，且边缘光滑、过渡自然，避免产生边缘锯齿和过度的人工效应。

英文摘要：

An image interpolation method based on structure component bidirectional diffusion is proposed in this paper, by which the edge diffusion in image magnification is effectively decreased and the sharpening edges are generated. In this method, the edges are enhanced by using the advanced coupling of bidirectional diffusion filter after contour stencils interpolation. In order to deal with the edge contours more precisely, the structure component of initial interpolated image is filtered after separating from the initial interpolated image via Morphological Component Analysis （MCA）. Furthermore, the coupling of bidirectional filter is improved to adaptively adjust the edge diffusion degree according to the edge gradient, and to control pixels value change along gradient direction more gently. The experimental results show that, the proposed method outperforms other comparing algorithms including the traditional interpolation algorithm and the related edge adaptive interpolation algorithms and several widely-used commercial software in terms of both objective and visual quality of the interpolation image. The method enhances the image sharpness effectively, and gains smooth edges, nature transition, also avoids producing the edge aliasing and overshoot artifacts.