摘要
针对传统的双线性插值图像放大算法存在的边缘模糊问题,提出一种改进算法:先计算插值点的双线性插值和最近邻点插值,然后以4个邻点的灰度方差构造权重,将二种插值进行加权融合获得最终插值结果。该算法既考虑到插值点与邻点之间距离关系,又考虑到邻点的灰度分布特性,有效地提高了放大图像的质量。实验结果验证了算法的有效性。
To solve traditional bilinear interpolated for the problem of blurry edge in image zooming algorithm, this paper proposes an improved algorithm. First, the algorithm calculates the interpolation point of the linear interpolation and the nearest neighbor interpolation, and then takes the four adjacent points as weight and finally makes weighted fusion the two interpolation to attaint the final interpolation result. This method not only takes into account the distance between interpolation point and neighbor point, but also thinks over the gray distribution of neighbor pixels, effectively improves definition of zoomed image. Experiment results show that the proposed algorithm is efficient.
出处
《激光与红外》
CAS
CSCD
北大核心
2009年第8期899-901,共3页
Laser & Infrared
关键词
图像放大
双线性插值
最近邻点插值
加权融合
image zooming
bilinear interpolation
nearest neighbor interpolation
weighted fusion
