基于阈值判断的区域指导插值算法  

Area-directed threshold-judgment image interpolation algorithm

作　　者：赵旦峰[1] 王博[1] 杨大伟[1] 

机构地区：[1]哈尔滨工程大学信息与通信工程学院,黑龙江哈尔滨150001

出　　处：《系统工程与电子技术》2013年第1期203-206,共4页Systems Engineering and Electronics

基　　金：国家高技术研究发展计划(863计划)(2011AA110203);黑龙江省科技攻关(GC12A305);哈尔滨市科技创新人才基金(2009RFXXG029);中央高校自由探索计划(HEUCF120804)资助课题

摘　　要：针对传统图像放大算法对边缘区域划分不准确造成图像质量较低,而自适应插值放大方法模型复杂且计算量大的问题,提出一种基于阈值判断的区域指导(area-directed threshold-judgment,ADT)插值算法。该算法利用阈值判断的方法对插值点周围进行区域划分,结合近邻法和众数法确定插值点所属区域,根据插值点类型不同,利用不同权值大小的线性插值公式实现自适应插值,提高图像质量。仿真测试结果表明,与基于协方差的自适应插值算法相比,ADT算法在图像质量降低不显著的前提下,能显著提高插值速度;与双线性插值算法相比,图像峰值信噪比平均增加2.19dB。A novel area-directed threshold-judgment （ADT） image interpolation mechanism is proposed for solving the area division problem in traditional magnification methods and the complexity of the model and com- putation in some adaptive image interpolation methods. The proposed algorithm exploits the threshold-judgment to classify some serial suitable regions according to the neighbor pixels around the current interpolated pixel. The region that around the interpolated pixel should belong to is decided by threshold combined with the nearest neighbor method and the mode method. According to the different types of interpolated points, the linear inter polation formulas with different weights are designed to achieve adaptive interpolation and improve the image quality. The simulation results show that the proposed algorithm gets a better visual effect. Compared with the edge-adaptive interpolation algorithm using covariance, the proposed ADT algorithm improves the interpolation speed significantly based on the premise that the image quality reduces tiny, and has a higher increase in peak signal-to-noise ratio by 2.19 dB on average compared with the bilinear algorithm.

关 键 词：图像放大 插值 自适应 阈值 区域指导 

分 类 号：TN911.73[电子电信—通信与信息系统]