摘要
研究了图像局部放大提高清晰度问题。图像放大一直是图像处理领域研究的重点内容之一,由于图像在放大过程中,容易出现局部模糊不清晰,产生锯齿状轮廓的图像,而传统的图像处理放大插值算法方法难以有效解决该问题,为了解决上述问题,提出了一种新的结合B样条插值和水平集重构算法图像局部放大算法。首先利用B样条函数对图像插值放大的快速算法,借鉴了增强图像高频信号对水平集分割的指导作用,并优化驱动水平集演化的内、外能量及曲线长度约束能量,采用基于水平集算法来平滑图像,重建图像的水平曲线且同时保持图像保真度。最后采用贝叶斯修复技术来修复放大后的图像,仿真结果表明,提出的技术可以有效的克服传统图像放大算法缺点,图像放大清晰度更高,插值效果好并且处理时间短。
Study the partial enlarged images to improve the clarity.The magnification of the image has been the key content in,research of image processing.The paper put forward a combined image magnification algorithm based on B-spline interpolation and level set reconstruct algorithm.Firstly,B-spline function was used for image interpolation algorithm,and the guidance effect of enhanced image frequency signals on level set segmentation was used for referenc.Then,the internal and external energy of driving level set evolution and the constraint energ of curve length were optimized.Based on the level set algorithm,the image was smoothed and the image horizontal curves were reconstructed while maintaining the image fidelity.Finally,the Bayesian restoration techniques was used to repair the magnified image.Simulation and experimental results show that the proposed technique can effectively overcome the traditional shortcomings of image magnification algorithm,and has higher image magnification clarity,good interpolation effect and shorter processing time.
出处
《计算机仿真》
CSCD
北大核心
2012年第5期269-272,共4页
Computer Simulation
关键词
局部放大
图像重构
水平集
样条函数
Local amplification
Image reconstruction
Level set
Spline function
