一种改进的局部自适应数字图像插值法  

AN IMPROVED LOCAL ADAPTIVE ALGORITHM FOR DIGITAL IMAGE INTERPOLATION

作　　者：陈小珊[1] 江雯[2] 杨帆[2] 陈更生[2] 吴秀龙[1] 周汀[2] 

机构地区：[1]安徽大学电子科学与技术学院,安微合肥230039 [2]复旦大学专用集成电路与国家重点实验室,上海201203

出　　处：《计算机应用与软件》2010年第5期246-249,共4页Computer Applications and Software

摘　　要：图像插值用于图像格式间的转换、改善图像质量等。经典的插值法对整幅图像无区别地使用一种插值函数,插值后的图像细节模糊,视觉效果不理想,抗噪能力差。针对这一问题,许多学者提出了基于边缘的自适应法以改善图像边缘,插值后图像视觉效果得到明显改善,但其抗噪能力没有提高。提出一种改进的局部自适应数字图像插值法ILSI(An Improved Locally Self-adaptive In-terpolation for Digital Image),该算法根据目标图像中源像素点的平滑程度对输出图像进行自适应插值。实验表明,该算法克服了经典插值图像存在的边缘锯齿、模糊效应及现有边缘自适应方法抑制噪声能力差的缺陷,插值后的图像呈现出良好的视觉效果。Image interpolation is used for format transformation and image quality improvement.Conventional methods adopt identical interpolation function throughout the whole image with no difference,the interpolated image has blurred detail,poor visual effects and weak noise-resistance.For solving the above problems,scholars suggested the edge-based adaptive interpolation methods for the amelioration of image edges,with it the visual effect of interpolated image has been improved noticeably.However,the ability of anti-noise is not enhanced.This paper introduces an improved local adaptive algorithm for digital image interpolation.In the algorithm,adaptive interpolation is made on inputted image according to the smoothness degree of the source pixel points in target image.The algorithm conquers the jagged edge and blurring effects in conventional methods and the flaw of existing edge-adaptive algorithm in poor capability of restraining noise,as shown in experiment.The interpolated target image presents preferable visual effects.

关 键 词：插值 自适应 抗噪 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]