基于Otsu阈值分割的边缘快速图像插值算法
|
摘要
为了满足视频监控、目标检测与识别过程中较高图像质量和较低算法复杂度要求,以及改善传统图像插值中细节模糊和边缘锯齿效应,文中提出一种基于Otsu阈值分割的边缘快速图像插值算法。利用Otsu算法,根据目标区域和背景区域的类方差最大,确定分割阈值,对非边缘区域进行双线性插值,边缘区域利用与待插值点周围6个或8个相邻降采样像素局部结构的多方向特点,自适应估计高分辨率像素值。实验表明,该算法运算复杂度低,很好保持了图像的边缘,获得了视觉质量较好的高分辨率图像。
A fast edge image interpolation algorithm based on Otsu threshold segmentation is proposed in this paper to meet the requirements of high image quality and low algorithm complexity during the process of video monitoring,and target detection and recognition,and improve detail fuzziness and saw-tooth effect during the traditional image interpolation. The Otsu algorithm is used to determine the segmentation threshold according to the maximum class variance between the target region and background region. The bilinear interpolation is conducted for the non-edge area. The multi-direction characteristic for the local structure of six or eight adjacent downsampling pixels around the point under interpolation is used for the edge area,so as to selfadaptively estimate high resolution pixel values. The experimental results show that the algorithm has low computational complexity,can well retain the image edge,and obtain high resolution images with good visual quality.
A fast edge image interpolation algorithm based on Otsu threshold segmentation is proposed in this paper to meet the requirements of high image quality and low algorithm complexity during the process of video monitoring,and target detection and recognition,and improve detail fuzziness and saw-tooth effect during the traditional image interpolation. The Otsu algorithm is used to determine the segmentation threshold according to the maximum class variance between the target region and background region. The bilinear interpolation is conducted for the non-edge area. The multi-direction characteristic for the local structure of six or eight adjacent downsampling pixels around the point under interpolation is used for the edge area,so as to selfadaptively estimate high resolution pixel values. The experimental results show that the algorithm has low computational complexity,can well retain the image edge,and obtain high resolution images with good visual quality.
引文
[1] SAJJAD M,EJAZ N,BAIK S W. Multi-kernel based adaptive interpolation for image super-resolution[J]. Multimedia tools and applications,2014,72(3):2063-2085.
[2]金海丁,周孝宽.数字图像自适应插值法[J].激光与红外,2006,36(9):907-910.JIN Haiding,ZHOU Xiaokuan. Self-adaptive interpolation algorithm for digital image[J]. Laser&infrared,2006,36(9):907-910.
[3] KEYS R G. Cubic convolution interpolation for digital image processing[J]. IEEE transactions on acoustics,speech and signal processing,1981,29(6):1153-1160.
[4]李春龙,潘海侠,王华峰.自适应立方卷积图像插值算法[J].北京航空航天大学学报,2014,40(10):1463-1468.LI Chunlong,PAN Haixia,WANG Huafeng. Adaptive cubic convolution based image interpolation approach[J]. Journal of Beijing University of Aeronautics and Astronautics,2014,40(10):1463-1468.
[5] LI X,ORCHARD M T. New edge-directed interpolation[J].IEEE transactions on image processing,2001,10(10):1521-1527.
[6] CHEN Meijuan,HUANG Chinhui,LEE Wenli. A fast edge-oriented algorithm for image interpolation[J]. Image and vision computing,2005,23(9):791-798.
[7]吴锡生,党向盈,赵勇.基于阈值控制的边缘自适应快速图像插值算法[J].计算机工程,2007,33(22):226-228.WU Xisheng,DANG Xiangying,ZHAO Yong. Fast self-adaptive edge-oriented interpolation based on threshold method[J].Computer engineering,2007,33(22):226-228.
[8] OTSU N. A threshold selection method from gray-level histograms[J]. IEEE transactions on systems,man and cybernetics,1979,9(1):62-66.
[9]张铮.精通Matlab数字图像处理与识别[M].北京:人民邮电出版社,2013.ZHANG Zheng. Proficient in Matlab digital image processing and recognition[M]. Beijing:Posts&Telecom Press,2013.
[10] TIAN Qichong,WEN Hao,ZHOU Chenhui,et al. A fast edge-directed interpolation algorithm[C]//Proceedings of the19th International Conference on Neural Information Processing. Berlin:Springer,2012:398-405.
[2]金海丁,周孝宽.数字图像自适应插值法[J].激光与红外,2006,36(9):907-910.JIN Haiding,ZHOU Xiaokuan. Self-adaptive interpolation algorithm for digital image[J]. Laser&infrared,2006,36(9):907-910.
[3] KEYS R G. Cubic convolution interpolation for digital image processing[J]. IEEE transactions on acoustics,speech and signal processing,1981,29(6):1153-1160.
[4]李春龙,潘海侠,王华峰.自适应立方卷积图像插值算法[J].北京航空航天大学学报,2014,40(10):1463-1468.LI Chunlong,PAN Haixia,WANG Huafeng. Adaptive cubic convolution based image interpolation approach[J]. Journal of Beijing University of Aeronautics and Astronautics,2014,40(10):1463-1468.
[5] LI X,ORCHARD M T. New edge-directed interpolation[J].IEEE transactions on image processing,2001,10(10):1521-1527.
[6] CHEN Meijuan,HUANG Chinhui,LEE Wenli. A fast edge-oriented algorithm for image interpolation[J]. Image and vision computing,2005,23(9):791-798.
[7]吴锡生,党向盈,赵勇.基于阈值控制的边缘自适应快速图像插值算法[J].计算机工程,2007,33(22):226-228.WU Xisheng,DANG Xiangying,ZHAO Yong. Fast self-adaptive edge-oriented interpolation based on threshold method[J].Computer engineering,2007,33(22):226-228.
[8] OTSU N. A threshold selection method from gray-level histograms[J]. IEEE transactions on systems,man and cybernetics,1979,9(1):62-66.
[9]张铮.精通Matlab数字图像处理与识别[M].北京:人民邮电出版社,2013.ZHANG Zheng. Proficient in Matlab digital image processing and recognition[M]. Beijing:Posts&Telecom Press,2013.
[10] TIAN Qichong,WEN Hao,ZHOU Chenhui,et al. A fast edge-directed interpolation algorithm[C]//Proceedings of the19th International Conference on Neural Information Processing. Berlin:Springer,2012:398-405.