图像去马赛克算法研究

图像去马赛克算法研究(论文12700字)
摘要：数码相机的生产厂商为了降低产品成本和硬件电路复杂程度，通常在单片图像传感器上覆盖彩色滤波阵列进行数据采样。为了获得一幅真彩色图像，需要进行插值或去马赛克来得到。在去马赛克算法中，图像高频部分容易出现失真，为了解决这类问题，需要我们提高去马赛克算法的质量。本文对几种经典算法进行了归类介绍和MATLAB实现的基础上，综合提出了一种改进的去马赛克算法，以边缘方向检测为基础，分步进行RGB三通道插值；接着对初步获得的重建图像进行后续操作，来抑制虚假色等现象。最后结合常见的算法性能指标PSNR和CIELAB空间的欧式距离 来分析数据层面的算法优劣。实验结果表明，本文提出的算法达到了较好的性能表现。
关键词：Bayer；去马赛克；边缘检测；MATLAB；峰值信噪比

ResearchonDemosaicingAlgorithm
Abstract:In order to reduce the cost and complexity of the hardware circuit, Digital camera manufacturers usually use a single image sensor covered by the color filter array to operate data sampling. In order to obtain a true color image, we need to do color interpolation or demosaicing process. In the demosaicing algorithm, the high frequency component of the image is prone to distortion.in order to solve such a problem, we need to improve the quality of demosaicing algorithm. In this paper, based on the introduction of several classical algorithms and MATLAB implementation, an improved demosaicing algorithm is proposed, which is based on the edge direction detection, and the three-channel interpolation is carried out step by step. Then, we continue doing Image follow-up operation to suppress false color and zipper effect and so on. Finally,we combine the common algorithm performance index PSNR and CIELAB space European distance to analyze the data level algorithm advantages and disadvantages.The experimental results show that the proposed algorithm achieves better performance.
Key words:Bayer CFA ;demosaicing;edge detection;MATLAB;PSNR

目 录
1绪论    1
1.1数码相机及马赛克图像    1
1.1.1数码相机构造    1
1.1.2马赛克图像的产生    3
1.2去马赛克算法研究现状    4
1.3论文的主要工作及结构安排    4
2数字图像处理基础    5
2.1色度学基础    5
2.2 彩色模型    5
2.2.1 RGB模型    5
2.2.2HSI颜色空间    6
2.2.3CIEXYZ颜色空间    7
2.2.4 CIELab颜色空间    7
2.3常见的插值失真现象    8
2.4 MATLAB及在数字图像处理中的应用    8
3经典的图像去马赛克算法    9
3.1双线性插值法    10
3.2色比色差恒定法    10
3.3基于梯度的插值算法    11
3.3.1 基于一阶的边缘检测算法    11
3.3.2基于二阶的边缘检测算法    12
3.4Adams-Hamilton自适应插值法    12
3.5 Lu算法    13
3.6本章小结    14
4改进的去马赛克算法    15
4.1改进的插值方法    16
4.1.2 插值G通道    17
4.1.3插值R/B通道    18
4.2后续处理    19
5插值算法的性能评价    20
5.1评估方法    20
5.1.1峰值信噪比PSNR    20
5.1.2CIELAB空间的欧式距离    21
5.3本章小结    24
参考文献.................................................25
致谢.....................................................26
附录.....................................................27