摘要
三维图像在获取、存储和传递时,由于受成像系统、传输方式、存储时间和存储介质等多种因素的影响,图像质量会产生一定的退化,严重的情况下会导致图像失真,影响使用。传统三维图像恢复方法多以线性分析为基础,假设条件多、运行耗时长、图像恢复效果差。为此,提出一种基于数据挖掘的三维图像无损恢复方法。依据图像变化的关联规则原理,分析三维图像成像的过程,并对其非相干成像条件做离散化处理;深度挖掘图像退化前后灰度变化的关联关系,并基于这种关联关系对原图像进行最大后验估计和先验计算,实现对三维图像的无损恢复。实验数据表明,提出的三维图像恢复方法运行时间短、均值误差低,具有良好的图像恢复效果。
In the acquisition,storage and transmission of3D image,the image quality degradation and even image distortion in severe situation may appear due to the influence of imaging system,transmission mode,storage time and storage medium,which may impact its usage.The traditional3D image restoration methods are mostly based on linear analysis,and have the dis-advantages of many assumed conditions,long running time and poor image restoration effect.Therefore,a3D image lossless re-covery method based on data mining is proposed.According to the association rules principle of image variation,the imaging pro-cess of3D image is analyzed,and the discretization is performed for its incoherent imaging conditions.The correlation of gray level variation before and after image degradation is mined deeply.On this basis,the maximum posteriori estimation and priori calculation are carried out for the original image to realize the lossless recovery of3D image.The experimental data shows that the3D image restoration method has short running time,low mean error and perfect image restoration effect.
作者
王丹
WANG Dan(College of Mechanical and Electronic Engineering,Huanghe Jiaotong University,Jiaozuo 454950,China)
出处
《现代电子技术》
北大核心
2018年第7期67-70,共4页
Modern Electronics Technique
关键词
关联规则
三维图像
图像失真
离散化处理
无损恢复
后验估计
association rule
3D image
image distortion
discretization
lossless recovery
posteriori estimation
