基于快速曲波变换的陶瓷显微图像处理  

CERAMIC MICROSCOPIC IMAGE PROCESSING BASED ON FAST CURVELET TRANSFORM

作　　者：刘国高[1] 

机构地区：[1]河海大学数理部,江苏常州213022

出　　处：《中国陶瓷工业》2008年第2期17-21,共5页China Ceramic Industry

基　　金：江苏省社会发展科技项目(编号:BS2007058);河海大学科技创新基金(编号:CC2006-02)

摘　　要：陶瓷内部结构信息对于陶瓷质量分析、生产工艺控制是非常重要的。由于陶瓷显微图像在采集和传输过程中不可避免地要受到光照分布不均匀、电子噪声等干扰因素的影响而使得图像的质量变差。需要首先对其进行去噪、增强处理,然后才能进行图像分析。曲波变换是在小波变换的基础上发展起来的一种新的多尺度分析方法,比小波更加适合分析二维图像中的曲线或直线状边缘特征,而且具有更高的逼近精度和更好的稀疏表达能力。快速曲波变换理论的提出也使得其理论更易理解和实现。因此,提出了一种基于快速曲波变换的图像去噪、增强方法,并将其引入陶瓷显微图像的处理中,然后按照分水岭算法进行粒度分割,得到陶瓷粒度分布的统计结果。实验结果表明,该方法是可行的,并且效果良好。Ceramic construction information is very important to ceramic quality analysis and manufacture control in ceramic industry. It is necessary to reduce noises and enhance edges for ceramic microscopic image processing at first. Curvelet transform is a new extension to wavelet transform in two dimensions. The directionality feature of curvelet transform makes it a good choice for representation of curves and edges in the image. The fast discrete curvelet transform theory makes it understood and implemented more easily. In this paper, an enhancement method is proposed to reduce the noises and enhance the edges by fast curvelet transform. This method is used for ceramic images pre-processing, and then watershed algorithm is applied to the segmentation. The grain size distributions can be obtained from segmentation images. It has been proved that this method is effective for the ceramic grain image.

关 键 词：陶瓷显微图像 快速曲波变换 图像去噪 图像增强 分水岭算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]