计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (8): 170-171.

• 图形、图像、模式识别 • 上一篇    下一篇

纤维板孔穴显微图像的检测方法

张剑飞,陈桂兰,岳 新   

  1. 黑龙江科技学院 计算机与信息工程学院,哈尔滨 150027
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-03-11 发布日期:2011-03-11

Method in detecting void microscopic image of cross-section of fiberboard

ZHANG Jianfei,CHEN Guilan,YUE Xin   

  1. College of Computer and Information Engineering,Heilongjiang Institute of Science and Technology,Harbin 150027,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-03-11 Published:2011-03-11

摘要: 针对纤维板孔穴显微图像的特点,提出了一种快速、准确的孔穴图像检测的新方法;该方法利用最大类间方差法初步分割出图像中的孔穴,结合GAP统计模型,以局部区域内多方向灰度分布函数的差别来进行边缘细化,确定最终边缘。实验结果表明,算法在一定程度上加强边缘检测的效果,并且具有很好的抗噪特性,取得了良好的检测效果。

关键词: 图像分割, 最大类间方差法, 孔穴, GAP统计模型, 边缘检测

Abstract: Aiming at the feature of the void microscopic image of the cross-section of fiberboard,a new algorithm is proposed which can detect the image effectively and accurately.This algorithm uses Otsu algorithm to segment the void from the microscopic image,then refines the void edge by using the GAP model which is based on the difference in distribution function of grey-level.The experimental results show that the algorithm improves the edge detection effect of void microscopic image with a good anti-noise performance,and achieves better detection effect.

Key words: image segmentation, Otsu algorithm, viod, GAP statistic model, edge detection