计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (10): 184-186.DOI: 10.3778/j.issn.1002-8331.2010.10.058

• 图形、图像、模式识别 • 上一篇    下一篇

路面破损图像的模糊增强方法研究

欧阳琰1,陈先桥1,初秀民2   

  1. 1.武汉理工大学 计算机科学与技术学院,武汉 430060
    2.武汉理工大学 水路公路交通安全控制与装备教育部工程技术中心,武汉 430060
  • 收稿日期:2008-09-27 修回日期:2008-12-05 出版日期:2010-04-01 发布日期:2010-04-01
  • 通讯作者: 欧阳琰

Research on fuzzy-enchancement algorithm of road damaged image

OUYANG Yan1,CHEN Xian-qiao1,CHU Xiu-min2   

  1. 1.Computer Science and Technology College,Wuhan University of Technology,Wuhan 430060,China
    2.Engineering Research Center of Transportation Safety,Ministry of Education,Wuhan University of Technology,Wuhan 430060,China

  • Received:2008-09-27 Revised:2008-12-05 Online:2010-04-01 Published:2010-04-01
  • Contact: OUYANG Yan

摘要: 针对传统路面破损图像模糊增强算法增强效果不明显、无法划分与背景灰度区域重叠的病害区域、控制参量难以设置、运算复杂等问题,提出了一种改进的路面破损图像模糊增强算法。首先对图像进行分块,并设计一种新的分段隶属度计算函数,使每个块区域内各像素点的灰度值完整细致地映射到隶属度上,能通过少量迭代次数获得一定的增强效果。在模糊逆映射上,采用灰度级扩展的逆变换函数,使增强效果进一步提升,消除了由于截断带来的灰度信息损失。新算法对多灰度级的路面破损图像的增强取得了很好的结果,并且控制参量无需人为设定。

关键词: 路面破损图像, 模糊增强, 模糊隶属度函数, 渡越点

Abstract: According to the shortcomings of traditional fuzzy-enhancement algorithm used to road damaged image,such as enhancement effect is not obvious,unable to divide the diseased region from background of gray area which overlapping the background,hard to set the control factor,difficult to compute.A new fuzzy-enhancement algorithm is proposed to overcome these drawbacks.By designing a new fuzzy membership function which makes the pixel gray value in the divided image fully and meticulously reflect to membership,the enhancement effects can be obtained during a few times of computer.In the fuzzy inverse mapping,the use of gray expansion of the inverse transforms function can improve the enhancement effect;eliminate the cut-off gray-scale brought about by the loss of information.The new algorithm for enhancing multi-gray-scale image of the damaged road surface can access to a very good result,and the control parameters don’t need man-made settings.

Key words: road damaged image, fuzzy-enhancement, fuzzy membership function, crossover point

中图分类号: