论文部分内容阅读
数字图像技术的发展使得路面病害破损的调查更加容易和安全,然而路面病害的自动识别技术还不能良好地适应当前复杂的路面情况。因此对传统的数字采集,人工参与判断的路面病害识别过程的效率提升尤为必要。针对这一需求,提出一种基于裂缝横断面灰度曲线图以及裂缝几何走势的裂缝信息提取方法。先获得横断面灰度曲线,从裂缝中心向两边寻找满足一定强度的边缘走势,由两端距离获得该断面裂缝宽度。通过一定数量的裂缝横断面信息综合得出该条裂缝的最大和平均宽度、裂缝长度等信息。该方法对于不同环境和人力参与下的裂缝判别有较好的稳定性,同时边缘强度的判别具有自适应功能。结果表明这种方法对于裂缝信息的识别有较好的效果。
The development of digital image technology makes the investigation of pavement damage easier and safer, however, the automatic identification technology of pavement diseases can not well adapt to the current complex road conditions. Therefore, it is necessary to improve the efficiency of the recognition process of pavement diseases by traditional digital acquisition and artificial participation. In response to this demand, a fracture information extraction method based on grayscale curve of fracture cross-section and geometric trend of fracture is proposed. Obtaining cross-sectional grayscale curve firstly, looking for the edge trend meeting certain strength from the center of the crack to both sides, and obtaining the width of the cross-section crack from both ends. Through a number of fracture cross-sectional information, the maximum and average width of the fracture, the length of the crack and other information are obtained. The proposed method has good stability for discriminating cracks under different environments and human interventions. At the same time, the discriminant of edge strength has an adaptive function. The results show that this method has good effect on the identification of fracture information.