基于数字图像识别的岩土体裂隙形态参数分析方法

Analytic method of morphological parameters of cracks for rock and soil based on image processing and recognition

为了准确高效地测量岩土体裂隙的各形态参数,结合计算机数字图像处理技术,提出一整套裂隙图像计算机识别和定量分析方法。通过对含有裂隙的图像进行二值化、桥接、去杂、智能识别等操作,获取裂隙网络节点以及各裂隙的主干,进而提取裂隙的宽度、长度、方向等裂隙形态参数,实现了裂隙图像的计算机定量分析。在此基础上提出一种评价裂隙网络连通性的方法。基于该技术思路开发的岩土体裂隙图像处理系统(CIAS)被成功应用于土体干缩裂隙图像的识别和形态定量分析研究中。研究表明,该方法可以更加科学、高效地提取岩土裂隙形态参数,为裂隙的定量分析和评价提供了可靠依据。

In order to measure the morphological parameters of cracks for rock and soil accurately and quickly, with the application of computer image processing technology, a set of recognition and quantitative analytic methods of crack images were introduced. Nodes and skeleton of crack network were traced out by the operations of binarization, crack restoration, noise reduction and intelligent recognition. Morphological parameters of cracks, such as length, width and direction, were computed by further recognition methods. Furthermore, a new method was proposed to evaluate the connectivity of crack network. A software, Cracks Image Analysis System （CIAS）, developed based on these technologies, was applied to the recognition and morphological quantitative analysis of soil crack images. The proposed method offered a reliable basis for the quantitative analysis and recognition of cracks, and thus the morphological parameters could be obtained more scientifically and efficiently.