中文摘要：

为有效检测结构物表面裂缝形态,基于图像分析技术,采取在结构表面黏贴纯色标定块,并借此通过投影变换来修正和标定原始图像,对修正后的图像进行骨架化提取、去除分支和毛刺等一系列处理后直接在图像上确定裂缝特征参数.基于所提方法,采用不同长度、宽度和倾角的线条来模拟裂缝,试验验证了图像修正和标定算法的有效性,并研究拍摄角度和光照条件对裂缝特征参数检测精度的影响.研究结果表明：基于不同角度拍摄的竖向和斜向模拟裂缝长度的识别精度分别为99.73%和99.64%,对宽度不小于0.2mm裂缝宽度的识别精度为95.84%和95.60%,表明该方法具有较高的检测精度;光线强度对裂缝宽度精度影响较大,光照强度在1400-3100lux时,识别精度较好,裂缝宽度识别的平均相对误差小于5%.

英文摘要：

This paper aims to find a efficient method for detecting the crack shapes on the structural surface. Based on the application of image analysis technology, a standard square attached on the structural surface is used as a reference mark to calibrate the original deformed images through the projection transformation, and then extracting the crack skeleton and removing the branches and burr, etc. Finally, the crack characteristic parame-ters in the image can be calculated. This method is verified by the simulated crack with lines of different length, width and angle of inclination and this paper also studies the influence of shooting angle and light condition on the detection accuracy of the crack characteristic parameters. Based on different shooting angles, the length iden-tification accuracy of vertical and oblique simulation crack can reach 99.73% and 99.64% respectively, and when the width is not less than 0.2 mm, the recognition accuracy of width is 95.84% and 95.60% respectively, which shows that this new method has high detection precision. The light intensity has a great influence on the crack width identification. When the light intensity is 1 400-3 100 lux, the average relative error of the crack width identification is less than 5%.