针对汽车发动机曲轴沟槽检测难度大、效率低等问题，提出了一种基于影像测量原理的曲轴沟槽检测方法，实现了沟槽 R角、宽度和深度三个特征参数的自动、快速、精密检测，并在此基础上开展工程应用。介绍了测量原理及检测方法；详细阐述了沟槽影像检测特征参数的自动提取技术，即面向曲轴沟槽特征影像的图像处理与数据处理方法，包括感兴趣区域提取、亚像素级边缘提取、几何基元的分割与拟合、关键几何特征提取等；基于检测样机平台，对检测方法进行了实验验证，并实现其工程应用。

In order to solve the great difficulties and problems such as low efficiency and low accuracy in measuring the groove of crankshaft in automotive engine, a new method of crankshaft groove measurement based on shadow castingtechnology was proposed. The characteristic geometric size including radius, width and depth of crankshaft groove could be measured automatically, rapidly and precisely with this method. The measuring principle and process were introduced, and the key image processing algorithms and data processing methods including region of interest extraction, subpixel-precise edge extraction, segmentation and fitting of geometric primitives, characteristic size extraction, etc. were all presented in detail. This method has been verified by the crankshaft groove measuring system developed by us, and has achieved its engineering applications.