目的 发展高性能、低成本基于图像处理技术的钢丝绳缺陷识别方法，实现钢丝绳磨损、锈蚀缺陷的检测。方法 首先采用邻域平均法提取钢丝绳的灰度区域特征，接着建立了基于熵统计函数以描述、量化灰度分布特征与灰度波动特征，在此基础上，选取良品、锈蚀缺陷与磨损缺陷三种类型的样品进行区域灰度、灰度分布、灰度波动特征的量化与提取。结果 经观测得到不同类型的样品在三维特征空间分布中具有明显的可区分性，基于该特点，可通过设置三维特征的阈值实现对钢丝绳合格品，锈蚀缺陷，磨损缺陷进行识别与区分。结论 该基于图像处理的检测方法可高效、准确地实现了对钢丝绳的锈蚀、磨损缺陷的识别，不仅具有学术价值，更具有实用意义，非常适用于在线检测。

The work aims to realize efficient inspection of wear and corrosion defects present in wire ropes by developing a high-performance and low-cost image processing technology-based defect rocognition method of wire ropes. Gray area features of the wire ropes were firstly extracted by neighborhood average algorithm. Then a entropy-based statistical function was established to describe and quantize grayscale distribution features and grayscale fluctuation features. On this basis, no-defective product, corrosion defect and wear defect samples were selected to quantize and extract regional grayscale, grayscale distribution and grayscale fluctuation features. The distribution of different types of sample in 3-dimensional characteristic space was observed to have obvious distinguishability. Based on this feature, corrosion and wear defects were detected and recognized by setting 3-dimensional characteristic threshold. The experimental result demonstrates that the image processing-basedinspecting method proposed in this paper can recognize corrosion and wear defects efficiently and accurately and is of both academic value and practical meanings, hence it is very fit for real time inspection.