基于X射线钢芯传送带图像的缺陷检测算法

Defect Detection Algorithm Based on X-Rays Steel-Core-Belt Images

钢丝绳芯输送带是煤矿的主要传输设备,然而煤炭生产常常会因为钢丝绳缺陷引起的崩断而中断.因此,为了保证安全生产,防止恶性事故发生,需要检测钢丝绳芯输送带的缺陷情况.本文以线性X射线阵列采集运行钢丝绳芯皮带的图像序列,根据其图像纹理特征,提出了一种线状局部二进制模式纹理编码算子(line-shaped LBP)检测钢丝绳芯图像缺陷,并在VC++平台上实现了钢丝绳芯输送带在线缺陷检测、存储、报告打印和显示缺陷的目标.大量实验结果表明,该算法检测精度平均值可达到90%.

The steel cord conveyor belt becomes the major coal transport equipment in coal mining enterprises and has become increasingly popular.But the production of the coal is often interrupted by the fraction of steel cords due to their defects.Therefore,in order to ensure safe production and prevent fatal accidents of coal,we need to detect defects of the steel cord conveyor belt.In this paper,we use Xrays linear detector array to collect image series of steel-rope-core-belt and propose a line-shaped local binary pattern texture coding operator（line-shaped LBP）to detect defects of the steel-cord-belt according to their texture features.We also implement an online defect detection system of steel cord conveyor belt on which defects can be detected,stored,report printed and targets displayed with VC＋＋ platform.A large number of experimental results show that the average precision of the algorithm to detect defects can reach up to 90%.