基于X射线气缸套的快速自动检测方法

Rapid automatic detection method of cylinder liner based on X-ray

为了能够对气缸套进行快速自动识别,提供了一种基于气缸套有无缺陷的快速自动检测方法。首先将待检产品内部最小目标置于最大旋转步长上进行全方位成像,提取各个待识别子目标或可变特征进行数据重排,对图像中不变子目标进行提取,分别建立特征和位置标准样本库。在识别中,根据其所在位置确认待件产品在特征标准库中的最优解,这样可以快速删除大量无缺陷的气缸套。对于部分存在缺陷的气缸套,采用改进的最大相关法。同时为了在降噪的过程引起不必要的失真,采用形态学滤波作进一步处理。实验表明,相比传统的技术而言,能更准确快速地对气缸套进行识别。

In order to identify the cylinder liner quickly and automatically.This paper provides a fast automatic detection method based on whether the cylinder liner is defective or not.Firstly,the minimum target inside the product to be inspected is placed on the maximum rotating step for omni-directional imaging,each sub-target or variable feature to be identified is extracted for data rearrangement,and the invariant sub-target in the image is extracted,and the feature and the variable sub-target are established respectively.Location Standard Sample Library.In recognition,confirm the optimal solution of the product in the feature standard library according to its location,which can quickly delete a large number of flawless cylinder liners.For some defective cylinder liners,the improved maximum correlation method is adopted.At the same time,in order to cause unnecessary distortion in the process of noise reduction,morphological filtering is used for further processing.Experiments show that compared with traditional technology,this paper can identify cylinder liner more accurately and quickly.