一种新型钢球表面检测方法的分析与系统实现
Analysis and Implementation of Novel Method for Steel Ball Surface Detection

钢球作为常用的球轴承滚动体,其表面质量直接影响轴承的性能及寿命。目前已研制出的钢球表面缺陷检测装置展开机构结构复杂,不适用于大批量检测。基于一种双图像传感器的球面全展开方法,分析了在球列状态下球面完全展开所需要满足的参数条件。搭建系统进行实验,实验结果表明,该系统利用双传感器同时分别对每粒钢球进行5次采像就能够对钢球表面进行精确有效的高速检测与分选,漏检率低,检测速度可达到14400粒/h,具有广泛的应用前景。
Steel ball is the rolling element of the ball bearings, whose surface quality directly affect the performance and longevity of the bearing. The developed systems of steel ball surface damage detection have complex unfolding mechanisms, which is not suitable for mass inspection. This paper is based on the sphere surface unfolding method with double image sensors, and the parameters that can fully unfold the sphere surface when the balls are analyzed in a queue. A system is built to illustrate the feasibility of this method. The result shows that the system accurately detect and sort the balls at a high speed of 14,400 pcs/h only take 5 images with each image sensor