飞行器软轴滚珠检测技术研究

Research on Ball Detection Technology of Aircraft Flexible Shaft

为解决飞行器滚珠式软轴的滚珠检测中使用传统人工或者半自动方法带来的大误差,长耗时等缺点,对飞行器软轴滚珠检测技术进行了研究,设计了一种基于图像形态学的滚珠自动识别算法.首先在实时成像中引入窗技术以提高图像的动态范围,再针对软轴图像过长的特点设计了一次成图技术,最后通过实验比较了多种主流图像分割方法,并据此开发了适用于滚珠式软轴的基于图像形态学的滚珠自动识别算法.实验结果表明,与传统检测方法相比,使用新方法后缺漏滚珠识别率提高了8.4%,计数分析时间也缩短为原先的3. 5%,更好的满足了工业检测要求.

To solve the inefficient and low-accuracy disadvantage of the ball detection of aircraft ball-type flexible shaft based on traditional manual or semi-automatic method,the technology of ball detection of aircraft flexible shaft is studied,and automatic recognition algorithm of balls based on image morphology is proposed. In this paper,first of all,tynamic range of images is improved by windowtechnique. Then one-time mapping technology is used to capture images considering the feature of of image’s large length. At last,automatic recognition algorithm of balls based on image morphology is proposed by comparing main image segmentation methods. Experimental results shows,the recognition rate of missing-balls is improved by 8. 4% and the ball counting time is reduced to3. 5% after using the newmethod. The newmethod meet the requirements of industrial detection better.