基于圆周灰度分布互相关的颗粒转速判别方法

Particle Rotation Speed Measurement Method Based on Cross-correlation of Gray Distribution

开发了一种基于颗粒圆周灰度分布互相关的计算机算法用于所拍流化床流场图像中颗粒转速的判别。该方法利用数字图像处理技术进行图像中颗粒的识别、筛选和定位,进而根据颗粒的圆周灰度分布在两幅图像中的变化计算颗粒的转速。通过对已知转速和旋转方向的小球的验证实验表明:所提出的转速判别方法能够比较准确地测量旋转轴垂直或近似垂直于拍摄平面的颗粒,并有效地过滤掉其他颗粒的干扰,保证了测量的可靠性。对转速测量的误差进行分析,提出了灰度修正方法以减弱颗粒灰度分布不均对转速测量的影响,指出考虑各因素后,所开发的转速判别方法的误差小于10%。

To automatically measure the rotation speed of moving particles from their images taken in a cold CFB riser with a high speed digital camera, a method based on cross-correlation of gray distribution of particle images was presented. The particles were identified, screened and located from the image sequence by using several digital imaging processing methods, and the rotation speed was determined by calculating the variation amount of particle＇s circular gray distribution in two consecutive images. A verification experiment by using a small sphere with known rotation speed and spin axis was also carried out. It is shown that the method is capable of measuring accurately the rotation speed when the direction of particle＇s spin axis is （nearly） perpendicular to the imaging plane. And meanwhile, the measurement data for other particles can also be filtered by using an appropriate threshold of correlation coefficient. A gray amendment method is used to eliminate the effect of non-uniform gray distribution of particle image on rotation speed measurement. The error for the measurement method is found to be smaller than 10%, which indicates a high reliability.