磨粒群盒维数计算过程中的参数优选法

A Parameter Optimization Method of Box Dimension Calculation for Wear Particle Group

针对磨粒群盒维数计算精度偏低的问题,提出盒维数计算过程中的参数优选法。研究图像大小、盒子尺寸边界、统计盒子方式、盒子尺寸增长方式对分形维数的影响规律,利用正交试验确定各因素对盒维数影响的主次顺序,得到影响因素的参数优选组合。计算验证表明:图像大小为500×500像素,盒子尺寸从1像素增长至整个图像大小,统计盒子引入的比例参数为0.4,盒子尺寸以2,4,6,8,…,2n等差增长时,盒维数计算精度能得到有效提高。

To solve the lower accuracy problem using box counting method for wear particle group fractal dimension cal- culation, a parameter optimization method in box dimension calculation was proposed. The effects of the factors such as the image size, the box size boundary, box-counting way, box size growth mode on the fractal dimension were analyzed. The primary and secondary order of the factors influencing the calculation accuracy was determined by an orthogonal test, the parameter optimization combination was acquired. The results show that, with image size being 500 x 500 pixel, box size boundary ranging from one pixel to the entire image size, scaling parameter being 0. 4, box size growing with arithmetic mode, the box dimension calculation accuracy is improved effectively.