摘要
作为主要成分的SiO_(2)是铁尾矿中最难熔化的部分。通过一组SiO_(2)的熔化时序图像来探究铁尾矿的熔化情况。首先,进行图像预处理,建立质心位置追踪模型和体积估算模型。将灰度图转化为二值图,利用卡尔曼滤波器进行质心追踪和对质心运动轨迹的描述。通过计算二值图中0像素的数量得到面积随时间变化的图像,拟合出熔化过程函数。结合二值图中目标物体的边界信息得到的参数量建立多元线性回归模型。估算颗粒扁平度和物体的体积,熔融速率通过颗粒集合的质量变化函数得出。
As the main component,SiO_(2)is the most difficult part of iron tailings to melt.A set of SiO_(2)melting time series images were used to explore the melting of iron tailings.First of all,the image preprocessing is carried out,and the centroid position tracking model and volume estimation model are established.Transform the grayscale image into binary image and use Kalman filter to track the centroid and describe the centroid trajectory.The image of area changing with time is obtained by calculating the number of 0 pixel in the binary graph,and the melting process function is fitted.According to the number of parameters obtained from the boundary information of the target object in the binary graph,a multiple linear regression model is established.The flatness of the particles and the volume of the object are estimated,and the melting rate is obtained from the mass change function of the particle set.
出处
《科技创新与应用》
2022年第3期50-52,共3页
Technology Innovation and Application
关键词
时序图像
卡尔曼滤波器
多元线性回归模型
质量变化函数
time series image
Kalman filter
multiple linear regression model
quality change function
