基于霍夫变换的烧结矿粒度识别研究

针对烧结矿生产时无法直接得到粒度大小和分布,人工检测的准确性和即时性不高等问题,提 出了一种基于图像增强和霍夫变换的烧结矿粒度识别方法。 该方法首先使用形态学开操作、图像像素点分 割、拉普拉斯图像锐化算子等方法进行图像增强,然后应用高斯滤波和图像边缘检测算法,最后用霍夫圆检 测算法进行烧结矿粒度检测,实时处理获取的图像,并检测出烧结矿的粒度大小和分布。 该方法可以快速检 测出图像中的烧结矿,其中图像像素点分割方法是根据烧结矿和背景的像素值设置分段函数进行分割,大幅 度减少图像中的噪声,提升了烧结矿和背景的对比度以及亮度,并且检测的准确性和即时性高,克服了人工 检测的弊端,准确率可达到 98%以上。 通过实验表明:该方法对提高烧结矿的生产效率、改善资源的利用、 降低人员成本具有积极作用。

Study on Particle Size Identification of Sinter Based on Hough Transform

In view of the fact that the size and distribution of sinter in production cannot be obtained directly and the accuracy and timeliness of manual detection are not high a sinter particle size identification method based on image enhancement and Hough transform is proposed. Firstly morphological open operation image pixel segmentation and Laplacian image sharpening operator are used for image enhancement. Then Gaussian filtering and image edge detection algorithm are applied. Finally Hoff circle detection algorithm is used for sinter particle size detection. The acquired images are processed in real time and the sinter particle size and distribution are detected. This method can quickly detect the sinter in the image and the image pixel segmentation method is based on the segmentation function of sinter and background pixel values which greatly reduces the noise in the image and improves the contrast and brightness of sinter and background and has high accuracy and timeliness of detection. This method overcomes the shortcoming of manual detection with an accuracy of up to 98%. The experiment shows that this method has positive effect on improving sinter production efficiency improving resource utilization rate and reducing personnel cost.