基于机器视觉的矿石块度智能识别及应用

给矿粒度特性以及矿石可磨性对半自磨过程的影响远大于常规碎磨流程，通过对给矿皮带上矿石图像进行矿石粒度特性分析，可从给矿源头上实现矿石块度实时检测。基于多尺度的矿石图像分析系统，实时采集半自磨给矿矿石图像，通过图像处理技术，实现矿石块度有效分割以及粒级统计，为半自磨给矿矿石块度控制提供数据支撑。

Intelligent identification and application of ore size distribution based on machine vision

In the semi-autogenous grinding operation, the ore feed of the semi-autogenous grinding machine is usually solid materials with a wide particle size distribution. The material in the cylinder is used as the grinding medium, and the continuous and strong impact grinding and peeling is carried out in the cylinder. In order to achieve the purpose of grinding. Therefore, the influence of ore particle size characteristics and ore grindability on the semi-autogenous grinding process is much greater than that of the conventional crushing process. By analyzing the particle size characteristics of the ore on the ore feeding belt, the real-time detection of the ore lumpiness is realized from the source of the ore feeding. Based on the multi-scale ore image analysis system, this paper collects the images of semi-autogenous grinding feed ore in real-time, and realizes the effective segmentation of ore lumps and particle size statistics through image processing technology, which provides data support for the control of semi-autogenous grinding feed ore lumps.