基于机器视觉的煤矸石检测研究

Research on Coal Gangue Detection Based on Machine Vision

针对传统煤矸石分拣机器人对输送带上不同位置煤矸石感知能力不足的问题,提出使用单激发多盒探测器(SSD)机器视觉算法对煤矸石进行检测的方法,提高分拣机器人的感知与分拣性能。实验结果表明,基于SSD的煤矸石检测算法实时性优于Faster-RCNN检测算法,且具有更高的准确性,进一步促进了煤矸石分拣机器人的智能化。

Aiming at the problem that traditional coal gangue sorting robots have insufficient perception of coal gangue at different positions on the conveyor belt, a method of detecting coal gangue using single shot multibox detector(SSD) machine vision algorithm was proposed to improve the perception and sorting performance of sorting robot. Experimental results show that the real-time performance of the SSD-based coal gangue detection algorithm is better than the Faster-RCNN detection algorithm,and it has high accuracy, which further promotes the intelligentization of the coal gangue sorting robot.