中南大学学报（自然科学版）2018，Vol.49Issue(1)：118-123,6.DOI:10.11817/j.issn.1672-7207.2018.01.016
一种改进的高斯混合模型煤矸石视频检测方法
An improved Gaussian mixture model for coal gangue video detection
摘要
Abstract
A new approach was put forward to realize the detection of coal gangue via the monitoring video of conveyer belt. Considering the complex scene and the poor video quality of the conveyer belt, coal gangue was detected and recognized by an improved Gaussian mixture model (GMM) which extracts and subtracts the background of the video. In order to improve the algorithm performance, the particle swarm optimization was employed to find the better parameters of GMM. The results show that the average discrimination ratio is 95.38％. The proposed method can effectively detect coal gangue in the coal flow on the conveyer belt, which is good for improving the quality of coal and the safe operation of conveyer belt.关键词
皮带运输机/高斯混合模型/粒子群优化算法/煤矸石检测Key words
conveyer belt/Gaussian mixture model/particle swarm optimization/coal gangue detection分类
信息技术与安全科学引用本文复制引用
程健,王东伟,杨凌凯,张美玲,郭一楠..一种改进的高斯混合模型煤矸石视频检测方法[J].中南大学学报（自然科学版）,2018,49(1):118-123,6.基金项目
国家重点研发计划项目(2016YFC0801406) （2016YFC0801406）
江苏省六大高峰人才项目(2017-DZXX-046) （2017-DZXX-046）
中国矿业大学学科前沿研究专项(2015XKQY19) (Project(2016YFC0801406) supported by the National Key Research and Development Program （2015XKQY19）
Project(2017-DZXX-046) supported by Six Talent Peaks Project in Jiangsu Province （2017-DZXX-046）
Project(2015XKQY19) supported by Research Program of Frontier Discipline of China University of Mining and Technology) （2015XKQY19）
