摘要
随着数字图像处理技术的发展,图像识别开始应用到煤矿的安全领域。基于背景差分法的研究,采用混合高斯模型的背景建模法,设计了井下视频图像中运动人员检测的流程和算法,来应对监测场景图像、光线强度的不断变化,并开发了实时监测平台,实现了对井下监控图像中人员进行识别和运动轨迹跟踪。实际应用结果验证,该算法能有效增强检测的效果,实现了在诸多不确定性因素的序列视频中较好的自适应构建背景,从而高效地检测出移动中的井下工作人员,并进行标注跟踪显示。
With the development of digital image processing technology, image recognition has been applied to the safety field of coal mine. Based on the research of background difference method and background modeling method of Mixture Gauss Model, this paper designs the process and algorithm of moving personnel detection in underground video images to cope with the continuous changes of monitoring scene images and light intensity, and develops a real-time monitoring platform, which realizes the recognition of personnel in underground monitoring images and trajectory tracking. Practical application results show that the algorithm effectively enhances the detection effect and achieves a better adaptive background construction in the sequence video of many uncertain factors, so as to effectively detect the moving underground workers and tag tracking display.
作者
王勇
Wang Yong(China Coal Research Institute,Beijing 100013,China;State Key Laboratory of Coal Mining and Clean Utilization,Beijing 100013,China;Research Center of Mine Safety Engineering and Technology,Beijing 100013,China)
出处
《电子测量技术》
2020年第1期28-31,共4页
Electronic Measurement Technology
基金
国家重点研发计划(2018YFC0808304)
煤炭科学技术研究院科技发展基金(2019CX-I-09)项目资助。
关键词
视频监控
人员图像识别
背景差分法
煤矿
Video surveillance
Personnel Image Recognition
Background subtraction method
Coal Mine
