基于RFID技术的双出口和单通道行人运动参数提取方法研究

发布时间：2018-03-11 13:29

本文选题：疏散　切入点：RFID　出处：《中国科学技术大学》2014年硕士论文　论文类型：学位论文

【摘要】：当发生火灾等灾害时,怎样有效地疏散人员,对于减少伤亡至关重要。人员运动特征是研究人员疏散的基础,而人员运动参数是研究人员特征的关键。人工统计和视频处理是提取行人运动参数的常用方法,但前者效率比较低下,后者拍摄时易受场地和环境限制,如在黑暗,有烟、雾等环境下,摄像机难以达到理想的拍摄效果。目前,RFID技术快速发展,在各领域广泛应用。因此,本文研究了如何将RFID技术应用于行人运动实验,并得到了一种快速提取行人运动基本参数的新方法。进而,组织了人员运动出口选择实验,以及可见光和非可见光两种场景下的单通道行人运动实验；根据提出的RFID方法得到出口人数、出口时间、出口流量与流率、行人速度等基本参数,并将该方法得到的行人运动基本参数与视频处理得到的参数进行了比较验证。通过实验结果的比较发现,90%的出口时间相对误差绝对值都小于0.08,所有的出口时间相对误差绝对值小于0.10；大约90%的出口流量相对误差的绝对值小于0.20；一个出口平均流量与平均流率的相对误差绝对值不超过3.55%,另一个出口平均流量与平均流率的相对误差绝对值不超过8.38%；几乎所有速度的相对误差绝对值小于0.30,大约70%的速度相对误差绝对值小于0.20。我们比较了RFID方法提取的两种场景下的单通道行人速度,大部分行人在可见光条件下的速度大于非可见光条件下的速度,这种情况与实际情况相符合。通过人员出口选择实验和单通道行人运动实验,验证了可以通过RFID技术快速有效提取行人运动基本参数,而且,本文提出的方法克服了传统的人工统计和视频处理方法的缺点与不足,这对人员疏散的研究具有重要的意义。
[Abstract]:How to evacuate people effectively when fire and other disasters occur is very important to reduce casualties. The characteristics of human movement are the basis of evacuation. Artificial statistics and video processing are common methods for extracting pedestrian motion parameters, but the former is inefficient, while the latter is vulnerable to site and environment constraints, such as dark, smoky, etc. In fog and other environments, it is difficult for the camera to achieve the desired shooting effect. At present, RFID technology is developing rapidly and widely used in various fields. Therefore, this paper studies how to apply RFID technology to pedestrian movement experiment, and obtains a new method to extract the basic parameters of pedestrian movement quickly. The experiment of selecting the exit of human movement and the experiment of single channel pedestrian movement under the two scenarios of visible and non-visible light were organized. According to the proposed RFID method, the number of people, the time of exit, the flow rate and the flow rate of exit were obtained. The basic parameters such as pedestrian velocity and so on are compared with those obtained by video processing and the basic parameters of pedestrian motion obtained by this method are compared with those obtained by video processing. Through the comparison of experimental results, it is found that the absolute value of 90% exit time relative error is less than 0.08, the absolute value of all exit time relative error is less than 0.10, the absolute value of about 90% exit flow relative error is less than 0.20, and one exit level is equal. The absolute value of relative error between average flow rate and average flow rate is not more than 3.55 and the absolute value of relative error between average flow rate and average flow rate in another outlet is not more than 8.38; the absolute relative error value of almost all velocities is less than 0.30, which is about 70% velocity phase. The absolute value of error is less than 0.20. We compare the pedestrian velocity of single channel in two scenarios extracted by RFID method. The speed of most pedestrians in visible light is higher than that in non-visible light, which is consistent with the actual situation. It is proved that the basic parameters of pedestrian movement can be extracted quickly and effectively by RFID technology through exit selection experiment and single channel pedestrian movement experiment. The method proposed in this paper overcomes the shortcomings and shortcomings of the traditional methods of manual statistics and video processing, which is of great significance to the study of evacuation.
【学位授予单位】：中国科学技术大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：X913.4

【相似文献】