The human dimension information is crucial for efficient building energy saving, health and productivity, comfort conditions and security management. A great number of studies have been developed for occupancy information measurement. However, existing review work has limited coverage on emerging images/videos based methods. In this paper, many occupancy measurement systems based on different sensors are reviewed, especially images/videos based methods. We conduct a comprehensive analysis based on different types (non-depth or depth cameras) and different installed locations (the room entrance or interior) of cameras. Considering the motion and static image information, we categorize these studies and compare the merits and limitations. As for other sensors, Wireless Fidelity (WiFi), Passive Infrared (PIR) sensors, carbon dioxide (CO2) sensors, electricity meters, this paper analyzes and discusses their applicable scopes and limitations. Sensor fusion method tends to perform better because different sensors can compensate each other. Moreover, future trends are presented, including the fifth generation mobile network (5G), cloud computing platform and artificial intelligence especially deep learning technologies

人的维度信息对于有效的建筑节能，健康和生产力，舒适条件和安全管理至关重要。已经针对占用信息测量进行了大量研究。但是，现有的审查工作对基于新兴图像/视频的方法的覆盖范围有限。在本文中，回顾了许多基于不同传感器的占用测量系统，尤其是基于图像/视频的方法。我们根据不同类型的摄像机（非深度或深度摄像机）和摄像机的不同安装位置（房间入口或内部）进行全面分析。考虑到运动和静态图像信息，我们将这些研究分类并比较其优缺点。至于其他传感器，无线保真（WiFi），被动红外（PIR）传感器，二氧化碳（CO2）传感器，电表，本文分析并讨论了它们的适用范围和局限性。传感器融合方法往往表现更好，因为不同的传感器可以相互补偿。此外，还提出了未来趋势，包括第五代移动网络（5G），云计算平台和人工智能，尤其是深度学习技术。