行为识别技术的研究与发展

Research and development of behavior recognition technology

行为分析有着广泛的应用前景,如智能视频监控、人机交互、自动识别报警、公共安全等方面,行为分析已成为相关领域的研究热点并有其潜在的经济价值。在人工智能和自动化操控迅速发展的当下,行为分析作为人工智能发展的中流砥柱也成为了国内外研究人员相继探讨的热点,关于人体行为分析的研究方式、模型算法和描述方法都取得了切实有效的发展。根据采用不同识别技术人体行为识别目前主流要分为四大类:基于计算机视觉的行为识别、基于传感器系统的行为识别、基于位置的行为识别和基于人物交互的行为识别。这篇文章主要探讨研究了行为识别技术和应用这2个方面的问题,综述了目前已有的技术情况,在探讨该方向各类技术的发展情况和研究现状的基础上,总结了当前行为分析仍然存在的问题和未来可能的发展前景。

Behavioral analysis has a wide range of applications, such as intelligent video surveillance, human-computer interaction, virtual reality, public security and so on, and it has great application prospects and economic value. Because of the urgent needs of artificial intelligence and behavior monitoring, the description methods and identification algorithms of behavioral analysis and understanding have developed rapidly and deeply. Human behavior recognition technology is mainly divided into four categories： behavior recognition based on computer vision, behavior recognition based on sensor system, behavior recognition based on location and behavior recognition based on human-device interaction. This paper mainly discusses the basic problems from such two aspects as behavior recognition technology and behavior recognition application, surveys the behavior recognition technology; after that, reviews the history and current development situation of all kinds of technical research. Finally, the paper summarizes the existing problems and future development direction of behavior analysis.