(完整word版)人体跌倒检测仪设计与实现.doc

基于模式识别的人体跌倒检测仪设计与实现摘要我国是世界上人口最多的国家，目前全国人口总数接近 14亿，受社会发展 的实际情况影响，我国除了是人口基数庞大的国家之外， 还逐年步入了人口老龄 化社会，通过数据统计就可以看到老龄化程度的不断加深， 老年人口数量占全国 总人口数量的比重将不断增长，老年人口基数日益庞大本次论文的研究目的就是要探索人体跌倒检测仪的设计理论和具体优化的 实施对策，通过实际的案例分析证明来检测设计的合理性， 并反馈出更为优化的 设计内容本次研究的理论意义在于：将仪器设计与现代化信息与经济发展融合 起来，从创新研发的角度来讲符合现代产业运作的内在要求， 切合受众接受的购买需求，其理论意义是填补检测仪设计空白的 实践意义在于：在我国人体跌倒 检测仪的研制和开发仍处于探索和试验阶段， 国内尚未真正出现具有独立自主研 发的检测仪设施，其实践意义对于提升老年人护理，减少社会养老负担等都十分 重要本次论文研究的主要内容包括六个部分： 第一部分，对所研究的内容、背景 意义、方法影响等进行了总结，并针对“跌倒行为”、“识别传感”、“系统应 用”等内容进行了文献综述整理第二部分，对相关概念进行了论述，包括：模 式识别、跌倒、跌倒检测仪等。

第三部分，开展了人体跌倒行为的特点研究第 四部分，进行了人体跌倒检测算法及实验研究 第五部分，提出了模式识别下人 体跌倒检测仪的设计策略第六部分，进行了总结和展望本次论文研究通过理论分析、实验研究、策略总结的基本过程，对模式识别 理论下的人体跌倒检测仪设计进行了相关论述， 最终得出如下结论：第一，模式识别下的人体跌倒检测仪设计离不开对重点概念的理解， 重点的概念内容包括跌 倒、模式识别和跌倒检测仪三方面第二，人体跌倒行为的特点的研究之中分析 了当前人体骨骼系统及特点，老年人跌倒行为模式与特点，为人体跌倒检测仪的 设计提供了人体模型的构建基础理论第三，通过人体跌倒检测算法及实验研究， 总结人体跌倒检测算法、阈值判断检测算法、模式识别检测算法的相关模式， 在实验结果的分析下证明了基于模式识别的计算方法，设计的人体跌倒检测仪要 考虑到跌倒时的具体身体表现数据和相关的敏感度， 提出了模式识别的多种计算机技术方法，这样设计出来的产品能够有效区分跌倒与其它日常生活行为动作， 避免出现误报的情况，干扰使用者的日常生活第四，提出了模式识别下人体跌 倒检测仪的设计策略： 应该完善模式识别下人体跌倒检测仪的设计实施过程， 进 行完备的信号数据处理， 构建完备计算的信号向量模， 充分利用现代化网络体系。

关键词：人体跌倒检测仪；模式识别；实验；对策On Human fall detector design and implementation of Pattern RecognitionAbstractChina is the world's most populous country, currently the country's population close to 1.4 billion, by the impact of the actual situation of social development, in addition to our huge population base country, year after year also entered the aging society, through statistics on You can see the degree of aging of deepening the elderly population accounts for the proportion of the number of the country's total population will continue to grow, increasingly large elderly population base.The purpose of this research paper is to explore the human fall detector design theory and implementation of countermeasures specifically optimized through actual case studies to prove to detect design is reasonable, and feedback the more optimized design content. Theoretical significance of this study lies in: the instrument design combines it with modern information and economic development, innovation and research and development from the perspective of compliance with the inherent requirements of a modern industrial operation, to meet the needs of the audience to accept the purchase, its theoretical significance is designed to fill the detector blank. Practical significance lies: in our human fall detector research and development is still in the exploration and testing phase, is not yet really the detector with independent research and development facilities, and its practical significance for enhancing elderly care, reduce social pension burden so importantThe main contents of this thesis consists of six parts: The first part of the research content, meaning the background, methods, impacts are summarized, and for the "fall behavior", "Recognition Sensor", "System", etc. conducted a literature review finishing. The second part of the relevant concepts were discussed, including: pattern recognition, fall, fall detector. The third part, to carry out the characteristics of the study of human behavior in the fall. The fourth part, were human fall detection algorithm and experimental research. The fifth part of pattern recognition is proposed under human fall detector design strategy. Part VI, a summary and outlook.This paper studies the theoretical analysis, the basic process of experimental research, strategy summary, the human body fall detector design pattern recognition theory under discussion were related, the final conclusions are as follows: First, the pattern recognition under the human fall detection device designed to focus on conceptual understanding is inseparable from the concept of the key points include falls, pattern recognition and fall detector in three areas. Second, the study of human behavior falls into the analysis of the current characteristics of the human skeletal system and features, behavior patterns and characteristics of falls in older people, as human fall detection device designed to provide a theoretical basis for constructing mannequin. Third, through human detection algorithm and experimental research fall, summarizes the human fall detection algorithm to determine the threshold detection algorithm, pattern recognition algorithms to detect the correlation model, in the analysis of experimental results prove that the calculation method based on pattern recognition, the design of the human fall detector to take into account the performance of specific body fall when the sensitivity of the data and the associated proposed pattern recognition method for a variety of computer technology, such products designed to effectively distinguish fall activities of daily living and other actions to avoid false positives cases,interfere with the user's daily life. Fourth, the proposed pattern recognition under human fall detector design strategy: should improve the human fall under pattern recognition detector。