中文摘要：

由于传统的定位系统受到慌乱中人员走动及火灾烟雾的影响，难以从采集的火灾图像中对受困人员进行准确实时识别，导致定位效果差。设计并实现了一种新的火灾图像中受困人员的定位系统。该系统主要由硬件和软件两部分组成，系统硬件的核心部分是ARM图像采集器以及S3C44BOX嵌入式处理器；主要通过MC9S12NE64以太网控制模块控制ARM图像采集器，将采集的图像传输到S3C44BOX嵌入式处理器进行处理，并对附件模块即通信模块进行了优化分析。软件设计部分主要对整个系统工作流程及以太网控制模块应用软件进行设计；最后通实验对设计的系统进行验证。实验结果表明，利用此系统进行火灾图像中受困人员定位，能够极大地提高受困人员定位的准确性，提高人员救治率。

英文摘要：

The traditional positioning system is difficult to recognize the trapped people in fire image timely and accurately due to the influence of the walking people and fire smoke. Therefore, a new trapped people positioning system in fire image was designed and implemented. The system is composed of hardware and software mainly. The core parts of the system hardware are the ARM image collector and S3C44BOX embedded processor. The Ethernet control module MC9S12NE64 is used to control the ARM image collector, transmit the acquired image to the S3C44BOX embedded processor for processing, and conduct an optimized analysis of the communication module. The working process of the whole system and the application software of Ethernet control module were designed. The designed system was verified in the experiment. The experimental results show this system used to locate the trapped people in the fire image can improve the accuracy of the trapped people positioning greatly and cure rate.