中文摘要：

目前,移动终端数据采集通常采用灾害要素分离存储、数据库关联管理的方式,其采集、数据管理过程繁琐且不便于数据同步传输。针对上述不足,提出一种有声可定位图像数据模型。实现集地理位置、属性信息、图像、语音等多源异构采集要素一体化综合表达;采用EXIF图像格式中的APP1-APP2标记段,以键-值方式对相关参数进行组织存储;介绍Android环境下语音的采样、IMA-ADPCM编码及EXIF格式的存储方法,在此基础上实现有声可定位图像移动采集系统。实测验证了该采集方法是便捷可行的,可准确、详尽、多方位记录灾害对象状态信息,能为灾害应急管理、分析应用提供较为全面完整的数据基础。

英文摘要：

In recent years,mobile terminal data collection usually adopts disastrous elements separate storage or relational database,which is complicated to get,manage and sharing these data.An acquisition method of the phonic-locatable image was proposed.This expression method contented many multi-source heterogeneous data,such as geographical position,attribute information,images and audio,and saved the data in APP1-APP2 marker segments of EXIF format using the key-value mode.The method of audio acquisition,IMA-ADPCM coding and EXIF information storage were introduced,and an acquisition system of the phonic-locatable image was realized.The test verifies the usability of the proposed method.