| Summary: | 碩士 === 淡江大學 === 資訊工程學系資訊網路與通訊碩士班 === 102 === With the rapid development of sensor technologies, along with the increasing popularity of mobile devices and wireless/mobile networks, the volume of data generated by human beings and all sorts of devices are getting larger and larger every day. It is without doubt that how to deal with the huge amount of data in an efficient way and to transform these data into useful information for people to make use of has become an important research topic. In this thesis, we focus on handling the large amount of environmental conditions data collected by various sensor devices. These data produced around us such as the temperature, the road conditions and the air quality can be numerically analyzed by utilizing the cloud computing technology. Therefore, we implemented a sensor data processing platform for intelligent cities based on Hadoop. We assume that sensor data are embedded in the image files captured by the vehicle drive recorders and the smartphones. With Android smartphones, users can upload the image files to the Hadoop cluster by the Separatags Android App we developed. After the image files are uploaded, we use the MapReduce framework to process them. Specifically, in the Map task we utilize the well-developed Hadoop Image Processing Interface (HIPI) library to extract the desired sensor data from the image files, and then in the Reduce task these sensor data are inserted into HBase. Besides, we use the Hadoop Distributed File System (HDFS) to store the street images captured by driving recorders installed in vehicles. People can then use their Android smartphones or standard web browsers to access the sensor data and the street images. In sum, the data processing platform we developed can be an important building block for constructing various useful and creative applications to serve people living in intelligent cities.
|
|---|
