Study On Method Of Object-oriented Vehicle Extraction In Remote Sensing Images

For decades, remote sensing technology at home and abroad has made rapid development. Researchers have studied on information extraction of little targets on the ground in high-resolution remote sensing images. Remote sensing image has the unique advantage of ground observation on a large area while the other traffic monitoring tools cannot. For this reason, information extraction in high-resolution remote sensing images has been a supplementary way to observe traffic rate of flow. Remote sensing technology has been used in traffic monitoring in recent years for a new direction.It is composed of several aspects as follows:(1) The development of vehicle information extraction in remote sensing images method of our country and abroad was investigated;(2) Study on the similarities and differences between the traditional methods of image analysis and object-oriented image analysis methods. Combined the image object characteristics of vehicles in the high-resolution remote sensing image with object-oriented analysis, the main vehicle object characteristics are determined for further use;(3) The object-oriented analysis method was used for vehicle information extraction test. Test on the ENVI FX software, and obtain a better vehicle extraction effect;(4) Using Qt and C++mixed programming, combined with ENVI secondary development technology, the vehicle information automatically extract system software development is completed. The software can finish the vehicle extraction in a single image, and do batch under a folder with similar images. It is easy to do secondary development using object-oriented software development approach;(5) Study on the development of vehicle information extraction technology in remote sensing images. With the development of remote sensing technology, terminal technology, cloud computing technology, three-dimensional GIS and Beidou navigation technology, a satellite, aviation and three in one city ground transportation monitoring network will be built in the near future. This can provide great help for monitoring of traffic flow, logistics and vehicle management, road emergency, public security, fire and medical pavement protection. Automatic extraction of vehicle information in the high-resolution remote sensing images can be used as a support technology to bring enormous social and economic benefits for building "intelligent transportation" and "smart city"...