Research On Urban Road Traffic Information Extraction Of The Shaded Area From High-Resolution Satellite Images

High-resolution satellite remote sensing brings new ideas to urban traffic information collection. With the improvement of the spatial resolution of remote sensing image, the image of vehicles on the road has become more and more clear. It can provide more traffic information, which is a good source of traffic flow data that is required for transport planning and management. But in high-density urban environment, building shadows blocked a lot of important information, which result in inaccurate results or even errors of image interpretation. These negative effects will reduce the acquisition accuracy of vehicle information, or even block the process of high-resolution satellite remote sensing in the application of urban transport information collection. This paper aims to study the technology of the vehicle information extraction of the shaded area from high-resolution remote sensing images. The main research contents are as follows:(1)Combining of shadow image characteristics of remote sensing and the needs of this article, this paper summarizes methods and procedures of image shadow detection based on brightness and shadow elimination of linear mapping grayscale; and tests the sample data of shadows area of the high spatial resolution remote sensing image to verify the effectiveness of the method; and provides data support for the later extraction of the vehicles.(2)Against to the incomplete vehicle information of shadow area, this paper proposed the method that combined classification of vehicle morphological rule and HSV color space analysis.and the idea of the method and implementation steps are constructed. First, it used the method based on vehicle classification to extract the image data of comprehensive vehicle and then used method of HSV color space analysis to extract the remaining portion of the vehicle be missed. The combination of these two methods to extract the vehicles of shadow region can get better results.(3)According to vehicle information extraction methods and steps which has been built in the chapter, this paper selects the Worldview-2 and Geoeye-1 images after shadow detection and elimination as the experimental data to extract vehicles and verify the effectiveness of the method, then uses the two indicators of vehicle extraction correct rate and vehicle extraction rate to evaluate the experimental results.This study complements the vehicle acquisition methods based on high-resolution satellite remote sensing data and improve the extraction accuracy, which is important for high-resolution satellite remote sensing image in the application of urban transport information collection.