Study on quality assessment of multi-source and multi-scale images in disaster prevention and relief

Yongan Xue,Mingmei Zhang,Jinling Zhao,Qinghai Guo,Rui Ma

2012-01-01

Disaster Advances

Abstract:Considering the frequent occurrence of various geological disasters in recent years, such as earthquake, debris flow, landslide, etc., remotely sensed images have been extensively utilized which provided optimal data sources for large-scale disaster monitoring and fast response. Consequently, the quick assessment of image quality is the prerequisite to select best remotely sensed images, but it has not been fully explored. In our study, geometric and color qualities were assessed using multi-source and multi-scale images. A layout scheme of check points was put forward to evaluate the geometric quality of images using a handheld global positioning system (GPS, Trimble Geo RTK). By selecting five national-level control points (Beijing 54 and Xi'an 80 coordinates), the positioning accuracy using the proposed scheme reached 4 cm better than 10 cm nominal accuracy and the number of required field survey as well as the layout scheme were also given, which can satisfy the accuracy requirement of plane position in filed survey. At the same time, an indicator (sigma 2) of brightness equalization was used to evaluate the color quality. The programme of color quality assessment was made using a combination of AcrGIS and C#. By selecting Geoeye and QuickBird images of Haiti earthquake, Spot 5, IKONOS and Unmanned Aerial Vehicle (UAV) images of geological disasters in Datong City and Xiangning County, Shanxi Province, their sigma 2 were evaluated. The final results demonstrated that when sigma 2 was less than and equal to 15, the brightness equalization was good and conversely it was bad when sigma 2 was greater than 100. This study is of significant importance to quickly select the optimal remotely sensed images from various kinds of images for assessing geological disasters when they occur.