IMAGE CORRECTION FOR DIGITAL MAPPING
Yubin Xin · 2006
Digital base maps generated from various source imagery, such as satellite imagery, airborne data, and aerial photograph, are becoming very useful and efficient. Digital map can provide detailed topographic, land-use and land-cover information on earth. With the growing requirements of automatically correcting raw data and generating digital maps, the technique which can be used for that purpose is demanded. This paper introduces a system containing a serial set of methods to correct satellite image or aerial data for mapping. The technique makes full use of georeference and sensor model information, such as ephemeris data, geometric model, and/or GPS/INS positioning information, to automatically register and orthorectify the raw image data. Through mosaicking process, a seamless mosaicking image or image tiles is produced. Semi-automatic and manual editing can be performed to produce a standard map to satisfy a Mapping branch requirement. This paper will briefly describe the mapping system concept and application that forms a production mapping flow. It will address on the Image Correction methodology including automatic image to image registration with initial direct georeference and DEM; automatic GCP error detection and sensor model refinement; orthorectification; and seamless cutline selection and mosaicking. The introduced image correction system is an automated system. It consists of multiple processing component functions in a scripting environment. The scripting work flow is highly customizable to deal with large data set. It is also capable of data accessing, retrieval and delivery to Enterprise database, i.e. Oracle 10G, or web via complementary subsystems.