Disparity Image Generation from UltraCamD Images With " Semi-Global Matching" Algorithm

Considering ongoing developments in Photogrammetry and Remote Sensing and attending to their applications such as digital true orthophoto generation from multi view and high resolution images, have already proven the urgency and necessity of approaches for generating DSM as input of these applications. In this research we describe the “Semi-Global Matching” algorithm and then implement and test if for generating an initial disparity image from airborne UltraCameD images. SGM uses a pixelwise, Mutual Information based matching cost compensating radiometric differences of input images. Pixelwise matching is supported by a smoothness constraint that is usually expressed as a global cost function. SGM performs a fast approximation by pathwise optimizations from all directions. Furthermore, post-processing steps for removing outliers, recovering from specific problems of structured environments and the interpolation of gaps are needed, to generate an accurate and precise DSM in the next steps for other applications.