Quality assessment of digital surface models extracted from WorldView-2 and WorldView-3 stereo pairs over different land covers

Abstract

Digital surface models (DSMs) extracted from very high resolution (VHR) satellite stereo images are becoming more and more important in a wide range of geoscience applications. The number of software packages available for generating DSMs has been increasing rapidly. The main goal of this work is to explore the capabilities of VHR satellite stereo pairs for DSMs generation over different land-cover objects such as agricultural plastic greenhouses, bare soil and urban areas by using two software packages: (i) OrthoEngine (PCI), based on a hierarchical subpixel mean normalized cross correlation matching method, and (ii) RPC Stereo Processor (RSP), with a modified hierarchical semi-global matching method. Two VHR satellite stereo pairs from WorldView-2 (WV2) and WorldView-3 (WV3) were used to extract the DSMs. A quality assessment on these DSMs on both vertical accuracy and completeness was carried out by considering the following factors: (i) type of sensor (i.e., WV2 or WV3), (ii) software package (i.e., PCI or RSP) and (iii) type of land-cover objects (plastic greenhouses, bare soil and urban areas). A highly accurate light detection and ranging (LiDAR) derived DSM was used as the ground truth for validation. By comparing both software packages, we concluded that regarding DSM completeness, RSP produced significantly (p<0.05) better scores than PCI for all the sensors and type of land-cover objects. The percentage improvement in completeness by using RSP instead of PCI was approximately 2%, 18% and 26% for bare soil, greenhouses and urban areas respectively. Concerning the vertical accuracy in root mean square error (RMSE), the only factor clearly significant (p<0.05) was the land cover. Overall, WV3 DSM showed slightly better (not significant) vertical accuracy values than WV2. Finally, both software packages achieved similar vertical accuracy for the different land-cover objects and tested sensors.

Evaluation of the Consistency of Simultaneously Acquired Sentinel-2 and Landsat 8 Imagery on Plastic Covered Greenhouses

“…The sun glint effect has been extensively discussed in aquatic satellite scenes [53][54][55], being previously reported for PCG roofs in a study about the quality of digital surface models extracted from very high resolution satellite stereo pairs (WorldView-2 and WorldView-3) [56]. In this case, the stereo pairs viewing geometry and its relationship with the sun position could induce specular reflection of sun light, thus causing unusually bright pixel digital values (sun glint effect) and thus contributing to increase the number of missing image matching points.…”

Architectural Graphics Research: Topics and Trends through Cluster and Map Network Analyses

“…Another representation possibility consists of the comparison of procedures to map urban areas using mobile and fixed systems [86]. In this context, photogrammetric techniques have been applied to obtain measurements related to the geomorphology of the terrain or the geometry of facilities in urban areas [87][88][89], as well as for the detection of volumetric or shapes changes due to erosion [90,91]. Figure 6.…”

Geometric Quality Evaluation Method for Multi-View Satellite Images in 3d Reconstruction

“…Researchers have made great efforts to reconstruct 3D models from satellite images. The quality of the 3D models reconstructed from satellite images is affected by various factors (Aguilar et al, 2019;Ozcanli et al, 2015), including type of cover objects (Poli and Caravaggi, 2013), sensors for satellite imaging Poli et al, 2015), 3D reconstruction software and algorithms (Facciolo et al, 2017;Qin, 2017;Shean et al, 2016), and so on. For the 3D reconstruction of a specified area, as the type of cover objects is fixed, the reconstruction quality mainly depends on the image acquisition (sensors for satellite imaging) and image processing (3D reconstruction software and algorithms).…”