Accuracy Assessment of a UAV Block by Different Software Packages, Processing Schemes and Validation Strategies

Vittorio Casella,Filiberto Chiabrando,Ambrogio Maria Manzino,Marica Franzini

doi:10.3390/ijgi9030164

Vittorio Casella, Filiberto Chiabrando + Show 2 more

Open Access

https://doi.org/10.3390/ijgi9030164

Copy DOI

Abstract

Unmanned aerial vehicle (UAV) systems are heavily adopted nowadays to collect high-resolution imagery with the purpose of documenting and mapping environment and cultural heritage. Such data are currently processed by programs based on the Structure from Motion (SfM) concept, coming from the computer vision community, rather than from classical photogrammetry. It is interesting to check whether some widely accepted rules coming from old-fashioned photogrammetry still holds: the relation between accuracy and ground sampling distance (GSD), the ratio between the vertical and horizontal accuracy, accuracy estimated on ground control points (GCPs) vs. that estimated with check points (CPs) also in relation to their ratio and distribution. To face the envisaged aspects, the paper adopts a comparative approach, as several programs are used and numerous configurations considered. The paper illustrates the dataset adopted, the carefully tuned processing strategies and bundle block adjustment (BBA) results in terms of accuracy for both GCPs and CPs. Finally, a leave-one-out (LOO) cross-validation strategy is proposed to assess the accuracy for one of the proposed configurations. Some of the reported results were previously presented in the 5th GISTAM Conference.

Highlights

The use of unmanned aerial vehicles (UAVs) for surveying purposes such as mapping, 3D modeling, point cloud extraction and orthophoto generation has become a standard operation in recent years
It is interesting to check whether some widely accepted rules coming from old-fashioned photogrammetry still holds: the relation between accuracy and ground sampling distance (GSD), the ratio between the vertical and horizontal accuracy, accuracy estimated on ground control points (GCPs) vs. that estimated with check points (CPs) in relation to their ratio and distribution
We show the name of the software used, the mean, standard deviation and root mean square error (RMSE) values of the differences between the photogrammetrically obtained object coordinates of markers and those determined by surveying; the analysis was performed for all the X, Y and Z components of the GCPs and CPs, if any

Summary

Introduction

The use of unmanned aerial vehicles (UAVs) for surveying purposes such as mapping, 3D modeling, point cloud extraction and orthophoto generation has become a standard operation in recent years. The large availability of so-obtained 3D models force the scientific community to explore and develop new tools for visualizing and sharing the achieved products and, as a consequence, open new scenarios strictly related to the use of virtual and augmented reality that for sure need to be considered as an important research topic even more related to the geospatial information [1,2]. Within this framework, the high quality of commercial off-the-shelf cameras, implemented in UAV platforms and the development of new generation software packages created an important revolution in the geomatics field. We have assisted to a transformation in the photogrammetric community that has adopted this new technique, demonstrating its potentiality in several contributions such as [13,16,17,18,19]

Objectives

Methods

Findings

Discussion

Conclusion