Hilmar Ingensand

Calibration and development for increased accuracy of 3D range imaging cameras

Abstract Range imaging has become a valuable technology for many kinds of applications in recent years. Numerous systematic deviations occur during the measurement process carried out by available systems. These systematics are partly excited by external and partly excited by internal influences. In this paper the following investigations will be presented in closer detail. First the statistics of the distance measurement of the analyzed range imaging cameras SwissRangerTM SR-2 and SR-3000 will be shown. Besides the question if the measurements are Gaussian distributed, the precision of the measurements will be shown. This aspect is of importance to answer the question if the mean value of a series of measurements leads to more precise data. Second diverse influencing parameters like the targets reflectivity and external as well as internal temperature are aimed. The dependency of the distance measurements with respect to the amplitude is one of the main aspects in this paper. A specialized set up has been developed in order to derive experimentally the detailed correlation, which is expressed in terms of linearity deviations. Besides the results of some specific aspects, an overview of the recommended calibration procedure is given. The reader of this paper will be enabled to understand the calibration steps needed to gain highly accurate data from the investigated range imaging cameras. Due to the fact that range imaging cameras are on their way to become state of the art in 3D capturing of the environment, it is of importance to develop strategies for the calibration of such sensors in order to enable users to revert to these principles for the sake of simplicity. Therefore, these strategies long for sophisticated approaches and reliable results of investigations. This paper will introduce such an approach to be discussed within the scientific and user environment. One of the main achievements of this work is the introduction of a method to significantly decrease the influence of temperature on the distance measurements by means of a differential measurement principle setup. The verification of the functionality is presented, as well.

Incorporating 2D tree-ring data in 3D laser scans of coarse-root systems

In times of global change biomass calculations and the carbon cycle is gaining in importance. Forests act as carbon sinks and hence, play a crucial role in worlds and forests carbon budgets. Unfortunately, growth models and biomass calculations existing so far mainly concentrate on the above-ground part of trees. For this reason, the aim of the present study is to develop an annually resolved 3D growth model for tree roots, which allows for reliable biomass calculations and can later be combined with above-ground models. A FARO scan arm was used to measure the surface of a tree-root segment. In addition, ring-width measurements were performed manually on sampled cross sections using WinDENDRO. The main goal of this study is to model root growth on an annual scale by combining these data sets. In particular, a laser scan arm was tested as a device for the realistic reproduction of tree-root architecture, although the first evaluation has been performed for a root segment rather than for an entire root system. Deviations in volume calculations differed between 5% and 7% from the actual volume and varied depending on the used modeling technique. The model with the smallest deviations represented the structure of the root segment in a realistic way and distances and diameter of cross sections were acceptable approximations of the real values. However, the volume calculations varied depending on object complexity, modeling technique and order of modeling steps. In addition, it was possible to merge tree-ring borders as coordinates into the surface model and receive age information in connection with the spatial allocation. The scan arm was evaluated as an innovative and applicable device with high potential for root modeling. Nevertheless, there are still many problems connected with the scanning technique which have an influence on the accuracy of the model but are expected to improve with technical progress.

Calibration of the fast range imaging camera SwissRanger for use in the surveillance of the environment

Many security & defense systems need to capture their environment in one, two or even three dimensions. Therefore adequate measurement sensors are required that provide fast, accurate and reliable 3D data. With the upcoming range imaging cameras, like the SwissRangerTM introduced by CSEM Switzerland, new cheap sensors with such ability and high performance are available on the market. Because of the measurement concept these sensors long for a special calibration approach. Due to the implementation of several thousand distance measurement systems as pixels, a standard photogrammetric camera calibration is not sufficient. This paper will present results of investigations on the accuracy of the range imaging camera SwissRanger. A systematic calibration method is presented which takes into consideration the different influencing parameters, like reflectivity, integration time, temperature and distance itself. The analyzed parameters with respect to their impact on the distance measuring pixels and their output data were determined. The investigations were mainly done on the high precision calibration track line in the calibration laboratory at ETH Zurich, which provides a relative accuracy of several microns. In this paper it will be shown, under which circumstances the goal accuracy of the sub centimeter level can be reached. The results of this work can be very helpful for users of range imaging systems to increase their accuracy and thus the reliability of their systems. As an example, the usefulness of a range imaging camera in security systems for room surveillance is presented.

A tool to model 3D coarse-root development with annual resolution

Dynamic root-development models are indispensable for biomechanical and biomass allocation studies, and also play an important role in understanding slope stability. There are few root-development models in the literature, and there is a specific lack of dynamic models. Therefore, the aim of this study is to develop a 3D growth-development model for coarse roots, which is species independent, as long as annual rings are formed. In order to implement this model, the objectives are (I) to interpolate annual growth layers, and (II) to evaluate the interpolations and annual volume computations. The model developed is a combination of 3D laser scans and 2D tree-ring data. A FARO laser ScanArm is used to acquire the coarse-root structure. A MATLAB program then integrates the ring-width measurements into the 3D model. A weighted interpolation algorithm is used to compute cross sections at any point within the model to obtain growth layers. The algorithm considers both the root structure and the ring-width data. The model reconstructed ring profiles with a mean absolute error for mean ring chronologies of <9% and for single radii of <20%. The interpolation accuracy was dependent on the number of input sections and root curvature. Total volume computations deviated by 3.5–6.6% from the reference model. A new robust root-modelling tool was developed which allows for annual volume computations and sophisticated root-development analyses.

Hydrostatic levelling systems: Measuring at the system limits

Abstract Three hydrostatic displacement monitoring system applications in Switzerland are discussed; the first concerns experience gained monitoring the foundation of the Albigna dam, the second relating to the underground stability of the Swiss Light Source synchrotron and the third concerning the deformation of a bridge near the city of Lucerne. Two different principles were applied, the Hydrostatic Levelling System (HLS) using the “half-filled pipe principle” developed by the Paul Scherrer Institute and the Large Area Settlement System (LAS) using the “differential pressure principle”. With both systems ground deformations induced by tidal forces can be seen. However, high accuracy of single sensors is not sufficient. A well-designed configuration of the complete system is equally important. On the other hand there are also limits imposed by installation logistics and by the environmental conditions. An example is the bridge monitoring application, where the acceleration along the bridge due to the passage of heavy trucks limits the feasibility of using hydrostatic levelling measurements.