3DPD: A photogrammetric pipeline for a PUSH frame stereo cameras

doi:10.1016/j.pss.2021.105165

Planetary and Space Science

Volume 198, April 2021, 105165

https://doi.org/10.1016/j.pss.2021.105165 Get rights and content

Highlights

•
3-dimensional reconstruction software for planetary surfaces.
•
High resolution DTMs of targets on the Mars surface provided by CaSSIS Stereo Camera images.
•
Photogrammetric pipeline for push frame instruments from CaSSIS to Bepicolombo stereo camera STC/SIMBIOSYS.
•
Archiving of the Digital Terrain Models obtained by the CaSSIS stereo pairs.

Abstract

An innovative photogrammetric pipeline has been developed by INAF-Padova for the processing of the stereo images from the CaSSIS (Colour and Stereo Imaging System) (Thomas et al., 2014). CaSSIS is the multispectral stereo push frame camera on board ExoMars TGO (Trace Gas Orbiter) which will image 1.5% of the Mars surface in stereo mode with a spatial resolution of 4.6 m/pixel: the highest resolution single pass stereo capability currently operating at Mars. Data acquisition started in April 2018. The camera is able to provide two images of the same target from two different points of view along the same orbit and within one minute. The telescope is mounted on a rotational stage and its boresight is oriented to 10° with respect to nadir direction. After the acquisition of the first set of images looking forward along track, the rotational stage is rotated by 180° and a second set of images (looking backward) is acquired. The stereo pairs can then be processed to provide the 3D topography of specific targets.

The suite of photogrammetry and imaging tools, named 3DPD (3Dimensional reconstruction of Planetary Data) (Simioni et al. 2017), is designed for processing stereo push frame data and producing the three-dimensional data for geomorphological analysis of planetary surfaces.

The workflow involves a MATLAB tool for the preparation of the inputs (the mosaicked images and the projection matrices) to be ingested into the 3DPD matching core software. The pipeline is in continuous development and routinely ingests a large number of images that CaSSIS is presently acquiring and will continue to acquire in the future. CaSSIS 3DPD products are the unique DTMs available nowadays and the stereo products have been considered in some scientific work (as described in Section 6.2). The same pipeline faces the need of a dedicated pipeline for the Mercury Global Mapping with the Spectrometers and Imagers for the Mercury Planetary Orbiter (MPO) BepiColombo Integrated Observatory SYStem (SIMBIO-SYS) (Cremonese et al., 2020).

Graphical abstract

Introduction

The aim of this article is to introduce the 3D reconstruction software for planetary surfaces developed (Simioni et al., 2017) by the research group of the Astronomical Observatory of Padova (OAPD-INAF). The group is involved in different imaging systems on board space missions, such as BepiColombo, ExoMars 2016 and JUICE, that are acquiring, or will acquire, stereo pairs of planetary surfaces.

BepiColombo has been launched on 20 October 2018 towards planet Mercury. SIMBIO-SYS, on board the spacecraft, is equipped with a stereo imaging channel (STC) designed to obtain a global coverage of the planet surface in stereo mode.

The ExoMars Trace Gas Orbiter (EM-TGO) mission, which has been launched on 16 March 2016, is equipped with the stereo camera CaSSIS (Colour and Stereo Imaging System) (Thomas et al., 2014) which has provided until 29th MTP (Medium Term Planning corresponding to 1 month of operation) 1613 stereo-pairs (August 2020) with different acquisition strategies and different filters modes.

The JUICE mission, which will be launched in 2022 with the camera JANUS on board will acquire stereo pairs of the surfaces of the icy satellites of Jupiter, primarily Ganymede.

The INAF-OAPD institute, as part of the CaSSIS team, has the responsibility for generating and archiving the Digital Terrain Models (DTMs). The DTMs archiving group (Cremonese and Re, 2018) implemented a Repository Dynamic Interface for the management of the requests of DTMs reconstruction and the delivery of the stereo products (https://cassis.oapd.inaf.it/archive/).

STC and CaSSIS use the same type of detector and are both based on the push-frame stereo acquisition mode. The photogrammetric procedures applied for the topographic reconstruction of the planetary surfaces in three dimensions are the same and most of the efforts devoted to the design and integration of the processes will be shared.

In the first Section, an introduction describing the instrumental context is proposed. Section 3 describes the developed software, the structure and all the implemented methods. Section 4 is dedicated to the user interface while the details of the base pipeline are illustrated in Section 5.

Section snippets

Pipeline context

Stereo photogrammetry is a digital image processing technique that allows to metrically extract information about the target observed; providing important support for geomorphological analysis and for the quantitative investigation of the features examined by the camera. Combining a variety of techniques, such as camera calibration, image matching and bundle adjustment, photogrammetry is able to derive 3D information from a pair or a set of overlapping images (Wu, 2017).

The result of this

3DPD pipeline

Fig. 1 shows the block diagram of the photogrammetric pipeline.

It consists of several steps:

-
the creation of the mosaiced images (see Section 3.1) starting from the framelets (single shot acquisitions),
-
the computation of the integer disparity map (see Section 3.2.2) (which define the parallax between the images),
-
the disparity refinement (see 3.2.3) at the sub-pixel level and
-
finally, the triangulation phase.

All these steps will be described in detail in the following sections.

GUI (Graphical user interface)

Even though the software has been designed to be as much automatic as possible, a user-friendly interactive processing is possible through the software’s Graphical User Interface (GUI). The interactive approach offers more control over specific and critical steps of the photogrammetric process.

Through the GUI it is possible to choose different steps of the workflow which have to be executed as well as to define related parameters, such as window template dimension, search area dimension, number

Technical description and architecture of the software

The software is developed in C SHARP language and it takes advantage of pre-processing tools developed in MATLAB.

The code is multithreaded and works on the Windows Operating System (Windows 7 SP 1 or later, 64 bit). It is recommended a configuration Intel Core i7 multi-processor with at least 16 GB of RAM.

The input framelets/images can be accepted in EDR (Experiment Data Records) format and/or in other more common image formats (JPG, TIFF, PNG) or even as MATLAB binary file. Camera parameters

Validation and performance

At the base of the validation procedure is the estimation of the performance and of the vertical precision by collecting statistics on the differences between the stereo DTM and a “ground truth” (another topographic data of higher accuracy that can be used as reference) (Da Deppo et al., 2006), in the following tests:

(i)
HERSCHEL_CRATER: based on a pair of synthetic images generated by using the HiRISE Texturized DTM.
(ii)
CASSIS_IMAGES: status of the 3D reconstruction with the CaSSIS images
(iii)
STC_SVS:

Conclusions

A photogrammetric stereo processing workflow has been developed for the 3D reconstruction from CaSSIS images. The pipeline includes a procedure for the image pre-processing management and different approaches for the stereo image matching. Two different tests based on synthetic and real images are presented to validate the pipeline. At the same time, the first stereo images acquired by the CaSSIS instrument have been processed to achieve the best possible accuracy relying on a preliminary

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This activity has been realized under the Italian Space Agency ASI-INAF agreement no. 2017-03-17. Under the contract the authors have made the pipeline available for the general users on proper workstation settled in their laboratories. We are currently considering to distribute a completely open release of the 3DPD at the end of the contract. A first SUM (Software User Manual) can be found as technical note (Simioni et al., 2020).

The authors wish to thank the spacecraft and instrument

References (48)

C.H. Acton
Ancillary data services of NASA’s navigation and ancillary information facility
Planet. Space Sci.
(1996)
H. Bay et al.
Speeded-up robust features (SURF)
Comput. Vis. Image Understand.
(2008)
R. Jaumann
The high-resolution stereo camera (HRSC) experiment on Mars Express: instrument aspects and experiment conduct from interplanetary cruise through the nominal mission
Planet. Space Sci.
(2007)
C. Re
Effects of image compression and illumination on digital terrain models for the stereo camera of the BepiColombo mission
Planet. Space Sci.
(2017)
S. Tulyakov et al.
Geometric calibration of colour and stereo surface imaging system of ESA’s Trace Gas Orbiter
Adv. Space Res.
(2018)
S- Baker et al.
Lucas-kanade 20 years on: a unifying framework
Int. J. Comput. Vis.
(2004)
P. Becerra et al.
Timescales of the climate record in the south polar ice cap of Mars
Geophys. Res. Lett.
(2019)
J.F. Bell
Calibration and performance of the Mars reconnaissance orbiter context camera (CTX)
Int. J. Mars Sci. Explor.
(2013)
K. Briechle et al.
Template matching using fast normalized cross correlation
L.P. Chew
Constrained delaunay triangulations
Algorithmica
(1989)

W.S. Ching

Normalized cross-correlation: the contrast-dependent problem and its solution

J. Electron. Imag.

(1995)

G. Cremonese et al.

The CaSSIS Digital Terrain Model generation and Archiving at OAPD

European Planetary Science Congress

(2018)

G. Cremonese et al.

March). First Mars surface stereo reconstruction with the CaSSIS stereo camera

G. Cremonese

The stereo camera on the BepiColombo ESA/JAXA mission: a novel approach

Adv. Geosci.

(2009)

G. Cremonese et al.

SIMBIO-SYS: scientific cameras and spectrometer for the BepiColombo mission

Space Sci. Rev.

(2020)

V. Da Deppo et al.

June). A novel optical design for planetary surface stereo-imaging: preliminary design of the Stereoscopic Imaging Channel of SIMBIOSYS for the BepiColombo ESA mission

V. Da Deppo et al.

Optical design of the single-detector planetary stereo camera for the BepiColombo European Space Agency mission to Mercury

Appl. Optic.

(2010)

A. Gruen

Adaptive least squares correlation: a powerful image matching technique

South Afr. J. Photogrammetry, Remote Sens. Cartogr.

(1985)

A. Gruen

Development and status of image matching in photogrammetry

Photogramm. Rec.

(2012)

H.U. Keller

OSIRIS–The scientific camera system onboard Rosetta

Space Sci. Rev.

(2007)

B.R. Kent

3D Scientific Visualization with Blender

(2014)

R.L. Kirk

High- resolution topomapping of candidate MER landing sites with Mars Orbiter Camera narrowangle images

J. Geophys. Res.

(2003)

R.L. Kirk et al.

The effect of illumination on stereo dtm quality:simulations in support of europa exploration. , XXIII ISPRS Congress, 12–19 July 2016, Prague, Czech Republic 10.5194/isprs-annals-III-4-103-2016[ES1] [ES1]Corrected a quote at line 671

ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences

(2016)

J.P. Lewis

Fast Template Matching, Vision Interface 95, Canadian Image Processing and Pattern Recognition Society, Quebec City, Canada, May 15-19

(1995)

Cited by (16)

Seasonal low albedo streaks at the edge of the Martian south polar layered deposits
2024, Icarus
Low Albedo Streaks (LAS) are narrow dark streaks observed at the edges of the South Polar Layered Deposits (SPLD) on Mars. We investigate the seasonal morphology, surface topography of LAS-rich areas using monochrome and multiband images from both the High Resolution Imaging Science Experiment (HiRISE) and the Color and Stereo Surface Imaging System (CaSSIS). We complement the image analysis with spectral data from the Compact Reconnassiance Imaging Spectrometer (CRISM). In addition, we analyze the thermal environment of LAS-rich areas with data from a Martian global climatological model. Our analysis suggests that streaks (i) originate from similar scarps and outcrops on the edge of layered structures at the end of local winter during different Martian years, (ii) lengthening in spring forming sinuous branches on terrain with rippled ridges, and (iii) eventually fade leaving distinguishable bright deposits within the streaked area. The surface topography analysis indicates that streaks advance on slopes between 4° and 34°, while the dust flows emanated from mass-wasting scars propagate on the slope steeper than 20°. Our spectral data analysis suggests that pyroxene- or basaltic-bearing materials are deposited on the surface mainly composed of dust by LAS. Thermal analysis indicates that LAS are active when the surface temperature rises from 145.5 $K$ to 162 $K$ , fading when the temperature increases suddenly and significantly. Finally, we evaluate several possible mechanisms to explain the formation of LAS. LAS might be formed by liquids, and such liquids might be brines. However, the presence of salts with lower eutectic temperatures seems to be necessary for the existence of brine on the Martian surface, and should be investigated at the target area in the future.
Change detection and monitoring of active Martian surface phenomena with the Colour and Stereo Surface Imaging System (CaSSIS) onboard the ExoMars Trace Gas Orbiter (TGO)
2023, Icarus
Mars is host to a variety of active surface processes that relate to changes in seasonal ice/frost, slope activity, wind and processes potentially relating to liquid water. Regular monitoring and change detection of these phenomena is crucial to not only provide us insights into present day Martian surface conditions, but also its past geologic and climatic scenarios. Visual comparisons of high-resolution remote sensing images of the surface from the Mars Reconnaissance Orbiter (MRO) have allowed detailed tracking and monitoring of these changes over time. The CaSSIS instrument onboard the ExoMars Trace Gas orbiter also provides a unique pathway to track areally-large surface changes in colour, that permits us to better constrain origin and evolution of various surface features. Since TGO operates in a non-sun-synchronous orbit, previously adopted qualitative methods like visual image comparisons for MRO cannot be adopted for CaSSIS-based change detection. Consequently, this study lists and describes in detail, the techniques that need to be adopted to conduct such change detection campaigns with CaSSIS, for the variety of active processes currently identified on Mars. This work also proposes and describes two additional semi-quantitative techniques for CaSSIS-based change detection, that may be used in concert with existing visual comparison methods to enable reliable change identification and tracking. It is observed that these supplementary methods work well in characterizing a variety of surface changes related to ice/frost and slope processes, and can provide reasonable spectral constraints to better understand their origin. These methods are not effective in providing quantitative constraints for purely physical changes. Instead, for long-period changes like aeolian bedform movement, digital image correlation techniques on ortho-rectified images, are recommended; while for short period changes like real-time dust devil activity, measurements based on CaSSIS stereo pairs of the region can be used to provide quantitative estimates of change.
In-flight radiometric calibration of the ExoMars TGO Colour and Stereo Surface Imaging System
2022, Planetary and Space Science
Citation Excerpt :
An imaging mode is the combination of parameters that defines a particular observation, these include the number of colour channels (up to four), image compression settings, and final swath size. In addition, CaSSIS is capable of acquiring nearly-simultaneous stereo images that can be combined to produce Digital Terrain Models of the surface and anaglyphs (Simioni et al., 2021, Re et al. this issue). Common to all possible imaging modes is the general scheme for colour imaging with CaSSIS which combines a “push-frame” technique with colour filters deposited on a glass window fixed in front of the detector (Fig. 1).
The Colour and Stereo Surface Science Imaging System (CaSSIS) of the ExoMars Trace Gas Orbiter returns on average twenty images per day of the Martian surface, most of them in 3 or 4 colours and some of them in stereo. CaSSIS uses a push-frame approach to acquire colour images, with four bandpass filters deposited directly above the sensor and an imaging cadence synchronized with the ground track velocity to cover the imaged area with tens of small, partially overlapping images. These “framelets” are later map-projected and mosaicked to build the final image. This approach offers both advantages and challenges in terms of radiometric calibration. While the collection of dark and flatfield frames is considerably enhanced by the frequent and fast acquisition of tens of successive images, mosaics assembled from the adjacent framelets highlight the straylight and changes in the bias of the detector. Both issues have been identified on CaSSIS images, with low intensities overall (up to a few %), but sufficient to generate prominent artefacts on the final assembled colour images. We have therefore developed methods to correct these artefacts that are now included into the radiometric calibration pipeline. We detail here the different steps of the calibration procedure and the generation of the products used for calibration, and discuss the efficacy of the corrections. The relative uncertainties on the bias and flatfield frames are low, of the order of 0.2 and 0.1%, respectively. The uncertainty on the absolute radiometric calibration is of 3%, which is quite low for such an instrument. The straylight adds an estimated ∼1% error to the absolute calibration. The residuals after corrections of the straylight and bias offsets are of the order of a few DNs to tens of DNs. As CaSSIS can observe the Martian surface in challenging illumination conditions to provide unique views of the surface at early and late local solar time, residuals from the straylight correction can become noticeable when the absolute signal is very low. As they appear at the level of the noise in very low illumination images, these residuals do not limit the scientific exploitation of the data. For most of the dataset, as the signal in well-exposed images reaches 8000 DNs in the panchromatic filter and thousands of DNs in the colour filters, the residuals are negligible and CaSSIS provides the best colour images available over many areas covered.
CaSSIS-based stereo products for Mars after three years in orbit
2022, Planetary and Space Science
Citation Excerpt :
The DTM spatial resolution is strictly related to the image quality and to the accuracy of the orientation data so the grid spacing depends on the acquisition conditions. For the time being, a nominal stereo pair acquired at 4.6 m/pix provides a DTM with 15 m/pix post sampling as for the case presented in (Simioni et al., 2021). The 3DPD software also permits the generation of DTMs “at every pixel” (on a grid with post spacing equal to the image resolution: 4.6 m/pix).
The Colour and Stereo Surface Imaging System (CaSSIS) on board the ExoMars Trace Gas Orbiter (TGO), operating in push-frame mode, provides multiband images at four different wavelengths thanks to a Filter Strip Assembly with a panchromatic filter and three broadband filters within the visible and near infrared range. The camera acquires stereo pairs fundamental for the initialization of the photogrammetric process to perform three-dimensional reconstruction of the Martian surface at the best resolution of 4.6 m per pixel for regions up to ∼400 km² in one imaging sequence.
The 3D points derived from the stereo processing are used to generate Digital Terrain Models (DTMs) with height accuracy on the order of one image pixel on ground allowing high-resolution morphometric studies and in general improving the understanding of the geology and geomorphology of the surface of Mars.
This work provides a review of the CaSSIS stereo products supported by a description of the applied methods and examines some specific approaches directed to science analysis. Furthermore, our development of methods is herein focused on the proof of concept and the performance of our dedicated pipeline. The DTM generation procedure has been implemented in a stereo photogrammetric pipeline by the team of the National Institute for Astrophysics-Astronomical Observatory of Padova (INAF-OAPd). The workflow is based on area-based image matching integrated in a multi-resolution approach where the quality of the image matching largely determines the quality of the output DTM. For this reason, the influence of the parameters involved in the matching process (i.e. number of tie-points, template sizes and shape models in matching) has been studied.
CaSSIS stereo products have been generated for approximately 0.1% of the surface of Mars and 16.3% of the total stereo images acquired so far. In this work, some scientifically interesting targets have been considered in the investigation to provide an overview of the quality of the stereo results.
The experimental studies related to stereo analysis frequently led to comparison tests since they represent the best approach for contributing to the methodological consolidation of the photogrammetric data processing. The quality assessment based on comparison with reference terrain data is very promising also in considering areas with different surface type and morphologies.
Pre-landslide topographic reconstruction in Baetis Chaos, mars using a CaSSIS Digital Elevation Model
2022, Planetary and Space Science
Planview detailed morphological analysis of martian landslides is usually performed using orbital imagery such as from the ConTeXt camera (CTX) at 6 m/pix, the Colour and Stereo Surface Imaging System (CaSSIS) at 4.5 m/pix or the High-Resolution Imaging Science Experiment (HiRISE) at 0.25–0.5 m/pix. However, topographic information is key to fully understand a landslide's formation mechanism and its mobility, by estimating the material volumes mobilised and the spatial distribution of erosion and deposition. Digital Elevation Models (DEM) are required to carry out these analyses; nevertheless, there is currently a gap in landslide-volume studies between those using Mars Orbiter Laser Altimeter (MOLA) dataset at ∼450 m/pix or HRSC at 50–200 m/pix and those using HiRISE data at 1–2 m/pix, which is only partially filled by CTX elevation data at ∼20 m/pix. The CaSSIS camera on board the ESA/Trace Gas Obiter (TGO) can be used to produce DEMs, but so far, such data have not yet been used to conduct a landslide volume analysis. Here, we use three reconstruction methods (semi-automatic, morphology-based and tilted) on a CaSSIS DEM to estimate the initial topography and hence the volume and the distribution of erosion and deposition of a 6 km long landslide in Baetis Chaos. Despite the complex topography of the surrounding area due to the presence of an ejecta deposit beneath the landslide, we were able to estimate the landslide's volume and mass distribution. Using a tilted plane as part of estimating the initial topography produced the best results. We evaluated the success by considering the quantifiable balance between erosion and deposition (given the uncertainties) and more subjectively by considering whether the volume distribution matched with what was expected based on the morphology in images alone. Therefore, we recommend the use of this method for individual landslide studies in complex topography where detailed knowledge of the deposit-thickness distribution is required. The semi-automatic reconstruction method produces satisfactory volume estimates and would be better suited to studies where hundreds of landslides are present. We found that CaSSIS data can be used to successfully conduct such analyses, providing additional DEM coverage to study martian medium-scale landslides or other landforms of similar scale (5–15 km) with the notable benefit that it provides single-pass stereo image acquisition.
Geology, in-situ resource-identification and engineering analysis of the Vernal crater area (Arabia Terra): A suitable mars human landing site candidate
2022, Planetary and Space Science
Citation Excerpt :
On a specific area located inside the Vernal crater (Fig. 5D), we planned and acquired a CaSSIS stereo pair characterised by a spatial scale of 4.0 m. In order to prepare the CaSSIS DTM, following the procedure described in Simioni et al. (2021), the undistorted panchromatic images have been first mosaicked after a bundle adjustment process for both intrinsic and extrinsic parameters with an on-ground grid size equal to the image pixel resolution. The matching workflow then focused on a feature-based extraction of the seed points.
A multidisciplinary study of an ancient area of Mars (Early to Late Noachian) located in Arabia Terra is presented, centred at 6°1′N, 354°54′ E and including the 55 km size Vernal crater. By means of different spatial scale imagery datasets and digital terrain models (MOLA, THEMIS, HRSC, CTX, CaSSIS and HiRISE), we prepare a high-resolution geological map of the study site. We highlight the different bedrock stratigraphy inside the Vernal crater which is of particular exobiological interest given the presence of putative ancient hot springs, as well as identifying multiple transverse aeolian ridges, inverted fracture networks and paleochannels, mounds, and a 58 m fresh crater located just outside Vernal crater rim. Within all low-latitude regions of Mars, the studied site presents the highest values (up to 16.0 wt%) of water equivalent hydrogen, hence suggesting that there is a widespread presence of in situ subsurface (at maximum depths of 1–2 m) natural resources, such as water ice and/or hydrated minerals. The equatorial location of the area results in the maximum surface temperature and the highest mean solar flux gatherable on the surface of the planet throughout the year. The interesting scientific case, coupled with the presence of in situ exploitable resources and the thorough accomplishment of all landing/roving engineering safety requirements, make the Vernal crater area a strong landing site candidate for future human exploration of Mars.

View all citing articles on Scopus

View full text

3DPD: A photogrammetric pipeline for a PUSH frame stereo cameras

Highlights

Abstract

Graphical abstract

Introduction

Section snippets

Pipeline context

3DPD pipeline

GUI (Graphical user interface)

Technical description and architecture of the software

Validation and performance

Conclusions

Declaration of competing interest

Acknowledgements

Planet. Space Sci.

Comput. Vis. Image Understand.

Planet. Space Sci.

Planet. Space Sci.

Adv. Space Res.

Lucas-kanade 20 years on: a unifying framework

Int. J. Comput. Vis.

Timescales of the climate record in the south polar ice cap of Mars

Geophys. Res. Lett.

Calibration and performance of the Mars reconnaissance orbiter context camera (CTX)

Int. J. Mars Sci. Explor.

Template matching using fast normalized cross correlation

Constrained delaunay triangulations

Algorithmica

Normalized cross-correlation: the contrast-dependent problem and its solution

J. Electron. Imag.

The CaSSIS Digital Terrain Model generation and Archiving at OAPD

European Planetary Science Congress

March). First Mars surface stereo reconstruction with the CaSSIS stereo camera

The stereo camera on the BepiColombo ESA/JAXA mission: a novel approach

Adv. Geosci.

SIMBIO-SYS: scientific cameras and spectrometer for the BepiColombo mission

Space Sci. Rev.

June). A novel optical design for planetary surface stereo-imaging: preliminary design of the Stereoscopic Imaging Channel of SIMBIOSYS for the BepiColombo ESA mission

Optical design of the single-detector planetary stereo camera for the BepiColombo European Space Agency mission to Mercury

Appl. Optic.

Adaptive least squares correlation: a powerful image matching technique

South Afr. J. Photogrammetry, Remote Sens. Cartogr.

Development and status of image matching in photogrammetry

Photogramm. Rec.

OSIRIS–The scientific camera system onboard Rosetta

Space Sci. Rev.

3D Scientific Visualization with Blender

High- resolution topomapping of candidate MER landing sites with Mars Orbiter Camera narrowangle images

J. Geophys. Res.

The effect of illumination on stereo dtm quality:simulations in support of europa exploration. , XXIII ISPRS Congress, 12–19 July 2016, Prague, Czech Republic 10.5194/isprs-annals-III-4-103-2016[ES1] [ES1]Corrected a quote at line 671

ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences

Fast Template Matching, Vision Interface 95, Canadian Image Processing and Pattern Recognition Society, Quebec City, Canada, May 15-19