Next Article in Journal
Integrated Reconstruction of Late Quaternary Geomorphology and Sediment Dynamics of Prokljan Lake and Krka River Estuary, Croatia
Previous Article in Journal
Imaging Cultural Heritage at Different Scales: Part I, the Micro-Scale (Manufacts)
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 15
Issue 10
10.3390/rs15102587
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

German Extermination Camps on WWII Reconnaissance Photographs. Orthorectification Process for Archival Aerial Images of Cultural Heritage Sites

by
Sebastian Różycki
*,
Artur Karol Karwel
and
Zdzisław Kurczyński
Faculty of Geodesy and Cartography, Warsaw University of Technology, 1 Sq. Politechniki, 00-661 Warsaw, Poland
*
Author to whom correspondence should be addressed.
Remote Sens. 2023, 15(10), 2587; https://doi.org/10.3390/rs15102587
Submission received: 28 February 2023 / Revised: 29 April 2023 / Accepted: 6 May 2023 / Published: 16 May 2023
(This article belongs to the Section Environmental Remote Sensing)
Browse Figures
Versions Notes

Abstract

:
Aerial photographs taken over the past 80 years are often the only record of topography and events that have been destroyed or obliterated. However, the lack of camera certificates for many historical photographs, and their physical degradation, often makes it challenging to correct them geometrically. In this paper, we present the process of orthorectifying archival Luftwaffe aerial photographs of the area of the Treblinka extermination camp from May 1944, based on a computer vision-based process and preprocessing techniques. Low-cost and easily accessible software was used, which allowed for the generation of a fully metric orthophotomap in a repeatable and accurate way. This process can be repeated for archival aerial photographs from other dates (for the Treblinka camp) and other extermination camps (Belzec and Sobibor).

1. Introduction

German Extermination Camps from WWII

A United States Holocaust Memorial Museum (USHMM) study indicates that there were about 42,500 Nazi German forcible detention camps (1933–1945) of different categories [1], through which 18 million prisoners from thirty different countries of the world passed. All of these camps were tasked (to varying extents and intensities) with the use of terror, the exploitation of the so-called “labour force”, and the destruction of the conquered peoples. A specific role in the genocide was played by a group of extermination camps: Belzec, Sobibor, and Treblinka II, which were established after the Wannsee conference in 1942. The three camps were part of the “Operation Reinhardt” framework and were also called “Operation Reinhardt camps” [2,3]. Extermination camps were designed specifically for the systematic killing of people, Jews mostly, who were delivered en masse on trains. However, people of other nationalities, including Poles, Soviet POWs, and Roma, were also killed. Gassing was the primary method of killing in the camps [4] and was conducted at permanent installations constructed for this specific purpose.
The primary problem in the archaeological investigations, monitoring, or restoration of the German extermination camps, i.e., Treblinka, Sobibor, and Belzec (Figure 1), is the Halakhic ban on any violation of human remains. This rule defines a specific methodology that should be used to investigate these extermination camps. The first step is locating and delineating gravesites as precisely as possible. Because of Halacha (Jewish Law), we can only use non-invasive techniques that allow for the discovery of places without serious interference in the ground [5,6,7].
The next problem is the camp topography. The construction of the camps themselves poses a significant challenge to the research being performed in their areas. They were established to mass exterminate groups of people and were not expected to function for long. Most buildings were prefabricated wooden barracks (e.g., Belzec, Treblinka). The only stone/brick buildings were the gas chambers [8]. The Germans were meticulous in removing all traces and evidence of the camps. Barracks, huts, and entire gas chambers were precisely disassembled and demolished [9] to transform the former camp area into “innocent-looking” land.
A good example is the fate of the Treblinka camp, where the Germans destroyed the camp infrastructure and set up a farm in its place [10]. Due to the lack of camp plans or other documents, research to date has mainly been based on the testimonies of witnesses. Many of these were collected several years after the war, which makes their interpretation difficult [11]. Most describe the killing process rather than focus on the spatial relationships of individual parts of the camp. The plans and sketches of the camps prepared by the witnesses were drawn from memory, are highly subjective, differ significantly from each other, and were not drawn to scale, which makes it hard to overlay them on spatial data (such as aerial photographs).
The photographic materials obtained during World War II (WWII) are unique resources that provide a record of the area’s topography and the changes in them in the middle and end of the 20th century [12,13]. The photographs were taken with the aim of carrying out air reconnaissance tasks, i.e., performing an analysis of ongoing military operations. These valuable materials often contain accidentally registered “events”, which can be used for purposes other than the analysis of hostilities. Aerial photographs are essential materials that allow us to observe and analyse changes in the surrounding space over several decades [14,15,16,17,18]. Cultural heritage sites are unique areas which should be protected, and the priority areas that require monitoring, restoration, and management should include the places related to the remembrance of the victims of WWII. The current research pays attention to the forgotten [19,20,21]: the sites of prisoners’ executions, namely concentration and extermination camps.

2. Aerial Reconnaissance Photos of Extermination Camps

2.1. WW2 Aerial Reconnaissance Archives

The National Archives and Records Administration of the United States (NARA) and the National Collection of Aerial Photography in Scotland (NCAP) hold about 1.2 million German aerial photographs from WWII. These photographs cover Europe, significant areas of the Soviet Union up to the Caspian Sea, and North Africa (e.g., Tunisia and Egypt) [22]. So far, no accurate and reliable catalogue has established the share of German photographs held by the Scottish and American archives. Some series of photographs are in both archives, while others are available only in one of them. The physical condition of the stored images in the archives is also different. In NARA, the photographs are most often made available in paper prints; in Scotland, they come in the form of microfilms or original films [23,24]. This situation makes it difficult to use the collections and requires knowledge and considerable experience in using these valuable materials. Another essential aspect of archival collections is the characteristics of the photographs taken for air reconnaissance.
German air reconnaissance was mainly tactical and divided between individual air fleets (German: Luftflotte) [25]. In the case of Eastern Europe, the photographs taken by the Luftwaffe were obtained mainly during the hostilities. For many situations (e.g., from the suspension of land warfare to a fixed front line), a series of photographs taken every few days are available. Reconnaissance flights were aimed at obtaining information from the front and the support area. Therefore, flights along the major railway lines or roads prevail. The photographic overlap in such cases does not exceed 30–50%. It decreases to a few percent once the direction of the flight changes. When the flight is aligned, it returns to the nominal value. The second type of flight obtained pictures as blocks of photographs.
In most cases, both longitudinal and lateral overlaps are retained. Similar photographs (in blocks) are available for larger cities in Poland. The photographs were possibly taken to create photoplans (German: Bildplan), which can be assigned to earlier years of the war as most of these flights were operated from 1940 to 1943.
In the case of the allied forces (Joint Air Reconnaissance Intelligence Centre, Medmenham, UK), the reconnaissance unit was highly centralised and focused on various reconnaissance products. The raids were conducted pointwise and aimed to cover photographs of objects with a block, which, in subsequent stages, were thoroughly interpreted. The allied forces’ photographs mainly contain industrial facilities (such as factories and refineries) and military facilities (e.g., seaports and airports) [25,26,27].

2.2. The Archival Aerial Photographs of German Extermination Camps from WWII in Poland

For the reasons described in Section 1 (Introduction), the most difficult areas to carry out research on are those of former extermination camps. For these objects, no building documentation has been kept (as opposed to the Auschwitz camp, for example [28]), and no camp buildings and barracks have been preserved (as opposed to the camps in Majdanek or Stutthof). During the research on Treblinka, Belzec, or Sobibor, aerial photographs were considered to be significant supporting material.
In the case of extermination camps, German aerial photographs were used as a data source during the multi-season archaeological research at the camp in Sobibor [8,29,30,31]. Aerial photographs taken in 1940 and 1944 were analysed and interpreted. In the rich literature on the subject, there is no information on the use of corrected archival aerial photographs for this camp. The selected photographs were only subjected to interpretation and were not used as measurement data.
The corrected archival aerial photographs obtained by the Luftwaffe were used by Staffordshire University archaeologists to conduct research from 2007 to 2015, covering the areas of both the Treblinka I work camp and the Treblinka II extermination camp [32,33]. The correction process was performed using the ArcMap software based on polynomial correction (rectification process), which did not include the denivelation of the terrain. The first comprehensive search query of photogrammetrical materials for the camps in Treblinka was carried out by the Warsaw University of Technology team in 2017. Both archives described in Section 2.1 were queried. A set of available photographs (from 1940 and 1944) was obtained and used during archaeological and geophysical research from 2017 to 2022 [34].
For the camp in Belzec, archival aerial photographs were not used during the archaeological research carried out from 1997 to 1999. Archaeological research on the camp preceded the construction of a monument commemorating the victims [35]. After a search of the literature, there were no results for available aerial photographs of the extermination camp in Belzec.
Within the framework of this article, a search query was conducted for the camps in Sobibor and Belzec. The results comprise the available photographs (reconnaissance units of the Luftwaffe) and are presented in Table 1. The scale was obtained from archival queries and is an approximate piece of information. It allows for a quick assessment of the usefulness of this material for future interpretation processes. The table also contains the photographs available for the Treblinka camp [34].
When analysing the results obtained, it is worth paying attention to the dates of the flights available. Aerial photographs taken during the operation of the camps have not been preserved. In April 1942, the extermination camp in Sobibor began its operation and it ceased its activity in December 1943. The second camp in Belzec started its activity in March 1942 and closed in June 1943. The third and largest extermination camp was established in Treblinka, probably in June 1942, and was closed in November 1943. The camps in Belzec and Sobibor have pictures that were taken before their construction. Such a set of photographs will allow for a time analysis that considers the changes in the terrain coverage which occurred between 1940 and 1944. For Treblinka, the quality of the photographs from 1940 does not allow for their interpretation [34]. Photographs from flights dating back to mid-1944 are available for all extermination camps. The approximate scale of the photographs and the good weather conditions prevailing at the time will allow for the future analysis of the area after the already-closed camps.

3. Geometric Problems in Archival Photo Processing

It is difficult to overestimate the importance of air reconnaissance photographs from WWII and remains of that period for historical research. In the context of the study of extermination camps, such photographs are part of the few objective sources of information about these camps, especially resulting from conscious destruction and the covering up of traces of such camps at the end of the war. Reconnaissance photographs have undeniable interpretative and measuring potential.
Historians and archaeologists know the possibilities of interpretation and measurement, but they often lack knowledge of professional use based on modern possibilities, including digital photogrammetry. The problem can be the access to source reconnaissance photographs of an exciting area and objects. From the technical point of view, the use is usually limited to the flat-fitting of photographs in existing map studies. However, the potential opportunities are much more significant [36].
Cooperation with specialists interested in such photographs indicates that the most desirable product, which would be useful for further analyses, is a digital orthophotomap developed on their basis. Such an orthophotomap retains all the interpretative qualities of original pictures and is additionally a cartometric product (it has a georeference, and each pixel has terrain coordinates in the adopted frame of reference). Such a map in digital form gives unlimited possibilities of spatial analyses with the use of other geodata, e.g., through the superpositioning of orthophotomaps with other geodata (other maps, photos, results of interpretation, archaeological research, results of geophysical research), up to current remote sensing data (such as multispectral images, altitude models, etc.). Such analyses can be conducted using the currently available software in a GIS environment.
The development of orthophotomaps from aerial photographs in the classical approach requires [37]:
  • Digital photographs (e.g., scanned photographs on film or scanned paper contact prints or images acquired directly with digital cameras);
  • Knowledge of the elements of the internal orientation of the camera (EIO);
  • Knowledge of the elements of the external orientation of the camera (EEO) or having the coordinates of the group of points identified in the photographs and measured in the field (photo control points);
  • A digital terrain model (DTM) of the image-covered area.
A review of the reconnaissance photographs from WWII, including the photographs of German extermination camps (see Table 1), indicates that the state of available photographs is very diverse, and their map (photogrammetric) elaboration usually differs from typical scenarios (workflow) in the development of modern aerial photographs.
Reconnaissance images do not usually constitute a block of images in today’s classic sense (images with sufficient longitudinal coverage in series and coverage between strips). Instead, we often deal with single photos. Reconnaissance photographs during WWII were made with reconnaissance cameras (not measuring cameras). The photographs were intended to serve (and served) the needs of military reconnaissance (photointerpretation), not the need to create cartometric photogrammetrical products based on them (such as altitude models, orthophotomaps or vector maps). This means that the camera design and shooting capacity addressed the needs of photointerpretation (not precise measurement). The consequence of this priority was the decision to use cameras with smaller frame sizes, often with longer focal lengths, equipped with focal plane shutters, thus causing additional geometric distortion in the images. The priority was to reduce the camera’s weight at the expense of the lower stability of the structure itself.
Usually, no camera certificates are available, i.e., the elements of the internal orientation of the camera and photos are not known. It is good to see the (usually four) fiducial marks and the nominal principal distance indicated in the background frame. As a result of the scanning or copying process, these elements are entirely or partially lost (e.g., one can see only three out of four fiducial marks because the photo format is larger than the A4 scanner used). Photographs are rarely available as original negatives or image diapositives (film contact copies). Most often, they are in the form of scans from paper contact prints. Such material is burdened with very significant geometric errors (deformations).
The main source of these deformations is the shrinkage of the photo paper (during the photochemical processing and in the ageing process of prints). These deformations are further enhanced by the scanning of paper contact prints on non-professional scanners (desktop scanners). From the point of view of the measuring potential of such materials, attention should be paid to the geometric nature of the deformation. These are affine-type deformations, which result in different scales along and across the direction of flight (due to various degrees of shrinkage of the photo paper in both directions). Such non-conformal deformations, which are not included in the development process, directly burden the final product of the study with geometric errors.
Pictures from WWII were taken about 80 years ago. In such photographs, it is challenging to identify details which would have survived to the present day, could be measured, and could serve as photo control points in the development process. This is particularly true for less-urbanised areas. It is a little easier to find the details that can be identified in archival and modern photographs or in orthophotomaps. Such details (candidates for natural photo control points) can be, for example, road intersections and breaks, preserved boundaries of cultivated fields, intersections of cultivated fields with roads, and other preserved terrain details.
The historical aerial photographs orthorectification process (HAPO) [38] is present in the literature in quite large numbers. The earliest aerial photographs used in the HAPO [39] were taken in the 1930s using Rb 18 camera, purchased from Zeiss by Portuguese photo companies. Currently, the photos are stored at the Instituto Geográfico do Exército (IGeoE) in Lisbon, Portugal [40].
Photos taken during World War II (from 1943–1944) have been used in several publications. Unfortunately, the process of correcting the photos was not described in them [41]. In some studies, the correction was performed based on 2D transformations in GIS programs [42,43].
Allied photos from 1945 that had been taken with an American K17 camera were used for a multi-temporal analysis of the Trentino region in north-eastern Italy [44]. The availability of large quantities of images taken after the war using US cameras is related to the orthophotomap generation program that was carried out in Western Europe between 1945 and 1946 as part of the Project Casey Jones program [45,46].
In the following publications, various applications of archival photos can be found, although these were acquired after the end of WWII, i.e., from the following years: 1951 [16], 1954 [47,48], 1959 [14,49], 1963 [50], and 1966 [51].
Studies on the geometry of aerial photographs taken by the Luftwaffe between 1939 and 1945 are underexplored in the literature. Undertaking this task is therefore of great importance, particularly considering the availability of these photos in open archives and their number (more than one million) [22].

4. Materials and Methods

4.1. Aerial Photographs Preprocessing

Modern digital photogrammetry methods allow for a different look at the problem in the development of archival photographs, including reconnaissance pictures from WWII. A typical, and the most desirable, product of such a study is a digital orthophotomap. This can result from processing (orthorectifying) only one photo (a relatively common case) or from processing a strip or block of images with mutual coverages. There are several problems involved in developing such photographs (see the previous chapter).
In a typical case, one can see a background frame containing fiducial marks and principal distance. The absence of the camera certificate makes it impossible to introduce the basic elements of the internal orientation of the camera (calibrated image distance and location of the principal point of the photo) and the lens distortion. In practice, however, this is not the most serious problem or source of errors. When scanned paper prints of archival photographs (which is a typical case) are processed, the main source of geometric errors is the shrinkage of paper prints and the process of scanning them on non-professional scanners. These cause significant affine errors (deformations) (see the previous chapter).
An analysis of many Zeiss and Leica (formerly Wild) camera certificates from the 1950s and 1960s conducted by the authors indicates the manufacturer’s high precision in mounting these cameras, especially the focal frame with the lens. This results in:
  • A minimal discrepancy (at the level of micrometres) between the position of the principal point of the photo and the origin of the image coordinate systems, defined by the fiducial marks;
  • The regular and stable position of the fiducial marks that create a square (with a deviation from the square at the level of micrometres).
Furthermore, lens distortions are moderately small (a few to several micrometres), especially for long-focus lenses. This means that the errors in the camera itself are practically negligible and two orders of magnitude smaller than the expected deformations from the paper shrinkage and the scanning process. Therefore, the lack of a camera certificate is not the main and dangerous source of errors.
These observations allow for the generation of a virtual camera certificate with zero lens distortion, zero image coordinates for the principal point, and fiducial marks that create a square. The affine transformation of the measured fiducial marks on the print scans to their nominal position (from the virtual certificate) will consider the significant and irregular shrinkage of the photo paper and the errors resulting from the use of a non-professional desktop scanner, with a negligible slight distortion of the lens. The above procedure can also be applied with only three of the four fiducial marks visible (which happens in practice).
Reconnaissance photographs from WWII are about 80 years old. Identifying details in such photographs that may meet the criteria of ground photo control points (GCP) is a typical problem in developing them. We usually have a full coverage of countries with a digital orthophotomap and a DTM from modern aerial photographs and aerial laser scanning (ALS). Such products are the best source for the identification of details on modern and archival photographs—candidates for the derivation of photo control points in the orthorectification of archival photographs (X and Y coordinates from the orthophotomap, height from the DTM). In addition, a modern DTM can be used in the process of orthorectification itself, assuming that the terrain (topography) has not changed significantly. Such a recommendation remains valid even in the case of archival photographs with mutual coverage, thereby allowing the generation of a DTM based on them.
When producing an orthophotomap, the first and most crucial stage is the orientation of the photographs. In the classical approach, the following stages can be distinguished in the orientation of archival aerial photographs: reconstruction of camera parameters based on the camera certificate (reconstruction of elements of internal orientation—EIO), relative orientation aimed at combining the photographs into one geometrically coherent block that consists in the automatic or semi-automatic measurement of tie points (pass points), external orientation based on natural ground control points identified in archival photographs and terrain or in newer maps.
The data obtained from the archival photos in the form of scanned paper contact prints are subject to large geometrical and radiometric errors. First of all, deformations of the paper base due to its shrinkage and radiometric errors caused by the fading of the scanned material due to its improper storage can be indicated here. Particularly undesirable is differential shrinkage, which causes affine distortions. The incorrect selection of scanning parameters and the low quality of the scanner introduce additional distortions which affect the accuracy of the orientation of archival photographs.
These problems became visible in our processed photographs. Due to the wrong arrangement of the photographs in the scanner, the photo frame was cut off with a partial loss of fiducial marks. There was also no camera certificate. These made the classical development process with the use of a commercial software impossible (e.g., Inpho or Orima).
It was then decided that the computer vision (CV) methods popular in recent years would be used, specifically the SfM (Structure-from-Motion) method, implemented in the easily available Agisoft Metashape software [52]. This method is resistant to local scale changes and image distortion. The so-called self-calibration is possible, i.e., processing without knowing the elements of the internal orientation of the photographs (without the camera certificate) and treating these elements as unknown in the adjustment process. An automatic search for pass points on a pair of photographs allows you to overcome poor radiometric image quality. This makes the Agisoft Metashape environment very useful for developing archival photographs as it is easily accessible and user-friendly [53,54].
The development process using SfM can be divided into two stages: the finding and matching of pass points (steps 1–3) and the iterative determination of relative orientation elements (steps 4–6) (Figure 2).
In generating an orthophotomap of the Treblinka area in the Agisoft Metashape v. 1.8.5 software, three stages can be distinguished: image orientation, digital terrain model (DTM) creation, and orthophotomap generation. The self-calibration option was used in the process of image orientation. In this process, the basic elements of the internal orientation of the photographs were treated as unknowns, along with a description of their affine deformations.

4.2. Case Study Area and Datasets

The former Treblinka II extermination camp is located in the Masovian Voivodship, 100 km northeast of Warsaw. The Germans built the Treblinka II camp in the middle of 1942, next to a village and railway station bearing the same name. The first transport of Jews deported from the Warsaw Ghetto arrived on 23 July 1942 [55]. After that, the camp became the site of the extermination of Jews deported from the Warsaw Ghetto and other ghettos established by the Germans in different towns and cities of the Warsaw District. Jews from Germany, Austria, France, Belgium, Yugoslavia, Greece, and the USSR were all sent to Treblinka II. The authorities used rail transport to bring the victims to their final destination, a railway platform directly neighbouring the camp [56]. The entire extermination process for each group of transported individuals, from their selection on the ramp to the removal of dead bodies from the chambers, lasted 1–3 h [4,55]. On 2 August 1943, an armed rebellion organised by the prisoners broke out in the camp. After the uprising in November 1943, the entire facility, including all installations, was dismantled. Some materials and equipment were taken away. It is estimated that around 800,000 people met their deaths in the Treblinka II extermination camp.
To carry out the experiment, we decided to use paper contact prints of photograph numbers 125 and 126, from a flight made on 15 May 1944 (Figure 3). They were scanned at a resolution of 600 DPI with the Epson Expression 10,000XL scanner. This means a scanning pixel of 42 µm, which, for a photonegative scale of 1:27,000, gives a terrain resolution of about 1.2 m. The used photos were characterised by negligible mechanical damage, including warping, scratches, and dust marks. The weather conditions during the flight were good. There is no atmospheric haze or clouds in the photographs.
The contemporary data employed in the study include data obtained via the use of LiDAR (Light Detection and Ranging; density of six points per sqm and an altimeter accuracy of about ±15 cm; acquisition date: 2011), as well as an orthophotomap (GSD 0.25 m; acquisition date: 2020). The LiDAR data came from a project conducted by the Head Office of Geodesy and Cartography that was carried out for Poland [57,58]. Use of these data is free for public administration entities and for scientific purposes.

5. Results

In the first stage of the experiment, the detection and description of keypoints were carried out. Thanks to the SfM measurement algorithms used in the program, a large number of points was measured. As a result of the automatic measurement, 3795 such points were detected, and 3145 remained after filtration (Figure 4).
This allowed us to build a coherent geometric block of two photos. On their basis, the relative orientation of the photos was determined. The accuracy of the relative orientation was 0.8 pixels.
For the absolute orientation (exterior orientation), we used a modern digital orthophotomap with a resolution of 0.25 m (terrain pixel) and localisation accuracy of 3 pixels (0.75 m), which covered the entire area of the country with an update cycle of every 2–3 years (map service of the Head Office of Geodesy and Cartography, coordinate system PUWG-1992). In the archival photographs and the modern orthophotomap, eight common terrain details were identified, which served as control points in the development process. Such details were intersections of road axes and intersections of parcel boundaries with local roads. Of these points, seven were treated as ground control points (GCP) and two as check points (ChP). The heights of these points were interpolated from the modern DTM.
Based on the control points, the external orientation of the photos was carried out. The accuracy obtained at the GCP was RMSXYZ = 1.0 m and RMSXYZ = 1.7 m at the ChP.
An attempt was made to produce a DTM from oriented pictures. A point cloud was created using Agisoft Metashape, which was to be used to create a DTM. The number of point clouds was 6,159,132. Unfortunately, the quality of this cloud was not satisfactory. The reason was the poor radiometric quality of the images, which caused high cloud noise. The obtained point cloud was characterized by numerous artifacts (holes) caused by the poor radiometry of the images. This is especially visible (Figure 5A) in the western part of the coverage of the images, as well as in the place of the open area (large white stain in the middle of the image), for which not enough points were measured. Upon analysis of the point cloud in the horizontal projection (Figure 5B), it can be seen that a large part of the points is located below the ground surface, which excludes the possibility of using it to create a DTM.
It was then decided that a DTM derived from LiDAR data covering the entire country would be used. It is a DTM in the GRID structure, with a resolution of 1 m (grid cell) and a height accuracy of 0.15 m. An orthophotomap was generated from archival photographs with a pixel size of 1.3 m.
The accuracy of the generated orthophotomaps was independently assessed using a resolution of 0.25 m (terrain pixel). On both orthophotomaps, eight controls points (Figure 6) were identified (other than those used as GCP for the orientation of archival photographs). The following accuracy for the generated orthophotomap was obtained: RMSX = 1.04 m and RMSY = 1.36 m. The results should be considered satisfactory and consistent with the obtained accuracy of the orthophotomap generation process [59].

6. Discussion

In the experiment, two Luftwaffe photographs with a stereoscopic coverage of the Treblinka II extermination camp were developed; these were obtained in 1944 by German air reconnaissance. These photos were available in the form of paper copies (contact prints), scanned on a non-professional scanner. The camera certificate was not available, and not all of the four corner fiducial marks were visible.
The camp area itself was heavily altered during its liquidation at the end of the war. In addition, the camp was located in a forestry and agricultural area. This means that it was not possible to find natural terrain features that could be identified after 80 years and used as a GCP for further development.
It should be emphasized that the presented situation is quite typical for photographic documentation, in the form of reconnaissance aerial photographs, of extermination camps from WWII. Furthermore, an even more typical case would be the development of only a single aerial photograph.
Luftwaffe photos in their current state would be difficult to develop using the classical approach. An approach based on the achievements of CV (Computer Vision) was thus used, based on the SfM (Structure from Motion) method, and implemented in the popular and available Agisoft Metashape software. This approach allowed us to find more than 3000 common keypoints on both images despite their poor radiometric quality.
The use of a modern orthophotomap and DTM developed from LIDAR data allowed for the identification of common terrain details in archival photos and orthophotomaps, which served as ground control points (GCP) [44]. This made it possible to georeference also archival Luftwaffe photos.
The inclusion of aerotriangulation with a self-calibration option in the development process made it possible to take into account significant deformations in the contact prints of archival photos (the shrinkage of the paper base) despite the lack of a camera certificate. The accuracy of such georeferencing was close to 1–1.5 pixels of the photos.
The orthorectification of archival photos with the use of modern DTM made it possible to assess the location accuracy of the orthophotomap obtained in this way through independent control in a modern orthorectification process. This accuracy was close to one pixel of the resulting orthophotomap.
These results, both from the georeferencing of archival Luftwaffe photos and the resulting orthophotomap, should be considered good as they are similar to those obtained in the development of modern digital photos.
An attempt was made to produce a DTM from oriented pictures. Unfortunately, the quality of the generated cloud was not satisfactory. Noisy DTM due to poor radiometry of archival photos are known in similar research studies [60]. Thus, it can be concluded that generating archival DTMs from Luftwaffe aerial photographs will be difficult, and the accuracies obtained will be low. Another problem may be the lack of photo coverage, which is quite common in reconnaissance flights performed by the Luftwaffe. Finding sorties characterized by good radiometric images and adequate coverage may become a serious challenge.
The obtained orthophotomap is a unique product documenting the area of the extermination camp. It is an ideal information layer that constitutes a background for other geodata: thermal photos, historical sketches, other contemporary photos, archaeological research, geophysical prospecting results, etc.

7. Conclusions

This article described the process of orthorectifying an archival Luftwaffe aerial photograph from May 1944, for the area of the Treblinka extermination camp, based on a computer-vision-based process and preprocessing techniques. Low-cost and easily accessible software was used, which generated a fully metric orthophotomap in a repeatable and accurate way. This process can be repeated for archival aerial photographs from other dates (for the Treblinka camp) and other extermination camps (Belzec and Sobibor). The generated orthophotomap is a material that can be used during (a) the analysis and reproduction of the topography of camps, (b) a search for mass graves, and (c) an analysis of the areas adjacent to the camps. The orthophotomap also allows for the recognition of objects in the field (based on GPS measurements) with an accuracy of about 1.5 m. This will enable the performance of geophysical and/or survey research on designated objects while interpreting archival photographs. Such an approach is fundamental when analysing camp objects which were deliberately destroyed, particularly those that have undergone the process of commemoration and the passage of time.
The orthophotomap is an ideal background (reference) layer in created spatial information systems. Other thematic layers and geodata can be superimposed (additional archival and contemporary photographs, results of their interpretation, analyses, GPR surveys, etc.).
The authors of this article are currently working on the orthorectification of multitemporal photographs for change detection. Experiments will include the SfM method, which will consider image orientations on points detected by different types of detectors and determine and transfer orientations between images acquired at different times.

Author Contributions

Methodology, Z.K. and A.K.K.; software, A.K.K.; validation, S.R.; resources, S.R.; writing—original draft preparation, Z.K., S.R. and A.K.K.; visualisation, S.R.; writing—review and editing, S.R, Z.K. and A.K.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Minister of Education and Science Republic of Poland, grant number NdS/544021/2021/2022 under the program: Science for Society. We acknowledge financial support by Faculty of Geodesy and Cartography within the funding programme Open Access Publishing.

Data Availability Statement

The data from Head Office of Geodesy and Cartography are available publicly. The photographs available in the National Archives and Records Administration are shared on a public domain basis. The data can be requested from the corresponding authors.

Acknowledgments

We would like to express our last thanks to the anonymous reviewers whose comments and suggestions helped improve and clarify this article.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Theune, C. Shadow of War: Archaeological Approaches to Uncovering the Darker Sides of Conflict from the 20th Century; Sidestone Press: Leiden, The Netherlands, 2018. [Google Scholar]
  2. Gross, Z.; Stevick, E.D. Introduction to the Open File: Holocaust education—International perspectives: Challenges, opportunities and research. Prospects 2010, 40, 17–33. [Google Scholar] [CrossRef]
  3. Writte, P.; Tyas, S. A new document on the deportation and murder of Jews during “Einsatz Reinhardt” 1942. Holocaust Genocide Stud. 2001, 15, 468–486. [Google Scholar] [CrossRef]
  4. Arad, Y. The Operation Reinhard Death Camps, Revised and Expanded Edition: Belzec, Sobibor, Treblinka; Indiana University Press: Bloomington, IN, USA, 2018. [Google Scholar]
  5. Różycki, S.; Nieradko, A.; Karczewski, J.; Schwarz, A. The use of non-invasive techniques in locating graves of Holocaust victims; the Rejowiec case study. Teledetekcja Sr. 2016, 54, 51–60. [Google Scholar]
  6. Klein, A. Rebuilding of a road across an old Jewish cemetery in northern Poland. In Construction for a Sustainable Environment; Sarsby, R., Meggyes, T., Eds.; CRC Press: London, UK, 2009; pp. 360–365. [Google Scholar]
  7. Stone, L. Quantifying the Holocaust: Hyperintense kill rates during the Nazi genocide. Sci. Adv. 2019, 5, eaau7292. [Google Scholar] [CrossRef]
  8. Webb, C. Sobibor Death Camp: History, Biographies, Remembrance; Columbia University Press: New York, NY, USA, 2017. [Google Scholar]
  9. Schute, I. In de Schaduw van een Nachtvlinder. Een Archeoloog op Zoek naar Sporen van de Holocaust; Prometheus: Amsterdam, The Netherlands, 2020. [Google Scholar]
  10. Lehnstaedt, S. Der Kern des Holocaust: Belzec, Sobibór, Treblinka und die Aktion Reinhardt; CH Beck: München, Germany, 2017. [Google Scholar]
  11. Wienert, A. Camp Cartography: On the Ambiguity of Mapping Nazi Extermination Camps. Zeitgeschichte 2018, 45, 575–598. [Google Scholar] [CrossRef]
  12. Ferguson, L. Aerial archives for archaeological heritage management: The aerial reconnaissance archives–A shared European resource. In Proceedings of the 11th EAC Heritage Management Symposium, Reykjavík, Iceland, 25–27 March 2010. [Google Scholar]
  13. Cowley, D.; Ferguson, L. Historic aerial photographs for archaeology and heritage management. In Proceedings of the Space Time and Place, III International Conference on Remote Sensing in Archaeology, Tamil Nadu, India, 17–21 August 2009. [Google Scholar]
  14. Kulha, N.; Pasanen, L.; Aakala, T. How to Calibrate Historical Aerial Photographs: A Change Analysis of Naturally Dynamic Boreal Forest Landscapes. Forests 2018, 9, 631. [Google Scholar] [CrossRef]
  15. Hajdukiewicz, H.; Wyżga, B. Aerial photo-based analysis of the hydromorphological changes of a mountain river over the last six decades: The Czarny Dunajec, Polish Carpathians. Sci. Total Environ. 2019, 648, 1598–1613. [Google Scholar] [CrossRef]
  16. Michałowska, K.; Głowienka, E. Multi-Temporal Analysis of Changes of the Southern Part of the Baltic Sea Coast Using Aerial Remote Sensing Data. Remote Sens. 2022, 14, 1212. [Google Scholar] [CrossRef]
  17. Đorđević, D.R.; Đurić, U.; Bakrač, S.T.; Drobnjak, S.M.; Radojčić, S. Using Historical Aerial Photography in Landslide Monitoring: Umka Case Study, Serbia. Land 2022, 11, 2282. [Google Scholar] [CrossRef]
  18. Jao, F.-J.; Chu, H.-J.; Tseng, Y.-H. Historical Image Registration and Land-Use Land-Cover Change Analysis. Environments 2014, 1, 181–189. [Google Scholar] [CrossRef]
  19. Carr, G.; Colls, C.S. Taboo and sensitive heritage: Labour camps, burials and the role of activism in the Channel Islands. Int. J. Herit. Studies 2016, 22, 702–715. [Google Scholar] [CrossRef]
  20. Jasinski, M.E. Predicting the past—Materiality of Nazi and post-Nazi camps: A Norwegian perspective. J. Histor. Archaeol. 2018, 22, 639–661. [Google Scholar] [CrossRef]
  21. Kamola, A.; Różycki, S.; Bylina, P.; Lewandowski, P.; Burakowski, A. Forgotten Nazi Forced Labour Camps: Arbeitslager Riese (Lower Silesia, SE Poland) and the Use of Archival Aerial Photography and Contemporary LiDAR and Ground Truth Data to Identify and Delineate Camp Areas. Remote Sens. 2020, 12, 1802. [Google Scholar] [CrossRef]
  22. Cowley, D.; Standring, R.; Abicht, M.J. Landscapes through the lens: An introduction. In Landscapes through the Lens: Aerial Photographs and the Historic Environment; Cowley, D., Standring, R., Abicht, M., Eds.; Oxbow, GB: Oxford, UK, 2010; pp. 1–6. [Google Scholar]
  23. Going, C.J. Signposts for GX–looking for prints. AARGnews 2006, 32, 30–31. [Google Scholar]
  24. Going, C.J. A neglected asset. German aerial photography of the second World War period. In Aerial Archaeology: Developing Future Practice; Bewley, R., Rączkowski, W., Eds.; NatoScience Series: Amsterdam, The Netherlands, 2002; pp. 23–30. [Google Scholar]
  25. Różycki, S. Charakterystyka kamer rozpoznawczych i zdjęć lotniczych z okresu II wojny światowej. Arch. Fotogram. Kartogr. Teledetekcji 2010, 21, 353–361. [Google Scholar]
  26. Davies, P. Imagery in the UK: Britain’s troubled imagery intelligence architecture. Rev. Int. Stud. 2009, 35, 957–969. [Google Scholar] [CrossRef]
  27. Cowley, D.C.; Ferguson, L.M.; Williams, A. The Aerial Reconnaissance Archives: A Global Aerial Photographic Collection. In Archaeology from Historical Aerial and Satellite Archives; Hanson, W., Oltean, I., Eds.; Springer: New York, NY, USA, 2013. [Google Scholar]
  28. Dwork, D.; Pelt, R.J. Auschwitz, 1270 to the Present; WW Norton & Company: London, UK, 1996. [Google Scholar]
  29. Bem, M. Sobibór Niemiecki Ośrodek Zagłady 1942–1943; Muzeum, B., Ed.; Hitlerowskiego Obozu Zagłady w Sobiborze: Sobibór, Polska, 2012. [Google Scholar]
  30. Gilead, I.; Haimi, Y.; Mazurek, W. Excavating Nazi extermination centres. Present Pasts 2010, 1, 12. [Google Scholar] [CrossRef]
  31. Schute, I. Collecting Artifacts on Holocaust Sites: A Critical review of Archaeological Research in Ybenheer, Westerbork, and Sobibor. Int. J. Hist. Archaeol. 2018, 22, 593–613. [Google Scholar] [CrossRef]
  32. Colls, C.S. Holocaust Archaeologies: Approaches and Future Directions; Springer: New York, NY, USA, 2015. [Google Scholar]
  33. Colls, C.S. Holocaust Archaeology: Archaeological Approaches to Landscapes of Nazi Genocide and Persecution. Ph.D. Thesis, University of Birmingham, Birmingham, UK, 2011. [Google Scholar]
  34. Różycki, S.; Michalski, M.; Kopówka, E. Obóz Pracy Treblinka I: Metodyka Integracji Danych Wieloźródłowych; Faculty of Geodesy and Cartography: Warsaw, Poland, 2022. [Google Scholar]
  35. Kola, A. Bełżec: The Nazi Camp for Jews in the Light of Archaeological Sources: Excavations 1997–1999; Council for the Protection of Memory of Combat and Martyrdom: Warsaw, Poland, 2000. [Google Scholar]
  36. Farella, E.; Morelli, L.; Remondino, F.; Mills, J.; Haala, N.; Crompvoets, J. The EuroSDR time benchmark for historical aerial images. In Proceedings of the International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLIII-B2-2022, ISPRS Congress, Nice, France, 6–11 June 2022. [Google Scholar]
  37. Giordano, S.; Bris, A.L.; Mallet, C. Toward automatic georeferencing of archival aerial photogrammetric surveys. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. 2018, 4, 105–112. [Google Scholar] [CrossRef]
  38. Pinto, A.T.; Gonçalves, J.A.; Beja, P.; Pradinho Honrado, J. From Archived Historical Aerial Imagery to Informative Orthophotos: A Framework for Retrieving the Past in Long-Term Socioecological Research. Remote Sens. 2019, 11, 1388. [Google Scholar] [CrossRef]
  39. Redweik, P.; Roque, D.; Marques, A.; Matildes, R.; Marques, F. Recovering Portugal Aerial Images Repository. Int. Arch. Photogramm. Remote Sens. 2009, 38, 1–4. [Google Scholar]
  40. Gonçalves, J.A. Automatic orientation and mosaicking of archived aerial photography using structure from motion. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 2016, 40, 123–126. [Google Scholar] [CrossRef]
  41. Jaworek-Jakubska, J.; Filipiak, M.; Michalski, A.; Napierała-Filipiak, A. Spatio-Temporal Changes of Urban Forests and Planning Evolution in a Highly Dynamical Urban Area: The Case Study of Wrocław, Poland. Forests 2020, 11, 17. [Google Scholar] [CrossRef]
  42. Fabris, M. Monitoring the Coastal Changes of the Po River Delta (Northern Italy) since 1911 Using Archival Cartography, Multi-Temporal Aerial Photogrammetry and LiDAR Data: Implications for Coastline Changes in 2100 A.D. Remote Sens. 2021, 13, 529. [Google Scholar] [CrossRef]
  43. Marić, I.; Panđa, L.; Faričić, J.; Šiljeg, A.; Domazetović, F.; Marelić, T. Long-Term Assessment of Spatio-Temporal Landuse/Landcover Changes (LUCCs) of Ošljak Island (Croatia) Using Multi-Temporal Data—Invasion of Aleppo Pine. Land 2022, 11, 620. [Google Scholar] [CrossRef]
  44. Nocerino, E.; Menna, F.; Remondino, F. Multi-Temporal Analysis of Landscapes and Urban Areas. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 2012, 39, 85–90. [Google Scholar] [CrossRef]
  45. Boyd, R.J. Project Casey Jones’: Post-hostilities Aerial Mapping; Iceland, Europe, and North Africa, June 1945 to December 1946; Strategic Air Command: Ashland, NE, USA, 1988. [Google Scholar]
  46. Alomar-Garau, G.; Grimalt-Gelabert, M. Air Photo Interpretation for Spatial Analysis of Heritage Agrarian Structures in Mediterranean Settings as Sea-Breezes Proxy-Data. Application to the Island of Mallorca. Remote Sens. 2021, 13, 4652. [Google Scholar] [CrossRef]
  47. Seccaroni, S.; Santangelo, M.; Marchesini, I.; Mondini, A.C.; Cardinali, M. High Resolution Historical Topography: Getting More from Archival Aerial Photographs. Proceedings 2018, 2, 347. [Google Scholar]
  48. Persia, M.; Barca, E.; Greco, R.; Marzulli, M.I.; Tartarino, P. Archival Aerial Images Georeferencing: A Geostatistically-Based Approach for Improving Orthophoto Accuracy with Minimal Number of Ground Control Points. Remote Sens. 2020, 12, 2232. [Google Scholar] [CrossRef]
  49. Nebiker, S.; Lack, N.; Deuber, M. Building Change Detection from Historical Aerial Photographs Using Dense Image Matching and Object-Based Image Analysis. Remote Sens. 2014, 6, 8310–8336. [Google Scholar] [CrossRef]
  50. Grottoli, E.; Biausque, M.; Rogers, D.; Jackson, D.W.T.; Cooper, J.A.G. Structure-from-Motion-Derived Digital Surface Models from Historical Aerial Photographs: A New 3D Application for Coastal Dune Monitoring. Remote Sens. 2021, 13, 95. [Google Scholar] [CrossRef]
  51. Bożek, P.; Janus, J.; Mitka, B. Analysis of Changes in Forest Structure using Point Clouds from Historical Aerial Photographs. Remote Sens. 2019, 11, 2259. [Google Scholar] [CrossRef]
  52. Markiewicz, J.; Abratkiewicz, K.; Gromek, A.; Ostrowski, W.; Samczynski, P.; Gromek, D. Geometrical Matching of SAR and Optical Images Utilizing ASIFT Features for SAR-based Navigation Aided Systems. Sensors 2019, 19, 5500. [Google Scholar] [CrossRef] [PubMed]
  53. Karwel, A.K.; Markiewicz, J. The methodology of the archival aerial image orientation based on the Structure-from-Motion method. Sens. Mach. Learn. Appl. 2022, 1, 1–19. [Google Scholar] [CrossRef]
  54. Zawieska, D.; Markiewicz, J.; Kopiasz, J. Development of true orthophotomaps of the fortified settlement at Biskupin, Site 4, based on archival data. Archaeol. Prospect. 2019, 26, 1–28. [Google Scholar] [CrossRef]
  55. Kopówka, E.; Rytel-Andrianik, P. Dam im Imię na Wieki (Iz 56, 5): Polacy z Okolic Treblinki Ratujaący Żydów; Wydawnictwo Sióstr Loretanek: Siedlce, Poland, 2011. [Google Scholar]
  56. Kopówka, E. Treblinka—Nigdy Więcej; Muzeum Walki i Męczeństwa w Treblince: Siedlce, Poland, 2002. [Google Scholar]
  57. Bakuła, K.; Ostrowski, W.; Zapłata, R.; Kurczyński, Z.; Kraszewski, B.; Stereńczak, K. Zalecenia w Zakresie Pozyskiwania, Przetwarzania, Analizy i Zastosowania Danych LIDAR w Celu Rozpoznania Zasobów Dziedzictwa Archeologicznego w Ramach Programu AZP. Available online: https://repo.pw.edu.pl/fileView.seam?fileName=2018_Baku%C5%82a_i_inni_Zalecenia_odno%C5%9Bnie_99.pdf&entityType=article&fileTitle=2018_Baku%C5%82a_i_inni_Zalecenia_odno%C5%9Bnie_99.pdf+%2F+1+MB+%2F+&lang=pl&fileId=WUTb37b3e7a398747e088e943b554030b8e&cid=1178894 (accessed on 15 April 2023).
  58. Zapłata, R.; Bakuła, K.; Stereńczak, K.; Kurczyński, Z.; Kraszewski, B.; Ostrowski, W. Zalecenia Odnośnie do Pozyskiwania, Przetwarzania, Analizy i Wykorzystania Danych LiDAR w Celu Rozpoznania Zasobów Dziedzictwa Archeologicznego w Ramach Programu AZP—Między Teorią a Praktyk. 2018. Available online: https://repo.pw.edu.pl/fileView.seam?fileName=2018_Baku%C5%82a_i_inni_Zalecenia_odno%C5%9Bnie_99.pdf&entityType=article&fileTitle=2018_Baku%C5%82a_i_inni_Zalecenia_odno%C5%9Bnie_99.pdf+%2F+1+MB+%2F+&lang=pl&fileId=WUTb37b3e7a398747e088e943b554030b8e&cid=1179018 (accessed on 15 April 2023).
  59. Kurczyński, Z.; Falkowski, P.; Uchański, J. Ortofotomapa zniszczonej Warszawy. Mag. Geoinf. GEODETA 2007, 2, 14–18. [Google Scholar]
  60. Giordano, S.; Le Bris, A.; Mallet, C. Fully automatic analysis of archival aerial images current status and challenges. In Proceedings of the Joint Urban Remote Sensing Event (JURSE), Dubai, United Arab Emirates, 1–4 March 2017. [Google Scholar]
Remotesensing 15 02587 g001 550
Figure 1. (A) Geographical location and aerial view of former German extermination camps: (B) Treblinka, (C) Sobibor, and (D) Belzec. The 2020 orthophotomap was used as a base for all the camps. The 2020 orthophotomap was used as a base for all camps [© Head Office of Geodesy and Cartography, NARA]. Coordinate grid: Polish Coordinate System 1992 (EPSG: 2180).
Figure 1. (A) Geographical location and aerial view of former German extermination camps: (B) Treblinka, (C) Sobibor, and (D) Belzec. The 2020 orthophotomap was used as a base for all the camps. The 2020 orthophotomap was used as a base for all camps [© Head Office of Geodesy and Cartography, NARA]. Coordinate grid: Polish Coordinate System 1992 (EPSG: 2180).
Remotesensing 15 02587 g001
Remotesensing 15 02587 g002 550
Figure 2. The incremental SfM procedure.
Figure 2. The incremental SfM procedure.
Remotesensing 15 02587 g002
Remotesensing 15 02587 g003 550
Figure 3. The outline of the aerial photographs taken in 1944 was used to generate the orthophoto (yellow) and the area of the current Treblinka camp commemoration (red trapezoid). The 2020 orthophotomap was used as a base [© Head Office of Geodesy and Cartography, NARA]. Coordinate grid: Polish Coordinate System 1992 (EPSG: 2180).
Figure 3. The outline of the aerial photographs taken in 1944 was used to generate the orthophoto (yellow) and the area of the current Treblinka camp commemoration (red trapezoid). The 2020 orthophotomap was used as a base [© Head Office of Geodesy and Cartography, NARA]. Coordinate grid: Polish Coordinate System 1992 (EPSG: 2180).
Remotesensing 15 02587 g003
Remotesensing 15 02587 g004 550
Figure 4. Distribution of keypoints on images 125 (A) and 126 (B). Points in blue—accepted; in grey—rejected; GCP—light green flag symbol with white dot (8 points); CP—light green flag symbol with red dot (2 points) [© NARA].
Figure 4. Distribution of keypoints on images 125 (A) and 126 (B). Points in blue—accepted; in grey—rejected; GCP—light green flag symbol with white dot (8 points); CP—light green flag symbol with red dot (2 points) [© NARA].
Remotesensing 15 02587 g004
Remotesensing 15 02587 g005 550
Figure 5. Point cloud obtained from Agisoft Metashape software: (A) vertical projection of dense cloud; (B) horizontal projection of dense cloud.
Figure 5. Point cloud obtained from Agisoft Metashape software: (A) vertical projection of dense cloud; (B) horizontal projection of dense cloud.
Remotesensing 15 02587 g005
Remotesensing 15 02587 g006 550
Figure 6. Spatial distribution of the eight control points on orthophotomap from 1944 (with transparency 70%). The Open Street Map was used as a base [© OpenStreetMap—Open Database License, Head Office of Geodesy and Cartography]. Coordinate grid: Polish Coordinate System 1992 (EPSG: 2180).
Figure 6. Spatial distribution of the eight control points on orthophotomap from 1944 (with transparency 70%). The Open Street Map was used as a base [© OpenStreetMap—Open Database License, Head Office of Geodesy and Cartography]. Coordinate grid: Polish Coordinate System 1992 (EPSG: 2180).
Remotesensing 15 02587 g006
Table
Table 1. Parameters of aerial photographs (Luftwaffe) for Extermination Camps.
Table 1. Parameters of aerial photographs (Luftwaffe) for Extermination Camps.
Extermination CampsImage DateScale of PhotonegativeFocal Length [mm]Format [cm × cm]
Belzec26.05.19401:28,000199.9118 × 18
Belzec11.07.19401:80,000??18 × 18
Belzec13.05.19441:14,000200.4730 × 30
Belzec15.09.19441:17,500752.3030 × 30
Belzec17.09.19441:15,500752.3030 × 30
Belzec18.11.19441:35,000200.6430 × 30
Belzec19441:40,000200.3630 × 30
Belzec19441:18,000??18 × 18
Belzec19441:13,000751.6130 × 30
Treblinka19401:80,000??18 × 18
Treblinka13.05.19441:33,000200.7230 × 30
Treblinka15.05.19441:27,000200.7230 × 30
Treblinka02.09.19441:50,000200.8630 × 30
Treblinka18.09.19441:44,000200.4330 × 30
Treblinka29.09.19441:44,000200.6030 × 30
Sobibor30.09.19411:10,000Bildplan
Sobibor28.03.19441:40,000??30 × 30
Sobibor28.03.19441:46,000200.8030 × 30
Sobibor28.05.19441:23,500200.8530 × 30
Sobibor29.05.19441:20,000201.1830 × 30
Sobibor30.05.19441:31,000200.6130 × 30
Sobibor??1:17,000210.7718 × 18
Note: “??” means that the date is unknown in these cases.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Różycki, S.; Karwel, A.K.; Kurczyński, Z. German Extermination Camps on WWII Reconnaissance Photographs. Orthorectification Process for Archival Aerial Images of Cultural Heritage Sites. Remote Sens. 2023, 15, 2587. https://doi.org/10.3390/rs15102587

AMA Style

Różycki S, Karwel AK, Kurczyński Z. German Extermination Camps on WWII Reconnaissance Photographs. Orthorectification Process for Archival Aerial Images of Cultural Heritage Sites. Remote Sensing. 2023; 15(10):2587. https://doi.org/10.3390/rs15102587

Chicago/Turabian Style

Różycki, Sebastian, Artur Karol Karwel, and Zdzisław Kurczyński. 2023. "German Extermination Camps on WWII Reconnaissance Photographs. Orthorectification Process for Archival Aerial Images of Cultural Heritage Sites" Remote Sensing 15, no. 10: 2587. https://doi.org/10.3390/rs15102587

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop