Skip to navigation – Site map

HomeNuméros9DossierClearing the Fog. Early Image Pro...

Dossier

Clearing the Fog. Early Image Processing and the Lunar Surveyor Imaging System (1966‑68)

Dissiper le brouillard. Les premiers traitements d’images et le système d’imagerie lunaire Surveyor (1966‑1968)
Frances Cullen
Translation(s):
Dissiper le brouillard. Les premiers traitements d’images et le système d’imagerie lunaire Surveyor (1966‑1968) [fr]

Abstracts

Over the course of the 1960s, the National Aeronautics and Space Administration (NASA) used some of its substantial resources to develop novel computer-based techniques for improving and enhancing its astronomical images. This article illuminates that early chapter in the history of the algorithmic image by describing the circumstances of the technology’s integration into the imaging system of one NASA program in particular: the Surveyor mission (1966–68). Surveyor used a television camera to capture images of the lunar surface, then telemetered those images back to Earth. There, the images were circulated using a diverse range of technologies that included magnetic tape, various film formats, and a combination of manual and computer-based processing techniques. The analysis shows how, as they were introduced into this already complex and diverse imaging system, these novel image-processing techniques operated as a sort of extended darkroom, essentially duplicating and augmenting an existing logic of information extraction and processing.

Top of page

Full text

1In 1959, the Soviet space program achieved something remarkable. Using a film camera, an automated darkroom, and a flying-spot scanner, its flyby space probe Luna 3 captured and transmitted to Earth the first pictures of the far side of the moon. The results were a revelation, both literally and figuratively. What had hitherto been shrouded in darkness was now available to human observation. On the strength of these images, the Soviets were able to produce a tentative early map of the distant lunar topography.

2This event undoubtedly represents a major landmark in the overlapping histories of astronomical image-making, space exploration, and the planetary sciences. At least from the United States’ perspective, however, it also premises a story in which Soviet shortcomings give way to American ingenuity. This is because Luna 3’s imaging system had created a problem in need of resolution. The informational value of its groundbreaking images had been compromised by the process of transmission, which had the deleterious effect of introducing frequency distortion and spurious noise.

  • 1 “Computer Process Brightens Surveyor Moon Pictures,” Jet Propulsion Laboratory, August 9, 1966, htt (...)

3A press release from 1966 recounts how a desire to improve on the quality of the Soviet pictures had spurred the National Aeronautics and Space Administration (NASA)’s early development of computer-based image processing technologies.1 Apparently the idea for such an application had struck Dr. Robert Nathan, a computer engineer who worked for NASA’s Jet Propulsion Laboratory (JPL), while he was reviewing the Luna 3 images in 1962. “It was quite clear,” he explained, “that extraneous noise had distorted their pictures and severely handicapped analysis.” The Soviets had attempted to correct for this shortcoming with the use of traditional, darkroom-based techniques, but Nathan was convinced that he could achieve better results by using a computer.

  • 2 Elizabeth A. Kessler, “Technology’s Palette: Voyager’s Eyes and the Hyperchromatic Enhancement of J (...)

4Luckily, his superiors at the JPL had both the means and the motivation to support his proposal. With their approval and the collaboration of his colleague, Fred Billingsley, Nathan founded a JPL subsidiary called the Image Processing Laboratory (IPL).2 In July of 1964, NASA’s Ranger VII contested the accomplishment of the Luna mission not only by using its television camera system to become the first American spacecraft to take a picture of the moon (fig. 1) but also by generating the first astronomical images to be digitally corrected and enhanced.

1. Image of the Moon taken by the American space probe Ranger VII, July 31, 1964

1. Image of the Moon taken by the American space probe Ranger VII, July 31, 1964

Greenbelt, NASA Space Science Data Coordinated Planetary Image Archives.

© NASA/JPL

  • 3 Jet Propulsion Laboratory, “Computer Process Brightens.”

5Within two years, some of those same Ranger images had been processed six or more times.3 NASA had also, by this time, launched its Surveyor program (1966–68). Under pressure to prepare for Apollo’s upcoming manned missions to the moon, Surveyor’s main purpose was to test the viability of the organization’s soft-landing technologies for use on the lunar surface; its successful operation on the ground had the effect of enabling the production of ever more and better lunar images. This, in turn, created the demand and the opportunity to develop ever more and better formulas for the correction and enhancement of all this visual data.

  • 4 Peter J. Westwick, “Visual Imagery in Solar System Exploration,” in Exploring the Solar System: The (...)

6The IPL, of course, did not disappoint. On the contrary, both throughout the Surveyor mission and beyond it, the group’s image processing program became both more sophisticated and more robust. The historian Peter J. Westwick judges that, by the 1970s, “JPL and the planetary program had perhaps the most advanced digital image processing capability in the country at the time.”4 That capability continued to develop and spread across the 1970s until, by the end of the decade, the ability to digitally enhance images had hit the commercial market.

7This article illuminates the early history of the digitally processed image by describing the circumstances of its integration into the imaging system of the Surveyor mission. Although Ranger might seem to present a more obvious choice for study, focusing on Surveyor instead allows my account of this technology’s formative years to circumvent the thorny issue of origins in favor of a more instructive emphasis on the vicissitudes of introduction.

8Specifically, the article highlights the fact that, despite the impressive technical accomplishments of the IPL, their new image processing techniques were at this point still experimental in nature and, as a result, relatively limited in application. Like Ranger before it, the Surveyor images had been captured by a television camera. Their circulation back on Earth then involved a diverse range of formats and technologies that included magnetic tape, various film formats, and a combination of manual and computer-based processing techniques.

9My analysis shows how, as they were inserted into this already complex, advanced, and constantly evolving imaging system, the new image processing techniques operated as a sort of extended darkroom, essentially duplicating and augmenting an existing logic of information capture and development. Even today, in the context of a recent digitization project, the algorithm automates a process of information extraction whose purpose is to enable further processing and analysis of the visual data in the Surveyor archive.

  • 5 W. Patrick McCray, “How Astronomers Digitized the Sky,” Technology and Culture 55, no. 4 (2014): 93 (...)
  • 6 In addition to the work of Westwick, the histories written by Kessler and Batchen have been crucial (...)

10Against a binary historical framework that would posit a straightforward transition from analog to digital, the historian W. Patrick McCray has argued that the history of astronomical knowledge production comprises “three overlapping data eras,” with the middle one—the “electronic” era, which, positioned between the “photographic” and the “digital,” is “transitional and adjoining”—characterized by a collaboration of analog and digital forms.5 It is in the spirit of this insight that this article contributes to the existing historiography a story not of the development of digital processing technologies at the JPL but of their logical and technical integration into an existing system of information capture and processing.6 In doing so, it also reveals something about the system’s epistemology of visual information: it is somehow both constant and malleable, disembodied and inextricable from the conditions of its own production.

Beginnings at the Image Processing Lab

  • 7 Billingsley, quoted in Kessler, “Technology’s Palette.”

11In 1965, Fred Billingsley, a JPL engineer who had collaborated with Nathan to found the IPL, delivered a paper at a conference on “Electronic Imaging Techniques.” In this paper, he reported on the program’s activities during its first few years of operations, which he characterized as a triumph. “This whole project grew out of a need to improve the pictures from the cameras which were used in the early days of the Ranger Project,” he wrote. “It was apparent from looking at the early camera tests that the pictures would not be satisfactory unless considerable improvement could be done.”7

  • 8 R. Cargill Hall, Lunar Impact: A History of Project Ranger (Washington, DC: Scientific and Technica (...)
  • 9 Jet Propulsion Laboratory, “Computer Process Brightens.”
  • 10 Raymond L. Heacock, “Lunar Photography: Techniques and Results,” Space Science Reviews 8, no. 2 (19 (...)

12In other words, the enterprise of the IPL was justified by the fact that NASA’s growing archive of in-flight images could not be made useful without the benefit of that group’s substantial interventions. According to NASA’s internally published history of the Ranger program, the Ranger camera team had achieved good effects by running their television pictures through programs that “removed spurious noise received with Ranger’s picture signal.” They also successfully applied algorithms that “enhanced contrasts in the photographs of the lunar surface by shifting the mean intensity level and expanding it to cover the full range of the gray scale from black to white.”8 By the time that Surveyor began, its own hazy television pictures could be brought by the new, digital techniques to such a resolution that scientists were able to calculate relief at the granularity of half a millimeter; this, in some instances, “apparently doubled the observable details.”9 The “processed” image presented in figure 2, for example, demonstrates how the interventions that were made on one particular photograph from Surveyor I could allow the camera team to make observations about the cohesive consistency of the soil.10

2. “Original” image taken by the Surveyor spacecraft and its “processed” version

2. “Original” image taken by the Surveyor spacecraft and its “processed” version

Illustrations taken from R. H. Steinbacher et al., “Surveyor I Mission Report”, part 3, “Television Data” (Pasadena, CA: Jet Propulsion Laboratory, 1966), 47 and 49.

© NASA/JPL

  • 11 Jet Propulsion Laboratory, Computer Process Brightens.”
  • 12 Kessler, “Technology’s Palette,” 1099.

13Such an impressive degree of accomplishment prompted Nathan to boast in 1965 that by devising corrective formulas and applying these to the enhancement of the video images, “we are removing the fog in our pictures of our planetary system.”11 Billingsley, for his part, went so far as to credit the development of such techniques for correcting and enhancing analog images with the success of the earliest lunar missions.12 By their collective account, then, the rise of computer-based image processing was not just magically revealing the secrets of the skies, it was also indispensable to the operations of the space exploration program.

14In the face of all this laudatory discourse about the wildly successful contributions of the IPL, the present analysis of Surveyor’s imaging system offers a more sober (which is to say, realistic) view of the actual state of computer-based image processing, at least before and during the narrow period of its operation. This is not to deny the veracity of accounts that stress the technology’s rapid state of development at this time: it is true that by the time the Surveyor mission launched, the IPL team had authored formulas that could alter images by correcting distortion, removing blemishes, and improving resolution. They could use these computer processes to stretch image contrast and create topographic maps of the moon’s surface.

  • 13 This phrase appears in several of the Surveyor mission television data reports. In the report for S (...)

15The published reports for the Surveyor mission note that “digital computer techniques [had been] developed and utilized in conjunction with the Ranger and Mariner photographic experiment and applied to Surveyor imagery”;13 by all accounts, the technology only became more advanced throughout the duration of the Surveyor mission. Indeed, reading these same reports in sequence does evince a sense of the computer-based techniques becoming, over even this short period of time, increasingly established as a component of NASA’s regular system for astronomical image production. And yet the role that digital processing had to play within the program’s system of visual knowledge production was, in practice, both limited and marginal.

  • 14 This phrase appears in every television data report. In addition to Gunter et al.’s Surveyor I repo (...)
  • 15 The Surveyor reports specify that the qualifier “processed” means “computer processed.” The report (...)

16This insight, too, is affirmed by the contents of the reports, which make it clear, precisely by stating it over and over, that the Surveyor team had corrected for distortions “by means of either manual or computer techniques.”14 In practice, preference was given to the former kind, by which the reports’ authors mean darkroom processing, on the one hand, and the hand-pasting of prints to form mosaics on the other. Thus, of the roughly 90,000 images that were taken during the Surveyor program in total, 1,042 of these were catalogued in the initial round of reports; and of these, only 150 are identified as “processed,” meaning computer enhanced.15

  • 16 Jason Davis, “Surveyor Digitization Project Will Bring Thousands of Unseen Lunar Images to Light,” (...)

17In addition to being relatively rare, the Surveyor team’s use of computer-based processing techniques was, in practice, belated, in the sense that it came into service late in the image processing workflow and only post-flight. The bulk of the image handling before that point took place without computer input and as a relay between film-based and televisual technologies. Competition for computer time, meanwhile, was at such a premium that if the IPL scientists wanted time with the machine, they were reduced to booking in slots overnight.16

18The corollary here is that while the new image processing technology must have made some important contributions to the Surveyor program’s overall record of success, these contributions would have been considerably more modest than the credit that was claimed for them. Instead, the technology’s early development was pursued in the spirit of experimentation, with the benefit of the vast network of resources and expertise that could be afforded by an organization like NASA. Its tentative and peripheral integration into the astronomical imaging systems of the day, meanwhile, was as a complement to rather than a replacement for another image-making technology: the television camera, a technology that was even then in a state of ongoing development and improvement itself.

The Television Camera

19The Surveyor program’s central objective was to test the viability of NASA’s Apollo design not only for soft landing and subsequent lift but also—and just as importantly—for safe operation on the surface of the moon. This meant, in part, that the program had to answer certain questions about the moon’s surface and atmosphere. The result of this mandate was a sequenced roster of spacecraft whose landing delivered to the lunar surface, in addition to the standard operational subsystems (of, for example, flight control, telecommunications, and propulsion), a complex system of gadgetry for the gathering of scientific data.

  • 17 Tom Lund, “The Surveyor Lunar Landing Mission,” in Early Exploration of the Moon: Ranger to Apollo, (...)

20Paramount among these data-collecting tools was the spacecrafts’ cameras and their related subsystems. The importance of the camera is reflected in the overall design of the spacecraft, which in many ways revolved around ensuring and enhancing the value of the imaging system’s visual outputs (fig. 3). For example, the spacecraft had photometric and colorimetric targets mounted on their surfaces, which could be included within image frames as photographic points of reference.17 In fact, so indispensable was the camera system to the mission that the first two of the seven Surveyor crafts were equipped with the television camera exclusively. Even when additional instruments were added to later Surveyor models (a soil mechanics surface sampler on Surveyor III; an alpha-scattering surface analyzer on Surveyor V), these additions had been selected largely to buttress and supplement the work of the camera.

3. “Model of the Surveyor VII spacecraft”

3. “Model of the Surveyor VII spacecraft”

Illustration from T. H. Bird et al., “Surveyor VII Mission Report”, part 3, “Television Data” (Pasadena: Jet Propulsion Laboratory, 1968), vi.

© NASA/JPL

  • 18 Klaus Brasch, “A Short History of Astrophotography: Part 2,” Journal of the Royal Astronomical Soci (...)
  • 19 Heacock, “Lunar Photography,” 223.

21Although the technology’s roots go back decades, astronomers did not begin to adopt the television camera until the late 1950s.18 NASA itself was not founded until 1958; even then, it was only in 1964 that they managed to use a television camera to take their first in-flight photograph of the moon. This newfound application created an ongoing pressure to improve, customize, and refine the specifics of the camera system design such that it could meet competing demands for quality, quantity, and range, all while operating under extreme conditions and at a substantial distance from Earth. Specifically, NASA’s camera engineers aimed to strike a balance between four main considerations: imaging systems needed to aim for a fast frame time, which would maximize the number of photographs taken; for a high number of scan lines, in order to maximize resolution; and for the lowest possible bandwidth, which would reduce the chance of interference in the transmission process undermining the quality of the image received. Engineers also had to consider the atmospheric conditions on the moon, with respect to their choice of component materials.19

  • 20 Lund, “Surveyor Lunar Landing Mission,” 89.

22In the case of the Surveyor imaging system specifically, it worked like this: a combined pan-head and tilting mirror allowed the camera the flexibility to cover a horizontal area of almost 360 degrees and an elevation of 60 degrees. Between the mirror and the lens, the pure image was made to pass through a wheel which could be rotated to set in place one of three different polarization filters, plus a neutral-density filter. All the camera’s many mechanisms—camera position, mirror angle, filter, zoom, focal distance, iris setting, and shutter time—could be controlled by the television camera team on Earth. Three different shutter modes (“normal,” which opened for about 150 milliseconds; “open shutter,” which opened for roughly 1.2 seconds; and “integrate exposure,” which stretched to several minutes) and two scan settings (200- and 600-line) were made available to accommodate the different lighting conditions and imaging objectives.20

  • 21 Bird et al., Surveyor VI Mission Report, 9.

23This range in parameters is a crucial context for understanding the way the images produced were received by NASA’s operators on Earth: namely, as highly mediated documents in need of delimitation and control. The Surveyor Television Data reports explain that “to derive maximum scientific information from a picture, it is necessary to have precise quantitative information on the camera that obtained the picture in terms of those parameters that describe the quality of the image.”21 To satisfy this requirement, each of the Surveyor cameras was subjected to a comprehensive calibration process prior to its launch. Camera parameter data was also collected while the craft was in operation and telemetered back to Earth along with the images, for use in the subsequent analysis of the pictures.

  • 22 Davis, “Surveyor Digitization Project.”

24When that pure optical image, having already refracted through the lens and mirror mechanism, struck the photosensitive surface of the vidicon tube, it was converted into a video signal and transmitted back to Earth. Then the image was recorded, along with all its calibration data, first onto magnetic tape and subsequently onto rolls of film.22 It is in these forms that the images could then be submitted for processing, either by the computer mainframe or in the JPL’s in-house photo lab.

  • 23 Heacock, “Lunar Photography,” 243.
  • 24 Westwick, “Visual Imagery,” 151.

25Again, the process of transmission was a major source of information distortion and loss, which is the main reason that the engineering team at the JPL worked to devise its corrective formulas. Such programs were primarily written (and, as we have seen, selectively applied) to correct images “for geometric nonlinearities and distortions, frequency or aperture response, photometric non-uniformities, and coherent noise”;23 as Westwick puts it, “like a photographer developing film in the darkroom, engineers had to apply digital processes to reveal images contained in the streams of digital bits.”24 In doing so, they did not just replicate the basic photographic logic, wherein raw visual information is extracted and developed. They also extended that logic, as the computer-enhanced image found itself operating in tandem and in concert not only with television technologies but with film-based ones too.

Image Processing as a Matter of Degree

  • 25 Bird et al., Surveyor VI Mission Report, 19.

26Upon their telemetering back to Earth, those television images had a famous first destination: they went direct to TV. During this heady period of politicized space exploration, NASA was in the practice of broadcasting its collected visual data live on television. It also received those images on “standard video monitors” that were set up in its headquarters.25 This immediate feedback measure facilitated informed real-time reactions on the part of the JPL engineers whose job it was to operate the craft in flight. In this way, the Surveyor images were involved, from the start, in a process of analysis and improvement; a first degree of feedback had already determined the form and substance of the Surveyor television pictures even before they had been made.

27From there, every version of a given Surveyor image frame seems to exist at a remove from the one that precedes it. The master positives, themselves derived from the tape recordings that had originally been received at NASA’s various tracking stations, already represent a first step in the process that saw Surveyor’s visual information successively, and repeatedly, reproduced and refined. All subsequent reproductions were then made from these. At any stage in that reproductive process, a given image could be assimilated into the service of a real-time workflow that merged, at increments, material techniques for image processing with cognitive methods of scientific analysis.

  • 26 The term “operations” is taken directly from the reports, where it basically describes image proces (...)
  • 27 Bird et al., Surveyor VII Mission Report, 27.

28This dynamic was at play from the very moment of initial receipt, since the JPL engineers did not just use the images on their screen to dictate their command of the lunar craft. That uncalibrated data was also printed as individual frames on a “continuously processed, directly recorded paper strip,” which was then used to prepare “real-time” or “operations” mosaics.26 This work facilitated the “immediate analysis of the surface features and camera performance”; its results are partially rephotographed and published in the reports but are specifically described as being “not of archival quality.”27

  • 28 This wording is taken from the Surveyor VI report (p. 29). Descriptions of the mosaics appear, with (...)

29As higher-quality prints began to come in from the tracking stations, the JPL team was able to use the calibration data that came with it to construct mosaics, thereby achieving ever-greater degrees of accuracy. The mission reports divide the mosaics into a number of categories accordingly. “Semi-improved” mosaics were prepared during the mission and represented a greater degree of accuracy; “improved” mosaics were created post-mission and “represent[ed] the final or best quality mosaic information.” 28“Special” mosaics, meanwhile, harnessed the highest level of skill, attention, and time to yield the highest level of detail.

30The default through all of this image handling was to rely on what one report refers to as “wet-processed prints,” but there was another option too. Most photographs were, as I have noted, developed in-house at the JPL. If a team member wished to subject an image to the process of computer-based enhancement, then this added an extra step to the process. They could input their image to the IBM mainframe, which would digitize and process it according to one of its library of preprogrammed scripts.

  • 29 Westwick, “Visual Imagery,” 153.

31At this stage in its history, the JPL engineers did not have access to interactive digital screens. The computer would spit out the processed image printed on another strip of film. Before they could see it, the team member would have to wait even longer than the hour it might take for the program to run, as it took another one to two days to do the processing in the photo lab.29 Only then could a “processed” image be integrated into a manually constructed mosaic (fig. 4).

4. Mosaic made in part from “processed” photos of Surveyor I, November 11, 1966

4. Mosaic made in part from “processed” photos of Surveyor I, November 11, 1966

Pasadena, Jet Propulsion Laboratory Archives.

© NASA/JPL; © Jet Propulsion LaboratoryArchives

32The algorithm’s integration within the overall system, then, was not just limited and marginal, it was also logical. This is to say that the specific function was both duplicate and additive, analogous to and sequential within that of the system as a whole. It opened new possibilities for a heightened degree of image enhancement and processing, while slotting that function neatly into a system for which that had already been the main objective; it is just that, like the famous Hitchcockian dolly zoom, the further the imaging technologies can bring us from the raw captured data, the clearer our view of the scene.

  • 30 Davis, “Surveyor Digitization Project.”

33The hybrid nature of this system is aptly embodied by the figure of the individual image frame (fig. 5). Once the second-generation copies had been made and circulated, the master negatives were clipped and catalogued at the JPL.30 There, they might meet with a range of possible fates: they could be accessed for information, sent to the darkroom for development, or brought to the IBM mainframe for processing.

5. Surveyor VII image, together with its metadata, 1968

5. Surveyor VII image, together with its metadata, 1968

70 mm film, Tucson, University of Arizona, Lunar and Planetary Laboratory.

© NASA/JPL; © Jet Propulsion LaboratoryArchives

  • 31 “Surveyor Digitization: Decoding the Code,” Space Imagery Center, Lunar and Planetary Laboratory, U (...)
  • 32 Space Imagery Center, “Decoding the Code.”

34This range of possible uses is reflected in the appearance of the individual frame, less than half of which is actually occupied by the subject image. The rest of the space is filled with calibration information. In the immediate vicinity of the image, this data is human-readable: to the image’s left is a grayscale strip and to its right the image metadata, in this instance presented as a combination of numbers and text. Beyond that, to the right, this same metadata is represented in two different machine-readable codes. A binary dot code conveys “the technical parameters of the image frame; such as azimuth, elevation, focal length, focus and exposure parameters.”31 On the far right, a bar code records “the identifying information for the image,” including the associated mission and the day and time of exposure.32

35Precisely because, for the scientist and engineer, they represent nothing more than a record of information, these image frames speak to an epistemology of knowledge production in which visual information is something that travels, disembodied, in a relay from format to format. They also demonstrate, however, that this same information is an indelible artifact of the technologies that created it. Thus if, on the one hand, the Surveyor television images could not be made useful until they could be made legible, then equally they could not be made legible unless the conditions of their production were brought into view. This is true whether the “perceiving” agent is a human or a machine.

Latent in the Archive

  • 33 Davis, “Surveyor Digitization Project.”
  • 34 Davis.

36Again, Surveyor, like so much of NASA’s activities at the time, was embroiled in the heat of the Space Race. This imperative to “put a man on the moon,” as it is so often put, created a need for speed. It also placed a heightened value on the kind of information that could help the organization prepare for those all-important Apollo missions to come, such that the processing of pictures of, for example, spacecrafts’ footpads, which apparently yielded information about how far they had sunk into the lunar soil, received priority over the more conventionally valuable scientific imagery and “pretty pictures.”33 The launch of NASA’s Apollo mission then quickly turned everyone’s heads.34 The bulk of the Surveyor archive was set aside, where it has been left almost entirely unanalyzed and unseen for a period of more than fifty years.

6. “Beautiful photo” of the Earth taken by Surveyor III during the total lunar eclipse, April 24, 1967

6. “Beautiful photo” of the Earth taken by Surveyor III during the total lunar eclipse, April 24, 1967

Houston, Lunar and Planetary Institute.

© NASA/JPL; © Lunar and Planetary Laboratory

  • 35 Jason Davis, “Surveyor Digitization Project Hints at Long-Lost Lunar Treasures,” The Planetary Soci (...)

37In 2014, the NASA-affiliated, University of Arizona–based Lunar and Planetary Laboratory (LPL) announced that it had attained a grant from the Lunar Advanced Science and Exploration Research program to digitize and catalog its impressive archive of 90,000 Surveyor images. The program’s purpose was to make these “long-lost lunar treasures” available to the public, complete with the calibration metadata, via publication on NASA’s Planetary Data System (PDS).35 The accomplishment of this goal would require the automation of further algorithmic interventions on a mass scale; ideally, of course, it would also invite further intervention at the hands of a larger scientific community. In other words, the algorithm enacts a process of information extraction, and this with a mind to enabling further processing of the Surveyor images.

  • 36 “Surveyor Digitization: Background,” Space Imagery Center, Lunar and Planetary Laboratory, Universi (...)

38The LPL’s project report justifies its own importance by pointing out that whereas “these image data and derived information are still being used to support lunar science and exploration,” less than 2 percent of that archive has ever been seen.36 This oversight is not simply a function of the historical context of the archive’s creation, which had so quickly drawn the attention of the scientific community on to the next dataset, but also has to do with the material conditions of the archive’s storage. At the time that the Surveyor program was in operation, the JPL was not in possession of the storage capacity that would have been needed to view its enhanced images effectively on a computer monitor, much less to keep them for posterity. Even those few computer-processed images that it did contrive to produce had to be converted back into print. This, then, is how they were stored for the decades to come: as film in canisters, stacked on shelves in a basement at the University of Arizona. This is a relatively stable means of storing the Surveyor images, but it also had the unfortunate effect of screening the visual data they contained from any ongoing activities of extraction and processing.

  • 37 Space Imagery Center, “Background.”

39When the LPL launched this digitization project in 2015, their first step was therefore straightforward: they began by scanning the archived image frames. Initially, the team projected that this stage of the project could be completed in a quick three-month spurt; and even though the process took twice as long as was anticipated, it was still completed over a six-month period in 2015.37 After that, the project’s technicians were faced with the more daunting task of figuring out how to manage—how, indeed, to process—the massive image cache that resulted from this.

40That is because, rich in information though they were, the 90,000 image files that the LPL now possessed could not be made useful until the team had devised an algorithm that could automate the process of converting the multiplicity of codes that had been inscribed onto the surface of the printed frames into a format compatible with the PDS database. This was a procedural matter, a system demand; without the attendant metadata, the files could not be uploaded to the data retrieval system. It was also, however, an epistemological one. There can be no making sense of the information contained by these image frames that is not buttressed by a knowledge of the means of their creation. This latest algorithmic intervention, in other words, extends the logic of the Surveyor imaging system once again, activating the visual information sitting latent within its archive precisely by making visible the parameters and structures of the system through which it was captured.

  • 38 Jason Davis, “New Lunar Mosaics Emerge as Surveyor Digitization Project Continues,” The Planetary S (...)
  • 39 Dr. Shane Byrne, personal communication to author, September 26, 2024.

41This proved to be a challenging task. As of May 2016, the team was “finalizing a computer algorithm to examine each individual frame and catalog the accompanying metadata—camera specifics like focal length, azimuth and elevation, and temperature—that will make the data set suitable for scientific use.”38 As of September 2024, the LPL hopes that they will be able to upload their data to the PDS in the spring of 2025.39

42Even without this outcome, however, all is not lost. The Surveyor digitization project was apparently stymied in its effort to make the program’s imaging system manifest on the platform of the PDS. It is my hope that this paper fulfills that purpose otherwise—that is, by making visible through critical, if not computational, means the algorithmic image’s historical implication within a diverse and sophisticated system of visual knowledge production.

Top of page

Notes

1 “Computer Process Brightens Surveyor Moon Pictures,” Jet Propulsion Laboratory, August 9, 1966, https://www.jpl.nasa.gov/news/computer-process-brightens-surveyor-moon-pictures.

2 Elizabeth A. Kessler, “Technology’s Palette: Voyager’s Eyes and the Hyperchromatic Enhancement of Jupiter and Saturn,” Technology and Culture 62, no. 4 (2021): 1099.

3 Jet Propulsion Laboratory, “Computer Process Brightens.”

4 Peter J. Westwick, “Visual Imagery in Solar System Exploration,” in Exploring the Solar System: The History and Science of Planetary Exploration, ed. Roger D. Launius (New York: Palgrave Macmillan, 2013), 152.

5 W. Patrick McCray, “How Astronomers Digitized the Sky,” Technology and Culture 55, no. 4 (2014): 935–36.

6 In addition to the work of Westwick, the histories written by Kessler and Batchen have been crucial to my analysis. Geoffrey Batchen, “Electricity Made Visible,” in Negative/Positive: A History of Photography (Abingdon, UK: Routledge, 2020), 115–27.

7 Billingsley, quoted in Kessler, “Technology’s Palette.”

8 R. Cargill Hall, Lunar Impact: A History of Project Ranger (Washington, DC: Scientific and Technical Information Office, NASA, 1977), 310.

9 Jet Propulsion Laboratory, “Computer Process Brightens.”

10 Raymond L. Heacock, “Lunar Photography: Techniques and Results,” Space Science Reviews 8, no. 2 (1968): 243 and 245.

11 Jet Propulsion Laboratory, Computer Process Brightens.”

12 Kessler, “Technology’s Palette,” 1099.

13 This phrase appears in several of the Surveyor mission television data reports. In the report for Surveyor I, it can be found on page 8. In subsequent reports, the page varies, but it is always the section on “camera calibration.” Stephen Z. Gunter et al., Surveyor I Mission Report: Part III, Television Data, Technical Report No. 32-1023 (Pasadena, CA: Jet Propulsion Laboratory, California Institute of Technology, 1967); Surveyor Project Science Staff, Surveyor III Mission Report: Part III, Television Data, Technical Report No. 32-1177 (Pasadena, CA: Jet Propulsion Laboratory, California Institute of Technology, 1968), 9; and Thomas H. Bird et al., Surveyor V Mission Report: Part III, Television Data, Technical Report No. 32-1246 (Pasadena, CA: Jet Propulsion Laboratory, California Institute of Technology, 1967), 9.

14 This phrase appears in every television data report. In addition to Gunter et al.’s Surveyor I report, Surveyor Project Science Staff’s Surveyor III report, and Bird et al.’s Surveyor V report, I have accessed the following: Thomas H. Bird et al., Surveyor VI Mission Report: Part III, Television Data, Technical Report No. 32-1262 (Pasadena, CA: Jet Propulsion Laboratory, California Institute of Technology, 1968); and Thomas H. Bird et al., Surveyor VII Mission Report: Part III, Television Data, Technical Report No. 32-1264 (Pasadena, CA: Jet Propulsion Laboratory, California Institute of Technology, 1968).

15 The Surveyor reports specify that the qualifier “processed” means “computer processed.” The report for Surveyor VII particularly notes that “processed, as used throughout, means digitally processed.” Bird et al., Surveyor VII Mission Report, 20.

16 Jason Davis, “Surveyor Digitization Project Will Bring Thousands of Unseen Lunar Images to Light,” The Planetary Society, October 24, 2014, https://www.planetary.org/articles/20141024-surveyor-digitization-project.

17 Tom Lund, “The Surveyor Lunar Landing Mission,” in Early Exploration of the Moon: Ranger to Apollo, Luna to Lunniy Korabl (Cham, CH: Springer Praxis Books, 2018), 89.

18 Klaus Brasch, “A Short History of Astrophotography: Part 2,” Journal of the Royal Astronomical Society of Canada 111, no. 6 (December 2017): 253.

19 Heacock, “Lunar Photography,” 223.

20 Lund, “Surveyor Lunar Landing Mission,” 89.

21 Bird et al., Surveyor VI Mission Report, 9.

22 Davis, “Surveyor Digitization Project.”

23 Heacock, “Lunar Photography,” 243.

24 Westwick, “Visual Imagery,” 151.

25 Bird et al., Surveyor VI Mission Report, 19.

26 The term “operations” is taken directly from the reports, where it basically describes image processing that is taken in real time, in service of the ongoing operations. The connection with the idea of the “operational image” is therefore indirect, but striking. Gunter et al., Surveyor I Mission Report, 35; Bird et al., Surveyor V Mission Report, 31; Bird et al., Surveyor VI Mission Report, 29; Bird et al., Surveyor VII Mission Report, 27.

27 Bird et al., Surveyor VII Mission Report, 27.

28 This wording is taken from the Surveyor VI report (p. 29). Descriptions of the mosaics appear, with differing degrees of specificity, in every one of the television reports.

29 Westwick, “Visual Imagery,” 153.

30 Davis, “Surveyor Digitization Project.”

31 “Surveyor Digitization: Decoding the Code,” Space Imagery Center, Lunar and Planetary Laboratory, University of Arizona, accessed September 8, 2024, https://sic.lpl.arizona.edu/surveyor-digitization/decoding-code.

32 Space Imagery Center, “Decoding the Code.”

33 Davis, “Surveyor Digitization Project.”

34 Davis.

35 Jason Davis, “Surveyor Digitization Project Hints at Long-Lost Lunar Treasures,” The Planetary Society, November 23, 2015, https://www.planetary.org/articles/20151123-surveyor-digitization-project-update.

36 “Surveyor Digitization: Background,” Space Imagery Center, Lunar and Planetary Laboratory, University of Arizona, accessed September 8, 2024, https://sic.lpl.arizona.edu/surveyor-digitization/background.

37 Space Imagery Center, “Background.”

38 Jason Davis, “New Lunar Mosaics Emerge as Surveyor Digitization Project Continues,” The Planetary Society, May 6, 2016, https://www.planetary.org/articles/20160506-surveyor-digitization-project-new-mosaics.

39 Dr. Shane Byrne, personal communication to author, September 26, 2024.

Top of page

List of illustrations

Title 1. Image of the Moon taken by the American space probe Ranger VII, July 31, 1964
Caption Greenbelt, NASA Space Science Data Coordinated Planetary Image Archives.
Credits © NASA/JPL
URL http://journals.openedition.org/transbordeur/docannexe/image/2502/img-1.jpg
File image/jpeg, 829k
Title 2. “Original” image taken by the Surveyor spacecraft and its “processed” version
Caption Illustrations taken from R. H. Steinbacher et al., “Surveyor I Mission Report”, part 3, “Television Data” (Pasadena, CA: Jet Propulsion Laboratory, 1966), 47 and 49.
Credits © NASA/JPL
URL http://journals.openedition.org/transbordeur/docannexe/image/2502/img-2.jpg
File image/jpeg, 379k
Title 3. “Model of the Surveyor VII spacecraft”
Caption Illustration from T. H. Bird et al., “Surveyor VII Mission Report”, part 3, “Television Data” (Pasadena: Jet Propulsion Laboratory, 1968), vi.
Credits © NASA/JPL
URL http://journals.openedition.org/transbordeur/docannexe/image/2502/img-3.jpg
File image/jpeg, 332k
Title 4. Mosaic made in part from “processed” photos of Surveyor I, November 11, 1966
Caption Pasadena, Jet Propulsion Laboratory Archives.
Credits © NASA/JPL; © Jet Propulsion LaboratoryArchives
URL http://journals.openedition.org/transbordeur/docannexe/image/2502/img-4.jpg
File image/jpeg, 455k
Title 5. Surveyor VII image, together with its metadata, 1968
Caption 70 mm film, Tucson, University of Arizona, Lunar and Planetary Laboratory.
Credits © NASA/JPL; © Jet Propulsion LaboratoryArchives
URL http://journals.openedition.org/transbordeur/docannexe/image/2502/img-5.jpg
File image/jpeg, 273k
Title 6. “Beautiful photo” of the Earth taken by Surveyor III during the total lunar eclipse, April 24, 1967
Caption Houston, Lunar and Planetary Institute.
Credits © NASA/JPL; © Lunar and Planetary Laboratory
URL http://journals.openedition.org/transbordeur/docannexe/image/2502/img-6.jpg
File image/jpeg, 111k
Top of page

References

Electronic reference

Frances Cullen, Clearing the Fog. Early Image Processing and the Lunar Surveyor Imaging System (1966‑68)Transbordeur [Online], 9 | 2025, Online since 26 February 2025, connection on 13 January 2026. URL: http://journals.openedition.org/transbordeur/2502; DOI: https://doi.org/10.4000/13dwz

Top of page

About the author

Frances Cullen

Frances Cullen is a historian of photography and a PhD candidate in McGill University’s Department of Art History and Communication Studies. Her doctoral thesis, “Real, Raw, and Revolutionary: Analog Photography’s True History in the Information Age,” traces the lineage of the term “analog” as a keyword for discussing histories of technological change in photography from the postwar period to today. She has published in Visual Studies and Philosophy of Photography.
Frances Cullen est historienne de la photographie et doctorante au département d’histoire de l’art et d’études de la communication de l’université McGill. Sa thèse de doctorat, intitulée Real, Raw, and Revolutionary. Analog Photography’s True History in the Information Age, retrace le parcours du terme « analogique » en tant que mot clé pour aborder l’histoire des changements technologiques dans le domaine de la photographie de l’après-guerre jusqu’à aujourd’hui. Elle a publié des articles dans les revues Visual Studies et Philosophy of Photography.

Top of page

Copyright

CC-BY-NC-ND-4.0

The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) may be subject to specific use terms.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search