Analysis of Lunar Orbiter Images Recovered From Analog Tape

D. R. Wingo1 and C. J. Byrne2, 1Skycorp Incorporated, P.O. Box 375 Moffett Field, CA,,
Submitted to 42nd Lunar and Planetary Science Conference
Introduction: The Lunar Orbiter Image Recovery Project (LOIRP) was founded in 2008 with funding from NASA ESMD to recover Lunar Orbiter images from the original 2″ analog magnetic tapes that had been held in protective storage by the National Archives and NASA for 40 years. Of the three central questions that had to be answered for project success, (can the tape drives be brought back to life, are the tapes any good, what is the quality of the data the best available), the final question, whether or not the analog image data on the tapes was superior in quality to the existing film was the ultimate criterion for success.

Continue reading “Analysis of Lunar Orbiter Images Recovered From Analog Tape”

More Boulders

This high resolution image, subframe 2128_H2, was taken by Lunar Orbiter 2 on 22 November 1966 at 20:18:27 GMT. Two areas containing craters filled with boulders have been highlighted. Enlarged versions of these locations are shown below. With a resolution of approximately 1 meter/pixel, the smallest boulders visible are several meters across.
LPI reference Images: [Medium] [Large]

Click on image to enlarge

Click on image to enlarge

Boulders in Lunar Orbiter 2 subframe 2111_H3

Click on image to enlarge.

This high resolution image, subframe 2111_H3, was taken by Lunar Orbiter 2 on 22 November 1966 at 13:49:11 GMT. As you can see from this full resolution enlargement, objects as small as a meter or two in size can be seen when the resolution of the original image is approx. 1 meter/pixel. The image below shows the location of this enlarged area.

Click on image to enlarge.

The Boulders of Copernicus

This image represents a portion of the central uplift within the crater Copernicus. The image, LOV-152-H1, was taken by Lunar Orbiter V on 16 August 1967 at an altitude of 103 km. The spacecraft was looking straight down at the crater as it snapped this picture series. The resolution of this image is 2.2 meters/pixel. [Click on image to enlarge]
Below, you can see the increase in contrast and resolution that LOIRP has attained when you compare the high resolution USGS image (left) and the one obtained by LOIRP on 10 December 2009 (right). [Click on image to enlarge]

Boulder Trails On The Moon

[Click on image to enlarge] Keith’s note: This image was taken on 21 November 1966 by Lunar Orbiter II at an altitude of 44 miles. The image is taken from frame 92, Framelet 445, and has resolution is 0.98 meters/pixel. As such the large boulder that has left a trail is around 6-7 meters in diameter. The image on the left shows the highest resolution image available online at LPI. On the right is the raw unproceesed image we retrieved this afternoon. While the large boulder’s trail is seen in both images, the details of that trail and the rest of the boulder field are much sharper in our newly retrieved image.
By coincidence this large boulder is very similar in size to “House Rock” a large boulder north of the Haughton Mars Research Station on Devon Island. That’s me standing on top if it.

This rock is a favorite place to pose for photos and was named after the large rock that Astronaut Jack Schmitt posed next to during the Apollo 17 mission.

Comparing USGS, LPI, and LOIRP Image Resolution (Update)

[Click to enlarge] Keith’s note: These images are taken from Lunar Orbiter II image LOII_092H1 Framelet 522. On the left is the highest resolution scanned version available online at LPI (or USGS). On the right is our partially processed version that we retrieved this morning. In addition to providing a much sharper image, note that our new image also allows contrast to be controlled such that features can seen in the areas that are darkened in the older image.
Right now we are focusing on retrieving Lunar Orbiter II images at the request of both LPI and USGS since high resolution (modern) scans have yet to be done for this mission’s images. We hope to have the entire image online later today complete with collapsed lava tubes (?) and giant boulders sitting in the middle of otherwise flat plains. With our recently enhanced (restored) original FR-900 tape drive and its recently restored sister drive we will soon have two fully functional drives in operation and will be able to (eventually) retrieve and release images on a daily basis.
I will be doing a live webcast tomorrow at 1pm PST wherein we give you a tour of our facility and watch as we pull a new image back from history and into the present. Details to follow.

Seeing Boulders on The Moon

[Click on image to enlarge] Keith’s note: Tonight we are testing out our newest Mac computer at the Lunar Orbiter Image Recovery Project located at NASA ARC. We’ll be using this machine (8 processors and 10 TB of storage) to do near-real time processing of imagery once we have pulled it off of original Lunar Orbiter analog data tapes using our restored FR-900 tape drives. We hope to do a live webcast this coming Thursday so that you can look over our shoulders as we bring another image to light for the first time in more than 40 years.
Tonight, as we were flying through a portion of one of the images we came across a boulder field. Here is the image archive at LPI – subframe H3, Framelet 323. The image was taken by Lunar Orbiter II on 20 Nov 1966 at an altitude of 52.2 miles with a ground resolution of 1.14 meters/pixel. The framelet image shown here is approximately 220 meters across. You can clearly make out a number of boulders around 1 meter in size sitting on the surface.

Newly Restored “Picture of the Century”: Lunar Orbiter 2’s View of Copernicus

The Lunar Orbiter Image Recovery Project (LOIRP) has released another iconic image taken during the Lunar Orbiter program in the 1960’s. This image, which shows the dramatic landscape within the crater Copernicus was often referred to as the “picture of the century” by many people at the time of its original public release in 1966.
This image was taken by the Lunar Orbiter 2 spacecraft at 7:05 p.m. EST on 24 November 1966 from an altitude of 28.4 miles above the lunar surface, 150 miles due south of Copernicus. At the time this image was originally released most views of the lunar surface involved looking straight down. Little, if any, sense of the true elevation of lunar surface features was usually available. This photo changed that perception by showing the Moon to be a world with tremendous topography – some of it Earth-like, much of it decidedly un-earth-like.
According to Wikipedia: “In 1966 the crater was photographed from an oblique angle by Lunar Orbiter 2 as one of 12 “housekeeping” pictures that were taken to advance the roll of film between possible astronaut landing sites being surveyed. At the time this detailed image of the lunar surface was termed by NASA Scientist Martin Swetnick and subsequently quoted by Time magazine as “one of the great pictures of the century.”
Time magazine said (“A New Look at Copernicus“): “Except for the black sky in the background, the photograph might have been mistaken for a composite of the scenic grandeur of Grand Canyon and the barren desolation of the Badlands of South Dakota. But when it was flashed unexpectedly onto a screen at a meeting of the American Institute of Aeronautics and Astronautics in Boston last week, sophisticated space scientists and engineers recognized the terrain immediately. It was a spectacular close up shot of lunar landscape. That photograph of the moon’s Crater of Copernicus, said NASA Scientist Martin Swetnick, is “one of the great pictures of the century.”
The following image is an interim version, with reprocessing and enhancements being made constantly. A larger, raw version (2.2 GB in size) is now online at NASA’s Lunar Science Institute. Larger view.

If you compare this new image with LPI’s high res version you can clearly see that this new image shows multiple striations in the surface, small boulders, landslides, shadows and a myriad of fine details simply not visible in the original. The LOIRP currently estimates that the resolution of this image is less than 1 meter/pixel. Larger view.

Dennis Wingo, co-lead of the LOIRP will make a presentation on this image and the LOIRP at the Lunar and Planetary Science Conference on Monday, 23 March 2009 at 8:30 am (session 102)
The LOIRP, funded by NASA’s Exploration Mission Systems Directorate and NASA’s Innovative Partnership’s Program, with support from Odyssey Moon, Skycorp Inc., SpaceRef Interactive Inc., ACES, and the NASA Lunar Science Institute, is housed at NASA Ames Research Center at Moffet Field, CA. This project has utilized original analog data tapes and restored tape drives to digitize original Lunar Orbiter project imagery. Utilizing computer technology unavailable at the time the images were originally taken, LOIRP has been able to produce images which greatly exceed the resolution of the images as they were first seen in 1966 and 1967.
The first image released by the LOIRP, the famous “earthrise image”, was made public in November 2008. It is anticipated with the release of this latest image of Copernicus, and further restoration of the original 40 year old hardware, that the pace of additional image releases will now increase.
Further information on the LOIRP can be found at
You can follow our image restoration process on Twitter at LunarOrbiter

Lunar Orbiter Image Recovery Project (LOIRP) Progress Report 19 February 2009

To: LOIRP Status
From: Dennis Wingo
Subject: Progress Report, Milestone ALERT! Folks Major milestones to report!
We have had a major milestone accomplished (well 98% of the way there). Figure 1 shows two framelets, from Lunar Orbiter II High Resolution image (we don’t know which one yet). Figure 2 shows our favorite image, the Lunar Orbiter 1 image of the Earth with two framelets that were captured from the undemodulated tape. The framelets that are stitched together are from LO-1-102.
This proves that we can demodulate random tapes from more than one Lunar Orbiter mission but it has also shown us what we have to do to get to full production mode. The demodulator must be ultra stable, however, the drive is still not to its original specification and so there is jitter in the signal. The new capstan motor should fix this as we have new belts, bearings, and the motor has been balanced back to its original specification. However, the the motor driver assembly that we built to use during the calibration process blew some power transistors and he had to come back down (he is in San Francisco) and pick up one of our spares. We have been building spares for the production process to keep the machines running and this is helping with the refurb process as well! We will have the capstan motor on Saturday and will install and align it on Monday.
Al Sturm is going back and doing some further refining of the demodulator design to account for some loss of demodulator lock on the signal found during the testing yesterday. This accounts for the striping that you see on the Lunar Orbiter II framelets that we captured and processed shown on the next pages. We have not found out what frame this is and it is possible (as I have looked through all of the LO-II images) that we have one that is not in the current database as one image is missing from the LPL database that may be ours. If this is true, then we are helping to fill out the complete LO image database for NASA. One final note, look at the final image and see the boulders sitting on the surface at the edge of the crater!
Gordon Woodcock has done some test automated assemblies of a couple of framelets which looks like it will result in a workable and at least semi-automated process for framelet reassembly.
Second Machine
We are continuing to install new caps and refurbished subassemblies that Ken and I have modified for better operation. It is our plan to finish the motor replacements, upgrades, and testing on the primary machine to get it ready for production before finishing up the second drive. This is as our experience gained on the first drive and the design upgrades will make the second drive relatively easy to get going after the learning curve of working on the first machine.
Figure 1 is an unknown high resolution image from Lunar Orbiter II. These are individual framelets and we are still searching to figure out which whole image that it is attached to. The lines through the images are artifacts from the demodulator that is not quite 100% working – yet. The image has boulders sitting on the surface quite easy to see. We have verified that the intrinsic resolution of these images is going to be very high, to be quantified further after we get the demod at 100% and the capstan motor replaced. Figure 2 is a composite from LO-I-102-H that we processed from the undemodulated tape.
Figure 1: Unidentified high resolution framelet from Lunar Orbiter II. Image LOIRP/NASA

Figure 2: composite from LO-I-102-H Larger image. Image LOIRP/NASA

We are closing in on satisfying all of the milestones from this phase of our project and we are looking forward to getting the go ahead and funding to enter the production phase. It is our plan to take a series of new images with us to the Lunar and Planetary Science conference in Houston to show what can be done with data mining these old data sets. We expect to be able to digitize any tape at will by next week and have a set of images that are the high priority outputs. I am extremely proud of our team and what has been accomplished and look forward to providing this raw data and finished products for NASA’s exploration and science efforts.
Respectfully Submitted
Dennis Ray Wingo

Lunar Orbiter Imagery and Apollo Landing Site Selection

Source: Apollo Expeditions to the Moon Chapter 5.5 – Mapping and Site Selection
Meanwhile the third member of the automated lunar exploration team had already completed its work. The fifth and last Lunar Orbiter had been launched on August 1, 1967, nearly half a year earlier. When JPL and Hughes began to experience difficulties with Surveyor development, and with the Centaur in deep trouble, NASA decided to back up the entire proaram with a different team and different hardware. The Surveyor Orbiter concept was scrapped, and NASA’s Langley Research Center was directed to plan and carry out a new Lunar Orbiter program, based on the less risky Atlas-Acena D launch vehicle. Langley prepared the necessary specifications and Boeing won the job. Boeing’s proposed design was beautifully straightforward except for one feature, the camera. Instead of being all-electronic as were prior space cameras, the Eastman Kodak camera for the Lunar Orbiter made use of 70-mm film developed on board the spacecraft and then optically scanned and telemetered to Earth. Low-speed film had to be used so as not to be fogged by space radiation. This in turn required the formidable added complexity of image-motion compensation during the instant of exposure. Theoretically, objects as small as three feet could be seen from 30 nautical miles above the surface. If all worked well, this system could provide the quality required for Apollo, but it was tricky, and it barely made it to the launch pad in time to avoid rescheduling.

A photo of the crater on the Moon,Tycho

The youngest big crater on the Moon is Tycho, which is about 53 miles across and nearly 3 miles deep. These Orbiter V photographs reveal its intricate structure. (Area in the rectangle above is pictured in higher resolution below.) A high central peak arises from the rough floor, and the crater wall has extensively slumped. The comparative scarcity of small craters within Tycho indicate its relatively recent origin. Flow features seen in both pictures could have been molten lava, volcanic debris, or fluidized impact-ejected material. Surveyor VII landed about 18 miles north of Tycho, in the area indicated by the white circle above. Enlargements of these pictures show an abundance of fissures and large fractured blocks, particularly near the uppermost wall scarp.

Continue reading “Lunar Orbiter Imagery and Apollo Landing Site Selection”