GeoTIFF is an industry standard for embedding geographic information in images. Beginning soon, HiRISE RDRs will include GeoTIFF info in the Jpeg-2000 files. All of the information about the image will continue to be in the RDR label (.LBL plain text file), but with this additional info in the JP2, image viewing software that supports GeoTIFF will be able to take advantage of it.

For example, such software could display the actual coordinates on Mars of the pixels you are looking at, allow you to measure features directly in physical units, or stitch together images based on their absolute location on the planet. A number of GIS (Geographical Information Systems) applications use GeoTIFF; many on our science team have been waiting patiently for this feature to be rolled out.

We have already begun to produce RDRs with GeoTIFF, and they will start appearing in our weekly releases. At some point, a major reprocessing effort will be underway to bring this feature (and others) to all of our pre-existing products.

This brings up the topic of versioning: namely, how to tell which version of a HiRISE product you are working with.

(more…)

Here are 40 RGB color images from the 1500 – 1600 orbit range of MRO.

View Images

There are, as always, many magnificent images here. Some of the noteworthy observations are:

PSP_001521_2025 and PSP_001501_2280: On the HiRISE web site you can see diagrams made by Tim Parker show the locations of various parts (lander, backshell, heatshield or parachute) for Viking Lander 1 and Viking Lander 2. It’s possible they aren’t in the color strip (I haven’t found them)!

PSP_001508_1245 and PSP_001510_2195: These two exhibit a “glow” pattern of saturated pixels due to high TDI (Time Delay Integration) settings on the blue-green CCDs. (All of the exposure settings are chosen for each observation based on a photometric model of the scene).

PSP_001538_2035: This is a rim-to-rim section across a crater called Tooting that is about 30 kilometers in diameter. It’s also interesting to note how the altitude of the rims, when combined with the large off-nadir roll angle (23 degrees), leads to an oddly bowed geometric projection. But it is correct; as the terrain rose, fell, and rose again from HiRISE’s angled point of view, the center of the ground track deviated slightly east or west from a true great-circle line.

PSP_001558_1325 and PSP_001593_2635: These dune fields are striking, forming incredible patterns.

PSP_001582_2245: Looking like a super-sized area of dried mud, the polygonal cracks in this image are amazing.

Updated (2008-Apr-10)

Starting with the 10/10 release, color images are included for the first time. We’ll describe how we process these in the days and weeks to come. But what I’d like to do first is give a brief description of all our product types as they currently are available. You’ve no doubt noticed a mind-boggling array of new options on our product pages. They now include what we call our “NOMAP” products; NOMAP means that they are not map-projected. In other words, not rotated to the direction of north, not mapped to a coordinate system, and not scaled to any particular geometric resolution.

I’ve prepared this ugly table that outlines each of the products now available (excluding the raw EDRs). So reading the columns from left to right: there are three types of “NOMAP” products, two types of lossy “QLOOK” (Quicklook) RDRs, and two types of lossless RDRs.

With that as a reference, now I’ll try to define everything more precisely.

“NOMAP”

Non map-projected product. Always lossy compressed for smaller size and quicker viewing. These are not formal Planetary Data System products; they’re “special”, meaning there is no PDS label and no Software Interface Specification describing them. Available for IRB, RGB and RED.

RDR

Reduced Data Record: reduced in the sense of refined or processed, not raw data. Formal PDS products with accompanying labels and a detailed SIS document describing their format and processing steps. Available both in lossless and quicklook formats for both RED & COLOR.

“QLOOK”

Quicklook: a special product that is a lossy compressed version of the RDR. In a normal RDR, all of the original data is retained. But with a quicklook, some of the highest resolution detail is discarded to make for quicker viewing.

RED

The image obtained by the red-filtered CCDs. It will be over the full swath width, typically data from all ten red CCDs. Covers the visible wavelength band from 550 to 850 nanometers.

IR

Infrared. Covers the near-IR wavelengths from 800-1000 nanometers.

BG

Blue-Green, visible wavelengths from 400-600 nm.

COLOR

A color RDR. It contains data from the IR, BG and center RED ccds. Typically this will be a skinny strip (”center swath”) inside a skinny strip, or as I like to say, the bacon-strip effect.

IRB

An enhanced color NOMAP. It has the same color bands as the RDR: IR, RED and BG.

RGB

An enhanced color NOMAP. It contains only data from the RED and BG. The blue is derived from the difference between the RED and BG. The color bands are RED, BG and the synthetic blue.

EDR

Experiment Data Record, a formal PDS product that is raw uncompressed data with a label header.

Note: we will be working towards making all of these products available for all prior releases.

Over at NASA Ames, the HiRISE Clickworkers program is in beta-testing. Anyone (this means you!) with a browser and a net connection can participate in the cataloging, or more precisely, keywording of HiRISE images.

This is an ambitious effort. Originally (years before HiRISE), Clickworkers was used to tag craters on Mars, helping pin down the relative ages of various regions. This time around, you identify a dozen or so possible feature types, then move on to the next image. So you have to be a little more discerning, though examples are provided.

I was just looking at the sizes of our images to date. We’re coming up on one thousand images that have been map projected. And it looks like we just recently passed the one million megapixel mark (one thousand gigapixels, or one terapixel!) in the geometrically projected ones (when rotated so that North is up, there tends to be a lot of empty pixels framing the images).

Assuming a standard screen size of 1.25 megapixels (1280×1024), that is 800,000 screenfuls. If you looked at one per second, it would take you almost ten days to view it all! But one thousand volunteers could get through it in a day, and spend 100 seconds per image, which seems reasonable. [Though of course they'll need time for sleep, etc!]

The idea of using human brain power as a sort of massively distributed computation engine (shades of The Matrix) has come a long way. Amazon’s Mechanical Turk pays volunteers for tasks such as identifying features, translating documents or answering questions. It was recently used in the search for a person (computer scientist Jim Gray) missing at sea. Volunteers viewed over a half million images, covering 3,500 square miles of ocean, though unfortunately his sailboat did not turn up.

Still, ‘crowdsourcing‘ (as Wired called it) seems like it will continue to be an efficient way to perform tasks that computers are currently very poor at. Here at the Lunar and Planetary Lab, it has also been used by Spacewatch to find Earth-approaching asteroids. So, essentially, you could help save the planet in a real-life version of the classic game Asteroids! Clickworkers also has a program where you can tag Mars Global Surveyor images, scouting interesting locations for HiRISE to target.

We can’t let the machines have all the fun!

To conclude our exploration of the pipelines that take raw channel files and create a beautiful, unmapped mosaic, let me introduce the Stitch pipelines: HiStitch and HiccdStitch.

The HiStitch pipeline combines the matching HiCal products for the same CCD into one more-or-less lined up CCD cube file. HiccdStitch combines these HiStitch cubes into RED, IR, and BG mosaics.

Both pipelines take some time, as overlapping pixels are accounted for and brought together. After these mosaics are created, additional steps create smaller jpeg files for easier viewing, and full-sized jpeg2000 files. We use these jpeg2000 files for validating our images.

There are later pipelines, but we first validate the HiccdStitch products: Did the previous pipelines work correctly? Did the uplink team command the camera correctly? Is there haze or clouds obscuring our view of the surface?

If everything looks good, and we have received the correct reconstructed SPICE ephemeris data, then the geometry pipelines are invoked. These pipelines project the images mathematically to a model of Mars and add geometry data to the images so that each pixel becomes a point on Mars with latitude and longitude coordinates.

The first HiRISE image data of the Primary Science Phase (PSP) arrived in Tucson last night sometime around 9 PM. Although we thought the first data might not arrive until early this morning, I was a little antsy and took a look from home around 9:40 PM to see a complete first observation ready for validation.

We are waiting for reconstructed SPICE ephemeris data, which comes out every Wednesday – starting next week – before sending these data through our geometry pipelines, and ultimately releasing them to the scientific community and public. Last time, we forced images through our geometry pipelines using predicted SPICE kernels; we do not want to double our workload by continuing that practice. The SPICE kernels released next Wednesday will cover some of the images captured this week.

Once the images have been visually and statistically validated and the matching SPICE kernels have arrived, one of the downlink folks will send the images through the geometry pipelines. We also need to get a select group of captions written and automatic caption information generated for the rest.

We are producing JPEG2000 products now in addition to smaller jpeg browse images, to be ready for our viewing client when it is ready for public release. However, there are many different JPEG2000 viewers and plugins already out there to start practicing with. One example is ExpressView from LizardTech.

Once we are on a roll, the data release will be steady and no one will be able to keep up with the wealth of Mars data coming in. Until the first public release of PSP images, we will try to provide here on HiBlog more details about the many tasks that must still be completed.

I decided the blog does not have enough pictures, so a few of us gathered around a MacBook Pro and said “Cheetos!”. Audrie is on the left, I’m next, Kite is next to me, and Tahirih is on the right. Yes, Kite has Princess Leia hair. No, I’m not a nerf herder. Who’s scruffy-looking?

Audrie, Tahirih and I did not previously appear in pictures on HiBlog because during transition imaging we were busy working in our offices and Tuvas for some reason did not visit us. We feel so left out (joking)! The three of us make up HiRISE Downlink Operations, which includes downloading new images, processing them, and image validation (the Student Validators also participate in this task). Audrie also works on instrument monitoring and safety. Tahirih also does most of the geometry processing. I also eat cheetos and chocolate cake. When Kite is not busy with HiRISE Uplink tasks – which is generally NEVER – she is blasting her way out of impossible situations that often involve walking carpets.

Some of you out there may be asking: what happens to a HiRISE image between the time that it is taken and the time that it is released to the public? Well, I’d like to give a summary here.

(more…)