Working with Mars Reconnaissance Orbiter HiRISE Data

From Isis Workshop

[edit] About MRO HiRISE

The Mars Reconnaissance Orbiter Mission

[edit] Launch and introductory mission info

Mars Reconnaissance Orbiter Artist's concept. Credit: NASA/JPL

NASA's Mars Reconnaissance Orbiter, launched in 2005, is searching for evidence that water persisted on the surface of Mars for a long period of time. While other Mars missions have shown that water flowed across the surface in Mars' history, it remains a mystery whether water was ever around long enough to provide a habitat for life.

[edit] Science goals of the mission

Mars Reconnaissance Orbiter seeks to obtain science data that will provide researchers with information for locating safe landing sites for future missions and tell researchers about the history of water on Mars. The spacecraft instruments will zoom in for extreme close-up images of the Martian surface in order to analyze minerals, look for subsurface water, trace how much dust and water are distributed in the atmosphere, and monitor daily global weather.

These studies will help determine if there are deposits of minerals that form in water over long periods of time, detect any shorelines of ancient seas and lakes, and analyze deposits placed in layers over time by flowing water. It will also be able to tell if the underground ice discovered by the Mars Odyssey orbiter is the top layer of a deep ice deposit, or if it is a shallow layer in equilibrium with the current atmosphere and its seasonal cycle of water vapor.

The orbiter's primary mission ends about three Earth years after launch, in November 2008. For details on all mission stages, see the Mission Timeline (NASA).

[edit] Science Instruments

During its two-year primary science mission, the Mars Reconnaissance Orbiter will conduct eight different science investigations at Mars. The investigations are functionally divided into three purposes: global mapping, regional surveying, and high-resolution targeting of specific spots on the surface.

Mars Reconnaissance Orbiter Instruments Artist's concept showing spacecraft instruments monitoring water cycle on Mars. Credit: NASA/JPL

• HiRISE (High Resolution Imaging Science Experiment) This high-resolution, visible-range camera can reveal small objects in the debris blankets of mysterious gullies and details of geologic structure of canyons, craters, and layered deposits.
• CTX (Context Camera) This camera will provide wide area views to help provide a context for high-resolution analysis of key spots on Mars provided by HiRISE and CRISM.
• MARCI (Mars Color Imager) This weather camera will monitor clouds and dust storms.
• CRISM (Compact Reconnaissance Imaging Spectrometer for Mars) This spectrometer covers the range of visible and near-infrared light, useful for identifying minerals, especially those likely formed in the presence of water.
• MCS (Mars Climate Sounder) This atmospheric profiler will detect vertical variations of temperature, dust, and water vapor concentrations in the Martian atmosphere.
• SHARAD (Shallow Radar) This sounding radar will probe beneath the Martian surface to see if water ice is present at depths greater than one meter.

[edit] Instrument Overview

HiRISE Instrument A comparision between the resolution of a camera aboard Mars Gloabal Surveyor and the HiRISE camera on Mars Reconnaissance Orbiter. Credit: NASA/JPL

HiRISE will image hundreds of areas of Mars' surface in unprecedented detail.

HiRISE operates in visible wavelengths with a telescope that will produce images at resolutions never before possible in planetary exploration. These high resolution images will enable scientists to resolve 1-meter (about 3-foot) sized objects on Mars and to study the morphology (surface structure) in a much more comprehensive manner than ever before.

From an altitude of approximately 300 kilometers above Mars, HiRISE will return surface images comprised of pixels representing 30 centimeters of the martian surface.

These high-resolution images provide unprecedented views of layered materials, gullies, channels, and other science targets, as well as possible future landing sites

[edit] Technical Details

HiRISE Observation: This image is one half (vertically) of a HiRISE observation scaled down to approximately 1/50th of its original resolution. It is of a small area inside Eberswalde crater in Margaritifer Sinus. Taken on November 8, 2006, the image is a composite of all 10 red detectors and the 2 blue-green detectors.

Full Resolution Subarea: This is a full resolution sub-area of the image on the left (indicated by the red outline). One pixel represents 25.6 cm on the surface of Mars.

This telescopic camera has a primary mirror diameter of 50 centimeters and a field of view of 1.15°. At its focal plane, the instrument holds an array of 14 electronic detectors, each covered by a filter in one of three wavelength bands: 400 to 600 nanometers (blue-green), 550 to 850 nanometers (red), or 800 to 1000 nanometers (near-infrared). Ten red detectors are positioned in a line totaling 20,028 pixels across to cover the whole width of the field of view. Typical red images are 20,000 pixels wide by 40,000 lines high. Two each of the blue-green and near-infrared detectors lie across the central 20% of the field. Pixel size in images taken from an altitude of 300 kilometers will be 30 centimeters across, about a factor of two better than the highest-resolution down-track imaging possible from any earlier Mars orbiter and a factor of five better than any extended imaging to date. As a rule of thumb, at least three pixels are needed to show the shape of a feature, so the smallest resolvable features in the images will be about a meter across for an object with reasonable contrast to its surroundings. The instrument uses a technology called time delay integration to accomplish a high signal-to-noise ratio for unprecedented image quality.

The Principal Investigator (lead scientist) for HiRISE is Alfred McEwen from the Lunar and Planetary Laboratory at the University of Arizona.

[edit] References & Related Resources

• HiRISE instrument description
• HiRISE instrument technical description (PDF)
• HiRISE mission web site
• Mars Reconnaissance Orbiter mission web site
• en.wikipedia.org/wiki/HiRISE

[edit] Processing HiRISE Data

Focal Plane Assembly Each CCD has 2048 pixels in the crossscan direction The 14 staggered CCDs overlap by 48 pixels at each end. This provides an effective swath width of approximately 20,000 pixels for the red images and 4,048 pixels for the blue-green and near infrared images. Credit: NASA/UA/Ball

Constructing a map projected HiRISE observation image is much more difficult than most instruments. There are 14 Charge-Coupled Devices (CCD) in the camera, ten red, two infrared, and two blue-green. When the camera electronics read out the image data, each CCD is broken in two halves (left and right channels). On the ground, the data for one observation ends up stored in 28 image files (14 CCDs x 2 channels per CCD). The goal of this lesson is to show the necessary steps to combine the CCD images into a single mosaic.

The process of constructing an uncontrolled HiRISE observation mosaic includes the following steps:

Level 0 - Data Ingestion

• Acquire and convert HiRISE image data files to the Isis image format.
• Add information to the Isis image in order to compute geometric properties such as latitude/longitude range and illumination angles of the image.

Level 1 – Radiometric Calibration and Noise Removal

• Convert raw pixel numbers to reflectance (irradiance/solar flux or I/F).
• Remove vertical striping.
• Repair tonal differences between left and right channels.

Level 2 - Projection

• Geometrically transform each CCD image into a map projected image.
• Mosaic the CCD images together to create a single HiRISE observation image.

The results of this process will be an uncontrolled mosaic. When we have finalized the software and procedure for creating controlled mosaics, information and documentation will be released on our site.

[edit] Level 0 Processing - Data Acquisition & Ingestion

There are three major components of level zero processing which are:

1. Acquisition of raw HiRISE image data for an observation. The Planetary Data System (PDS) archives these raw images in a standard format called the Experiment Data Record (EDR). For a single HiRISE observation there may be up to 28 EDR files which must be downloaded for processing.
2. Ingestion of the PDS EDR formatted images. That is, the conversion of the 28 files from PDS EDR image format to the Isis image format.
3. Initialization of each Isis image with SPICE (Spacecraft and Planetary ephemeredes, Instrument C-matrix and Event) kernel data. Recall this information is used to compute geometric properties about the HiRISE observation, such as the latitude/longitude range or illumination angles.

[edit] Acquiring HiRISE Data

There are several resources and tools for searching, previewing, and acquiring MRO HiRISE image data. The HiRISE team produces two types of official Planetary Data System (PDS) data products from the HiRISE instrument:

• Experiment Data Record (EDR). HiRISE EDRs contain the data collected by a single channel of the twenty eight available on the instrument and transmitted back to earth by the MRO spacecraft.
• Reduced Data Record (RDR). HiRISE RDRs contain a radiometric calibrated, map projected, mosaic of multiple channels from a single observation. These products are stored in JPEG 2000 format.

The following sites are useful for viewing and/or acquiring HiRISE data:

PDS Planetary Image Atlas, managed by the Jet Propulsion Laboratory (JPL) and USGS Astrogeology Research Program, allows the user to search through the MRO database for HiRISE, CTX, and MARCI products, and identify images based on various parameters, such as orbit number, geographic location, and mission phase. This tool includes both basic and advanced search capabilities. The site offers both EDR and RDR data products.

The HiRISE PDS Data Node at the University of Arizona provides access to all publicly released HiRISE EDR and RDR data products. All data products can be accessed in the traditional PDS volume format on the site.

For viewing HiRISE RDRs, you can use the HiRISE IAS Viewer to view RDR images and save snapshots. The tool also provides some basic tools for viewing metadata, adjusting the contrast, and navigation the image. This is a good tool to use for browsing for images based on your region of interest.

HiRISE Online Image Viewer allows you to select observations from a global map, preview images online, and download RDR image data at various resolutions in JPEG or JPEG 2000 format. This is a good tool to use for browsing for observations in the PDS Data Node directory structure.

This lesson covers the process of constructing a mosaic from PDS EDR images. If you are only interested in obtaining the preprocessed and constructed data, acquire the PDS RDR products using the above resources.

These tools for viewing and accessing HiRISE data will be discussed further in this lesson. If you'd prefer to skip straight to processing the data, jump to Importing HiRISE Data (below).

[edit] Related Resources

• JPEG 2000 overview on Wikipedia

[edit] File Naming Convention

Deciphering the filename can be helpful both when searching for data and also when managing files on your system. The PDS naming convention for channel EDRs is

PPP_XXXXXX_YYYY_CCD_CHANNEL.IMG

Where:

• PPP is the mission phase
  • AEB Aerobraking
  • TRA Transition
  • PSP Primary Science Phase
  • REL Relay
  • Exx Extended missions (if needed)
• XXXXXX is the orbit number
• YYYY is the target code, which refers to the latitudinal position of the center of the planned observation relative to the start of the orbit. The first three digits are whole degrees, the fourth digit the fractional degrees rounded to 0.5 degrees. (example Mars target codes on a descending orbit: 0900 for 90.0° or south-pole, 1800 for 180.0° or equator, 2700 for 270.0° or north-pole)
• CCD is the filter/CCD identifier RED0-RED9, IR10, IR11, BG12, BG13
• CHANNEL is the channel number 0 or 1

For example, PSP_002733_1880_RED5_0.IMG is an image centered at 188.0° latitudinal degrees from the start of the orbit around Mars, collected by channel 0 of CCD 5, red filter, during orbit 2733 of the primary science phase. See the Software Interface Specification for HiRISE Experiment Data Record Products (PDF) for additional details.

Similarly, the PDS naming convention for RDR mosaicked observation products is

PPP_XXXXXX_YYYY.IMG

[edit] Planetary Image Atlas

Screenshot of the Image Atlas 'Quick Search' options for Mars Reconnaissance Orbiter

The PDS Planetary Image Atlas, managed by the Jet Propulsion Laboratory (JPL) and USGS Astrogeology Research Program, allows the user to search through the HiRISE database and identify images based on parameters entered by the user. This tool includes both basic and advanced search capabilities.

The Planetary Data System(PDS) Imaging Node houses data from several planetary missions, and offers a variety of methods for accessing their holdings. For now, we will concentrate on acquiring MRO HiRISE data.

[edit] Search

The PDS Planetary Image Atlas provides a Product Search tool to interrogate the collection of HiRISE images. This tool lets us query information about each image and ignores the data that we have no interest in. A good way to reduce the number of images to look at is by defining an area of interest with latitude and longitude ranges. You can also restrict the search by choosing a minimum and maximum resolution. Remember to keep in mind the coordinate system (areocentric east for HiRISE) and units (generally meters for distances) required by the search tool. Launch the Mars Reconnaissance Orbiter Product Search to give it a try.

The table below lists the primary search parameters that can help you narrow down the number of images that are returned by a search of the PDS HiRISE image collection. The images to the right show screenshots of the MRO Product Search. Note there are two categories (the tabs above the search form) where these search parameters are found: Quick Search and Instrument.

Once you've made your search parameter selections, click the Get Count to see how many results your search will return, or Get Results to perform the search and access the results.

Screenshot of the Image Atlas 'Quick Search' options for Mars Reconnaissance Orbiter

Screenshot of the Image Atlas 'Quick Search' options for Mars Reconnaissance Orbiter

HiRISE PDS Data Node

[edit] Browsing by Volume

You can also go to the online data and Browse Online Data Volumes, which offers FTP access to the image data archive. This allows you to look at the image and text files in the archive, where you can find more helpful information. To give it a try, launch the Planetary Image Atlas in a new browser window. Now click the Mars Reconnaissance Orbiter Browse Online Data Volumes and choose a volume to look at. In each volume, data are organized into subdirectories by product type, mission phase, and orbit. Currently, the HiRISE PDS Data Node holds the data archive.

When you know the images that you would like to work with, you can go to this area with an FTP tool (or web browser) and download the images. The data areas of the FTP site are briefly described below:

• RDR or EDR - for Isis processing, you'll want to browse the EDR directory
  • PSP, etc. is the mission phase - Primary Science Phase (PSP) data is most likely what you'll want to use.
    • ORB_[start orbit]_[end orbit] refers the orbit range. These directories contain the data files.

[edit] Related Resources

• Planetary Image Atlas
• Planetary Image Atlas Help Pages

[edit] HiRISE Online Image Viewer

The HiRISE Team has created a really nice web-based tool for viewing HiRISE images. This view can be useful for searching for and previewing observation images in your region of interest. The tool is available on the HiRISE Online Image Viewer web page.

Shown in the screenshots below, this tool provides a navigable, global overview map that displays the locations of observations as squares color-coded by the age of the observation. Clicking on one of the squares brings up that observation in an image viewer. Both the overview map and the image viewer have navigation tools that allow you to pan and zoom.

On the image viewer page, there are links to download small and large JPEG image files, and huge JPEG 2000 image files of the image you are currently viewing.

Below the image viewer, you will also find a context map indicating the location of the HiRISE image. The base of the context image can be chosen from Odyssey THEMIS, Viking MDIM, or MGS MOLA data.

Global Overview Map: Map displays locations of observations as colored squares. Use the pan and zoom tools to move around the map. Click on a square to view the observation of that location.

Image Viewer: Image viewer has zoom and pan tools for navigating the image.

Context Map: Below the image viewer is a context map displaying an outline of the current observation.

[edit] HiRISE Image Access Solutions Viewer

HiRISE IAS Viewer, Open Remote File Image Selection Dialog

The HiRISE Image Access Solutions Viewer (IAS) Viewer can be useful for browsing the archive of observation mosaics (RDR products) when searching for data in your region of interest. The viewer can be launched from the HiRISE JPEG 2000 Viewing Tools page.

From the File menu, choose Open Remote File... to open the image selection dialog. An image selection dialog, like the one shown on the right, will appear.

Type in the following address in the Server field

hijpip.hinet.lpl.arizona.edu:8064

and click the Connect to Server button.

Once the application has connected to the server, a navigable folder menu will appear on the left side of the window. Click the plus sign (+) beside each folder to expand the folder tree:

• PDS
  • RDR
    • PSP

HiRISE IAS Viewer

The PSP folder contains folders for each set of orbits, named ORB_[start orbit]_[end orbit]. Expand the folder for the orbit range you'd like, the select the folder for the image of choice. Click the image file (the file ending with the file extension JP2) from the right side of the screen, and click the Open button to view the image in the main window.

Selecting an image will give you a display like that shown to the right, and you will have many options for manipulating the image.

Note: only the HiRISE RDRs can be opened and viewed with the IAS viewer.

The JPEG 2000 files used by this viewer are accessible via the JPIP server hijpip.hinet.lpl.arizona.edu:8064. Additional tools and information about JPEG 2000 can be found in the Wikipedia JPEG 2000 entry.

The Image Access Solutions Viewer is a product of ITT Visual Information Solutions. (Any use of trade, product, or firm names in Isis web pages, documents, or publications is for descriptive purposes only and does not imply endorsement by the U.S. Government.)

[edit] Importing HiRISE Data

HiRISE image First 4000 lines of an 80000 line EDR taken during orbit number 2733. The image has been compressed four times in the line and sample directions.

In order to work with HiRISE data in Isis, the HiRISE EDR file must be converted to an Isis cube file so Isis programs can read and process the data.

EDR files should always have a file extension of IMG. These files contain the image data as well as text describing the image data and the state of the instrument at the time the image was taken. The text is in the form of a standard PDS label (click to view example label file) located at the beginning of the file. Only the information needed by other Isis programs is transferred from the PDS label to the Isis cube label (click to view example label file).

[edit] Ingesting HiRISE EDR image into Isis

The program used to convert HiRISE EDR files to Isis cube files is hi2isis. The following example shows the command line usage. The resulting output file will be an Isis cube.

Example: ingesting a HiRISE EDR product into Isis:

hi2isis from=PSP_002733_1880_RED4_0.IMG \
       to=PSP_002733_1880_RED4_0.cub

The hi2isis program also converts the image header, prefix and suffix data to Isis Binary Large OBject (BLOBs) and has other parameters.

[edit] Related Isis Applications

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• hi2isis: converts a HiRISE EDR to Isis cube format

[edit] Adding SPICE

An important capability of Isis is the ability to geometrically and photometrically characterize pixels in raw planetary instrument images. Information such as latitude, longitude, phase angle, incidence angle, emission angle, local solar time, sun azimuth, and a many other pixel characteristics can be computed.

To compute this information, the SPICE (Spacecraft and Planetary ephemeredes, Instrument C-matrix and Event kernel) kernels must first be determined for the particular raw instrument image. These kernels maintain the spacecraft position and orientation over time as well as the target position and instrument specific operating modes.

To add SPICE information to your cube, run spiceinit application on the image so that camera/instrument specific applications (e.g., cam2map, campt, qview) will have the information they need to work properly. Generally, you can simply run spiceinit with your input filename and no other parameters:

 spiceinit FROM=my.cub

[edit] Related Isis Applications

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• spiceinit: adds SPICE information to the input cube

[edit] Level 1 Processing

Radiometric Calibration, Stitching, and Noise Removal

In this section we discuss how to create a level 1 HiRISE image. The process of generating a level 1 image involves:

• Radiometric calibration of the data so we have an image representative of an ideal image acquired by a camera system with perfect radiometric properties. Values in the resulting image represent the reflectance of the surface (I/F).
• Removal of systematic noise from the image. For HiRISE, this noise appears as vertical striping, referred to as furrows, which occur under certain observing conditions, and tonal mismatches among the data sets collected by adjacent channels in a CCD.

A special requirement for HiRISE is the reconstruction of the CCD data. That is, merging the left and right channel data from an individual CCD into a single image.

[edit] Overview of Radiometric Calibration

[edit] Why perform radiometric calibration?

Both vidicon cameras (such as those carried on-board the Viking and Voyager missions), and charge coupled device (CCD) cameras (such as on the Clementine, Mars Reconnaissance Orbiter, and other contemporary missions) produce digital images with the inherent artifact known as camera shading. Camera shading results from the non-uniform sensitivity across the field-of-view of the imaging instrument.

Flat field This image (acquired during pre-flight calibration by a Mars Exploration Rover Microscopic Imager) illustrates camera shading. Ideally, every pixel in the image should have the same DN. Radiometric calibration corrects this type of non-uniform brightness.

Perhaps the best way to illustrate camera shading is to imagine acquiring a digital image of a target of uniform brightness, say a screen that has been painted a uniform shade of gray. If the camera sensitivity across the fields-of-view were ideal (and the flat-field target exactly the same brightness everywhere), then the acquired digital image would have the same DN value for all the pixels in the image. However, because of the non-uniform sensitivity of the camera, the DN values of the resulting image will vary throughout the image array (see the example to the right). A typical camera may have as much as 20% variation across the field-of-view. Camera shading corrections are applied to an image that correct for the non-uniform sensitivity so that, in our flat-field observation example, the radiometrically corrected image would contain pixels of identical value.

[edit] What is radiometric calibration?

Radiometric calibration recalculates the DNs in an image based on numerous factors, such as the exposure time, known values for the camera shading based on flat-field observations, dark current (output current of a detector when no energy is incident on the detector, such as when the shutter is closed), and other factors describing the unique electronics design and characteristics of an imaging system. Camera sensitivity may be time dependent because of the drift of the camera sensitivity throughout the course of the mission. The camera sensitivity is also dependent on the filter, operating modes of the instrument, and temperature of the cameras. Additionally, the camera response may be non-linear at various brightness levels.

[edit] How is the image changed by radiometric calibration?

A radiometrically calibrated image has DNs in radiometric units that are proportional to the brightness of a scene. Radiometric calibration applications in Isis produce output values that represent either:

• Radiance - The amount of electromagnetic energy emitted or reflected from an area of a planet, in units of µw/(cm2*sr)
• Reflectance - The ratio of reflected energy to incoming energy (i.e. irradiance/solar flux, often simply called I/F). A reflectance would be 1.0 for an ideal 100% reflector where the sun and camera orientations are perpendicular to the reflecting surface.

Generally, Isis radiometric calibration applications that offer both reflectance and radiance output will output reflectance by default. Due to the fact that radiometric calibration is mission dependent, each mission supported by Isis has its own radiometric calibration application. A few examples of radiometric calibration applications in Isis are:

• vikcal: Viking Orbiter Visual Imaging System
• moccal: Mars Global Surveyor Mars Orbiter Camera

[edit] Radiometric Calibration of HiRISE Data

Currently, the radiometric calibration of HiRISE images is a work in progress. The camera has 14 CCDs with separate readouts, and many different operating modes such as pixel binning and time delay integration. Understanding and solving the radiometric calibration for HiRISE is like solving for 28 individual cameras. The HiRISE Science Team is continually working on the calibration of their instrument, and updates to the Isis software will be made over time as the calibration sequence matures. For now, run hical on each Isis cube in the observation.

The following example shows the command line for calibrating the image from channel 0, red filter, CCD 5:

hical from=PSP_002733_1880_RED5_0.cub to=PSP_002733_1880_RED5_0.cal.cub

[edit] Related Isis Applications

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• hical: radiometrically calibrates HiRISE channel images

[edit] Merging Channel Images into a CCD Image

Recall the HiRISE instrument reads the data from one CCD into to separate channels. The next step in level 1 processing is to combine the two channel cubes back into an individual CCD image. The channels can be merged using the histitch program as follows:

histitch from1=PSP_00273_1880_RED5_0.cal.cub \
        from2=PSP_00273_1880_RED5_1.cal.cub \
        to=PSP_00273_1880_RED5.cal.cub

Left three images: Data from channels 0 (left) and 1 (center) of a red 5 image stitched together to create the full red 5 CCD image (right). The images shown here are scaled-down full HiRISE channel images. Right three images: Close-up on a portion of the images shown on the left. Data from channels 0 (left) and 1 (center) of a red 5 image stitched together to create the full red 5 CCD image (right). Note that in the final image, the data from channel 0 appears on the right, and the data channel 1 appears on the left.

[edit] Related Isis Applications

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• histitch: combines two HiRISE channels to form a single CCD image

[edit] Overview of Noise And Artifacts

Noise and artifact are terms used to describe speckles, spikes, reseaus, missing data, and other marks, blemishes, defects, and abnormalities in image data created during the acquisition, transmission, and processing of image data. The line between the definitions of artifact and noise is fuzzy (and often subject to opinion), and often the terms are used interchangeably. Some noise and artifacts are expected, even purposefully added, and can be removed during the radiometric calibration process.

[edit] What is noise?

In image processing, noise is a type of flaw or blemish in the image caused by:

• Telemetry data dropouts or transmission errors
• Malfunctioning or dead detectors
• Read noise native to the CCD system
• Coherent noise caused by spurious electronic signals from the operation of instruments onboard the spacecraft

Noise can take the appearance of speckling, missing data, random or orderly patterns, and other variations that cause the image to have a muddled appearance, or visually distracting blemishes or patterns. There are three categories of noise:

• Fixed-location noise always exist at the same location in the image array, with predictable positions. Fixed location noise can be cosmetically corrected by replacing the bad pixels with the weighted average of the unaffected neighborhood pixels. Fixed-location noise can result from malfunctioning or dead detectors.
• Randomly occurring noise results from data transmission errors causing data bits to be altered at random intervals in the image. The random noise produces discrete, isolated pixel variations or "spikes" and gives an image a "salt-and-pepper" appearance. Additionally, telemetry drop-outs can cause portions of an image to be completely missing. This type of noise is generally corrected using filtering techniques that recognize missing or anomalous data pixels and replaces these data points with a weighted average of the unaffected neighborhood pixels.
• Coherent noise can be introduced by spurious electronic signals produced by the operation of instruments onboard the spacecraft during image observations. The spurious signals interfere with the electronics of the imaging system causing coherent noise patterns to be electronically "added" to the images. For example, the shuttering electronics of the Viking cameras introduced a spurious "herring-bone" pattern at the top and bottom of the image. Noise-removal algorithms are designed to correct specific coherent noise problems such as this one.

[edit] What are artifacts?

Generally, image artifacts are a type of flaw or blemish in the image introduced during processing, intentionally introduced due to the design of system, or unintentional introduction of debris or energy external to the system. Examples of artifacts include:

• reseaus etched on the camera lens
• reseaus exposed on photographic film during pre-flight preparations for a mission
• minute dust specks located in the optical path or on the focal plane array
• cosmic rays and other charged particles impacting the sensor (particularly CCDs)
• fringe, ring, or visible patterns created during filtering, ratio analysis, and other enhancement processes
• quantization, checkerboarding, and other artifacts introduced by image compression algorithms during conversion from Isis cube format to a lossy image format or bit-type reductions that reduce the tonal resolution of the data

Most artifacts fit neatly into the categories of noise listed earlier and are corrected using many of the same processes. For example, dust specks create fixed-location blemishes, and cosmic rays cause random spikes. Reseaus are useful blemishes that are removed once they are analyzed for their locations within an image and the information saved for later processing.

Reseaus, dropped data, and salt-and-pepper This example shows a number of noise and artifact types. The large, regularly spaced black dots across the image are caused by the reseaus on the camera lens. The pattern of vertical black lines across the bottom of the image was caused by transmission data loss. The black and white speckles, called salt-and-pepper, are random noise.

Instrument error The black streak running down the right side of this image may have been caused by an unusual error in the line-scanner camera that acquired this image.

Transmission error A glitch during the transmission of this image caused the data to become garbled (upper right) and some data was completely lost (the black area across the middle)

Image compression Exporting the image to a lossy, compressed image format (using very high compression), the compression algorithm introduced a tiled pattern across the image.

[edit] Removing Vertical Striping and Channel Tone Differences

This step may be removed as the radiometric calibration matures.

The current radiometric calibration of the HiRISE data may reveal vertical striping noise in the CCD images. This is especially true for data collected by CCDs which have shown degradation in data collected over time (e.g., red 9). Additionally, tone differences caused by the the electronic read-out of the left/right channels may not be fully corrected by the calibration program. The cubenorm application can be used to remove both the striping and left/right normalization problems. The following example shows the command line for removing the noise and tone mismatch from the CCD image using cubenorm:

cubenorm from=PSP_002733_1880_RED5.cal.cub to=PSP_002733_1880_RED5.cal.norm.cub

Before cubenorm: This image depicts problems with both vertical striping (red arrows) and left/right channel tone problem.

After cubenorm: This image shows the results of the cubenorm application, which removes both problems.

[edit] Related Isis Applications

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• cubenorm: normalizes values in a image

[edit] Batch Processing

The ingestion, SPICE initialization, calibration, merging, and normalization must be run for each channel image. This would be incredibly tedious to run each application for every file! Luckily, batch processing is easy to do in Isis. Using the Isis batchlist command line option, a set of CCD images for a single observation can easily be processed through level 1 processing. The commands shown below create file lists to use as inputs to the applications and run those applications using the batchlist option.

ls *.IMG | sed s/.IMG// > cube.lis
hi2isis from=\$1.IMG to=\$1.cub -batchlist=cube.lis
spiceinit from=\$1.cub -batchlist=cube.lis
hical from=\$1.cub to=\$1.cal.cub -batchlist=cube.lis
ls *_0.IMG | sed s/_0.IMG// > cube2.lis
histitch from1=\$1_0 from2=\$1_1 to=\$1 -batchlist=cube2.lis
cubenorm from=\$1 to=\$1.norm.cub -batch=cube2.lis

[edit] Related Isis Applications & Documentation

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• Isis Command Line Usage: how to run Isis programs on the command line
• hi2isis: converts a HiRISE EDR to Isis cube format
• spiceinit: adds SPICE information to the input cube
• hical: radiometrically calibrates HiRISE channel images
• histitch: combines two HiRISE channel images to form a single CCD image
• cubenorm: normalizes values in a image

[edit] Level 2 Processing

Map Projection, Tone Matching, and Mosaicking

In this section, we'll discuss how to create a level 2 HiRISE image. At this point you should have 14 CCD images that are radiometrically calibrated and have had noise removed. This process involves the following:

• Geometric tranformation of CCD images from spacecraft camera orientation to a common map coordinate system
• Correction of tonal mismatches among the projected images. Prior to mosaicking the projected CCD images, we must fix tonal mismatches among the projected images. Currently, the radiometric calibration is not accurate enough to eliminate the tonal differences among the CCD images.
• Creation of a HiRISE observation mosaic from the tone matched CCD images to create the HiRISE observation image.

The results of this process will be an uncontrolled mosaic. When we have finalized the software and procedure for creating controlled mosaics, information and documentation will be released on our site. If you are unfamiliar with maps projecting images in Isis, review Learning About Map Projections.

[edit] Overview of Map Projecting Images

Converting a raw instrument image to a map projected image is a fundamental capability of Isis. The target definition, ground range, pixel resolution, map projection, and projection dependent parameters are all needed in order to produce a map projected image.

[edit] Projecting the Image

The cam2map application converts a camera (instrument) image to a map projected image. cam2map will automatically compute defaults for most of the mapping parameters, so you only need to define a subset of the parameters in your map template (e.g. ProjectionName).

• Your cube must be data from a Isis-supported mission -- cam2map depends on camera/instrument information in the Isis system to perform map projections.
• If you are projecting several images with the same projection parameters, you can reuse the same map template for all your images simply by removing the latitude longitude range parameters (MinimumLatitude, MaximumLatitude, MinimumLongitude, and MaximumLongitude) from your map template.
• cam2map will automatically calculate parameter values for you -- all you really need is the projection name in your map template.
• If you're planning on mosaicking your projected images, make sure the PixelResolution is the same for all images. Some projections also require the CenterLongitude and CenterLatitude to be the same when creating a mosaic.

[edit] Related Isis Applications

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• maptemplate: set up a map projection parameter template for map projecting cubes
• cam2map: project a cube to a map projection

[edit] Projecting HiRISE Images

To assemble our HiRISE observation mosaic from the CCD images, we will use the application cam2map to convert each CCD image to a map projected image. We'll show here how to easily match the projection information among the CCD images. Alternatively, you can define your own map file using the maptemplate application. The first step is to project one of the CCD images, usually red 5 as it is in the center of the observation:

cam2map from=PSP_002733_1880_RED5.norm.cub to=PSP_002733_1880_RED5.sinu.cub

The default projection for cam2map is sinusoidal, so the resulting red 5 CCD image is now in sinusoidal projection. As with many Isis applications, cam2map has several parameters, but the default values for these parameters will result in good projection results.

The output cube contains all the mapping parameters we need for projecting the remaining CCD images so they can be mosaicked. To process the remaining CCD images, use the following command:

cam2map from=PSP_002733_1880_?.norm.cub \
       to=PSP_002733_1880_?.sinu.cub \
       map=PSP_002733_1880_RED5.sinu.cub pixres=map

[edit] Quick Tip

We use the red 5 sinusoidal cube as our map file. The remaining CCD images will use the same projection parameters as red 5, which is required for mosaicking the CCD images into our final observation image.

[edit] Related Isis Applications

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• maptemplate: generates a map file that can be used to project images
• cam2map: geometrically transforms a raw camera image to a map projected image

[edit] Tone Matching and Mosaicking

Tone matching among the CCD images will be necessary until the radiometric calibration procedure matures. To illustrate the need for tone matching, the example images on the right show a mosaic with and without the CCD image normalization. Here, we will make a mosaic of the red CCD images. The infrared and blue-green CCD images follow a similar process.

In order to tone match the CCD images, we use the equalizer application program to normalize the images.

1. Hold the tone of the red 5 CCD image. That is, the pixel values in the red 5 CCD image will not be changed. Create a list containing the filenames of the images that will be held. In this example, the red 0 through 4 and 6 through 9 CCD images will be normalized to the the red 5 image:

ls *RED5.cub > hold.lis

2. Create a list containing the filenames of red 5 images to be normalized in preparation for crating the mosaic. The contrast and brightness of these images will be updated so that tones match the red 5 image listed in the hold list. We are effectively normalizing to the red 5 calibration. Note the equalizer program does a statistical analysis among only the pixels in the overlap regions to determine the corrections to the contrast and brightness differences.

ls *RED*.cub > redCCD.lis

3. Run the equalizer application using the two lists created above as input. equalizer automatically generates output filenames with the extension .equ.cub. The output results will be ten cubes (red CCD 0 through 9 images) with the extension .equ.cub added to each filename.

equalizer fromlist=redCCD.lis holdlist=hold.lis

4. Finally we will construct the mosaic. A new list of the equalized cube filenames is required as input to automos. Note: Make sure you only add filenames for one observation (or adjacent observations) to the list file, otherwise all the images in the list will be mosaicked into a single output image.

ls *RED*.equ.cub > mosaic.lis
automos fromlist=mosaic.lis to=redMosaic.cub

[edit] Related Isis Applications

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• equalizer: tone match overlapping map projected images
• automos: automatically mosaic a list of map projected images

Full view of the red 5 mosaic without equalization

Full view of the red 5 mosaic with equalization

Close-up of red 5 mosaic without equalization

Close-up of red 5 mosaic with equalization

[edit] Exporting Isis Data

Naturally, once you're finished processing your image data and you have your final product, you'll want to use it in reports, papers, posters, web pages, your favorite Geographic Information System (GIS) or image analysis package, or simply share it with others. Since many other software packages can't read Isis cube format, you'll want to export your cube to a file type appropriate to your needs.

[edit] General Purpose: Desktop Applications and Web

isis2std (Isis to Standard formats) exports your cubes to a wide variety of popular image formats, such as PNG (Portable Network Graphics) and TIFF (Tagged Image File Format). No matter what desktop application you're using (Word Perfect, Illustrator, or GIMP, just to name a few), or what you're creating (posters, web pages, or a pretty background for your desktop), isis2std will handle it.

[edit] But There's More! GIS and Image Analysis Applications

But the icing on the cake is that isis2std also exports the necessary files for taking your image into GIS! If your cube has Mapping labels, isis2std will write a world file that provides GIS software, and other software which can take advantage of a geographically registered image, with the necessary information to properly display your image. Most applications designed to work with remotely sensed and map data (such as Arc products, Envi, and Global Mapper) can use TIFF, JPEG, and other images with world files. Several geographic software applications even have support for planetary data.

[edit] Astronomy Anyone?

isis2fits (Isis to FITS format) exports your cubes to Flexible Image Transport System (FITS) format, a standard in the astronomical community. There's lots of software designed to work with FITS (even a few common applications and software libraries , like ImageMagick and GIMP). In general, FITS format is useful for mission team members who are working with cruise data to analyze instrument calibration, camera pointing, and other factors. Since cruise image data frequently contain stars, astronomy software comes in quite useful for this type of work.

[edit] Just the Data, Ma'am

isis2raw (Isis to Raw format) exports an individual band in your cube to raw image format, a minimal format that contains just the data in the bit-type of your choosing. isis2raw is a good choice when none of the other formats Isis can export to will work for you. In order to import a raw image into another application, you will need to know the width (samples), height (rows), bit-type, and endianess of your raw file in order to import it correctly, so remember to keep track of the parameters you used to export your cube. Related Isis Applications

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• isis2std: Export a cube to popular image formats, such as TIFF and PNG, with GIS world files
• isis2fits: Export a cube to FITS format, the standard image format of the astronomical community
• isis2raw: Export a cube to raw format

[edit] Special Considerations: Exporting HiRISE images

HiRISE mosaics are very large and our export application isis2std does not handle large PNG or JPEG images. You will need to decrease the size of the image to export by either cropping or reducing.

[edit] Anaglyphs

See: HiRISE Anaglyphs

[edit] Related Isis Applications

See the following Isis documentation for information about the applications you will need to use to perform this procedure:

• reduce: scale down a cube
• crop: pull a region out of a cube