Gnome CCD Camera Imaging for Linux

Version 0.4 Last modified 5/9/2001


Table of Contents:


Introduction:

The Gnome CCD Camera Imaging application, gccd, is a basic camera,  telescope, and filter wheel controller.  It is not meant to be an astronomical  image processing application.  However, over time it might start to include more and more image processing functions.  In order to make the best images, gccd can control telescope slewing and filter wheels for color imaging.  One-shot color cameras are supported by allowing the full color frame to be split into individual color frames.  More than one CCD camera can be controlled to allow for guiding using one camera while integrating on another.  The Starlight Xpress MX series of cameras can self-guide so that guiding and integration can occur on the same camera.  Multiple exposures can be combined using a number of arithmetic operationsto create calibation frames.  By enabling auto-registration, multiple short exposures can be added to simulate a longer single integration.  The one-shot color cameras will split its colors before registration.

Web Site:

For the latest news and updates: http://schmenk.is-a-geek.com

Feedback:

Send me mail: astroccd@sbcglobal.net

Main Image View:

The main view of gccd is a MDI window that can display multiple images  in either a single notebook view or multiple top level windows, one foreach image.  Images are manipulated, opened, and stored from the mainview. 

Load Image:

FITS format is the only graphics files that can be accessed through gccd .  Any FITS format file can be loaded.

Acquire Image:

Acquiring an image brings up the CCD control panel.  Look in the Acquire  Image Control section for a full description of this function.

Image Properties:

Image properties are basically the FITS file header records.  Some are static and cannot be edited.  Others are fields used to identify the image and other pertinent information.  The NAME field is the base filename without any directory path or extension.  Use this to help identifythe image with a glance.  It must also be unique among the loaded images.  The static fields define the format and parameters of the exposure.

Save Image:

FITS format is the only graphics file format that images can be saved with.   The default extension is .fits but this can be changed using 'Save As'.   The FITS header can be edited with the ' ImageProperties' menu item.

Working Directory:

The working directory is the default directory used to load and save images.   It can be set through the Preferences drop-down menu.  Once set, itis retained accross invocations.

COM Port Settings:

Both the telescope slewing control and the filter wheel control require access  to a serial port.  If 'None' is selected, that device will not be enabled  in the CCD Control panel.  Since gccd does not run with root priveledges, the serial port device files, /dev/ttyS*, need to allow access to regular user applications.  Use the chmod program to alter the priveledge access bits to the serial device files.

View Multiple Document Interface Settings:

The Gnome MDI window can be viewed in three diffent modes.  Notebook, toplevel, and modal.  Notebook and modal only display one image at a time in a single main window.  The notebook view has tabs along oneside that can be clicked to bring any image to the front.  Modal doesn't have tab settings but the 'Window' menu has the list of all images loadedand can be selected with the menu items.  The toplevel view gives each image its own window.  This can display all the images at once, but requires lots of screen real-estate.  The default view mode is defined in the Gnome Control Panel.  The view mode will remain persistent accross invocations once set, regardless of the Gnome Control Panel.

View Image Colorized:

One-shot color cameras and monochrome images taken through filters can be  displayed using the color information recorded from the camera and/or filter  wheel.  The colorized image is only an aid in seeing the color in the image because it hasn't been calibrated.  An image processing program is needed to adjust the color balance.

View Image Contrast Stretched:

Many images don't use the full dynamic range of the DAC so they can be very  dark.  The contrast stretch setting simply does a linear stretch of the images' min and max values to the display.  This does not change the image data in any way, it is simply a visual aid.


View Image Aspect Corrected:

This displays the downloaded image with a 1:1 aspect ratio.  The pixel size is used to adjust the displayed image.  It does not affect theimage in memory or when it is saved.

Flip Image:

An image can be flipped in the horizontal or verical direction.  This in normally done to re-orient the image because of an optical path that flips  the image.

Invert Image:

An image can be inverted, basically making a negative of the original.  Some astronomers find this a better way to view faint images.  This can be done twice to restore the original image.

Remove Background:

In the event of much light pollution or when taking stacked images through a regular SLR camera lens, there can be a lot of background accumulation.  This option finds the most common pixel value (usually the background) and subtracts 95% of this value from the image.  It removes most of the background contribution while leaving enough for faint features.

Remove Noise:

When using uncooled cameras or making a very long dark frame, unwanted noise can creep into the image.  This option removes spurious pixel values while leaving most of the image untouched.  Make sure to save the image before applying this option to make sure undersamples features aren't removed as noise.

Splitting One-ShotColor Frames into Individual Color Frames:

One-shot color cameras achieve a full color image with a single integration  because of a color matrix overlaying the pixel grid of the CCD.  Individual  color frames can be synthesized by analyzing the color matrix and copying  the pixels of each unique color to a new frame.  Colors that aren't represented because of the matrix ordering are created from an average of their nearest neighbors that match the color.  The recombination of the color frames and their calibration is left to the image processing program.


Acquire Image Control:

The Acquisition Control contains all the options for imaging.  It has three main control areas.  The main exposure options are available at the bottom of the panel.  Exposure duration and camera selection aswell as status are always visible.  On the right is the telescope control panel.  It is always available for moving and slewing the telescope.  Options for inverting the right ascension and declination can be set to match the image view.  Finally, the notebook view contains all the advanced options for integration, guiding, and filter wheel control.

Select Camera:

All identified cameras are avialable in a drop down list.  The displayed  camera is what will be used for integrating.

Exposure Duration:

The exposure duration can be selected using three units: milliseconds, seconds,  and minutes.  This allows for integrations from instantaneous to many hours.

Exposure Begin/Cancel:

Exposures are started with the main Begin button at the bottom of the panel.   Once the integration begins, the button label changes to Cancel.  The button can be pressed again to abort the exposure.  If the Find &Focuspanel is active, the integrated frame will be sent tothe find/focus window instead of being saved.

Exposure Option:

Options for the integrated image

Binning:

The image can be binned vertically and/or horizontally at 1X, 2X and 4X.  Depending on the camera, the binning will occur inside the CCD chip or digitally  as the image is read.  One special bin mode for thevertical direction is the 1/2 X bin.  This is a bin mode for the StartlightXpress MX series of cameras that use the Sony interline CCDs.  This mode reads out the even and odd fields seperately and interleaves them for double the vertical resolution.  All other cameras treat the 1/2 X bin identically to 1X bin.

Calibration:

Calibration can be carried out using bias, dark, and flat frames.  Bias frames are zero length integrations used to subtract out DC bias and read noise.  A standard dark frame, one that includes bias and has the same exposure time as the selected integration can be used in the bias frame option.   Scaled dark frames can be used as generic darks.  They should be exposed  for a least twice as long as the intended calibrated integration. Flatfields are used to counteract effects of dust in the optical path and vignetting.  If using the filter wheel flat field, this option should be left as (None) otherwise it will override the filterflat field.

Base Name & Number:

The base name and starting sequence number of acquired images can be sethere.  If you are going to shoot a sequence of images of M42, you mightenter M42 as the base, and 1 for the starting sequence number.  Thesequence number will increment after every exposure to keep the image names unique.

Multiple Exposures:

Many back to back integrations can be made for various reasons.  The images can be individually sent to the Main View for capturing good seeing for planetary images  or combined to improve a deep sky image.
Frame Combine Modes:
The combine modes allow most common arithmatic frame combinations.  Summing,  averaging, median, minimum, and maximum operations are supported.
Auto-Registration:
The individual frames can be registered to the first image frame so all the  exposed frames can be combined.  If frames are kept seperate, theywill still be registered before being sent to the Main View.  When registering with a one-shot color camera, the combined color frame will be split into individual color frames before the registration takes place.  This keeps the color matrix information from being corrupted since registration will shift each frame by a sub-pixel amounts.

Find &Focus:

Achieving good focus is actually harder than it should be.  To help, the Find &Focus pane tries to make image downloads as short as possible  so the image frame rate allows real-time focus control.  The same features  make locating and centering objects easer as well.  Some cameras can also reduce the DAC precision to improve download time as well.

Image Scale:

By using the binning and windowing features of the CCD camera, the integrated  image can be quickly downloaded.  The image scale can be set from 4:1 to 1:4.  Zooming out to 4:1 uses binning to reduce download time and increase camera sensitivity.  This scale helps achieve gross focus.  By moving to the next scale level and adjusting focus along the way, good focus can be quickly achieved.  The focus image will be contrast stretched to help find the sharpest image.

Image Histogram:

The histogram of the image will help identify when best focus is achieved.   Minimum and maximum pixel values can be used as a focus quality measurement.   Because the focus pane automatically applies a contrast stretch, it cannot  be used to measure the relative brightness of the image.  Use the histogram  for quantitative focus feedback.

Guide:

Guiding is used for improving the tracking of your telescope.  Normally this is applied in conjunction with a telescope interface to make real-time adjustments during an exposure.  A seperate camera is required for the guiding unless you have one of the Starlight Xpress MX series of cameras.  These cameras have the ability to expose one field (odd or even) while guiding with the other.  GCCD was designed with this feature in mind.

Select Guide Camera:

This is the camera to be used for guiding.  Unless you have a Starlight Xpress MX series camera, it cannot be the same as the exposing camera.  One special note:  If you only want to autoguide for regular astrophotography or to just keep a star centered for visual tracking and you have only one CCD camera, load the dummy CCD camera tester.  Set your real camera for guiding and tester for integrating.  Set the exposurefor as long as you want.  You will not get a real image, of course,but the CCD camera will guide the telescope for you.

Self Guide Fields:

For use with the Starlight Xpress MX series of cameras.  Lets you select how to merge the even/odd fields of the CCD after guiding.  The last option, "Ignore Errors", is used for taking dark frames.  Normally the guiding algorithm will issue errors if it loses the guide star.  This skips the errors and continues with the integration.

Select Guide Object:

This allows for manual selection of a guiding object.  Usually you can let the software choose the brightest object, but this might not be desireable for tracking comets and other non-star objects.  You also need to manually select a guide object before training.

Clear Guide Object:

Use this if you manually move the scope to a different location.  If you move the scope through the scope control panel, the guide object will be automatically cleared.

Tracking Rates:

These are the rates the tracking routine uses to move the telescope to keep the guide object centered.  Usually they are automatically calibrated through training although they can be manually altered.

Train Tracking:

This is how the tracking rates are calibrated.  You must select a bright guide object that is fairly isolated before training.  The guide object should be close to the object you plan on imaging.  Tracking rates can be dependent upon where in the sky the telescope is pointing.  You may need to retrain if the telescope is slewed a great deal.  If you get a message saying the guide star was lost while training, adjust the training duration here.

Filter Wheel:

The filter wheel is used for color imaging.  Its use is fully integrated into the exposure process.  By selecting the "Auto-Sequence" option, all subsequent exposures will cycle through the included filters.  All exposures, including multiple, stacked, and otherwise will be shot through one filter before moving to the next filter where all the exposures will be repeated.  By selecting a current filter without "Auto-Sequence", all exposures will be shot through the current filter only.  Many filter wheels have a blank filter slot used for dark fields.

Color Selection:

The color selection allows for the proper processing of the exposed image to take place.  Color infromation is saved in the image and in the image name to help maintian records of the exposure.

Sequence Inclusion:

Only filters selected for sequence inclusion will be imaged through when "Auto-Sequence" is selected.

Exposure Percentage:

When auto-sequencing, some filter/CCD combinations may be more sensitive than others.  This setting allows for longer or shorter exposures through different filters.  The percentage is multiplied by the global exposure duration to calculate the filtered exposure length.

Flat Field:

Each filter should have its own flat field.  In order to select a flat field for each filter, make sure the "Flat Field" option in the exposure options is set to "(None)", otherwise it will override the filter flat field.

Miscellaneous Values:

These are values used internally to many of the algorithms in gccd .  Most will never need to be modified, but in case the algorithms fail, changing some of these values can get it to work for you situation.  Usually the field of view affects the algorithms more than anything.

Telescope Control:

Controlling the telescope is required for the auto-guiding feature to work properly.  The telescope control can be controlled through the internal routines of gccd or through mouse clicking on the arrows in the panel.   The arrows always reflect the state of the telescope control.

Slewing:

Slewing is useful for quickly moving the telescope.  Using the STAR2000 interface, slewing is done through moving in the desired direction, and then selecting the reverse direction as well.  This is a feature of the telescope mount and not of gccd.  It is known to work with Losmandy mounts, no others have been tested.  If slewing with STAR2000 causes you telescope to behave erratically, then this slewing control won't work for you.  Also, slewing is enabled globally, meaning if you are going to auto-guide and have slewing selected, you will quickly lose the guide star.  Make sure slewing is disabled whenever using the guide features, including training!

Reversing  Directions:

If you have an inverted CCD image, or you flip your scope while crossing the meridian, you may need to reverse the directions on one or both of the controls.

Select Interface:

Currently only LX200 style serial interfaces and STAR2000 style auto-guider interfaces are supported.  The manual option doesn't connect to any interface.  However the arrows remain active.  If you are diligent, you can mimic the arrows while auto-guiding.  You, in effect, become the interface.  The LX200 doesn't currently support any GOTO functionality.

Usage:

There are a few tips I've vome up while using and greating gccd.

Creating Calibration Frames:

Calibration frames are important, but a real pain to incorporate into a busy imaging session.  The best way to handle calibration frames is to create a library of frames that you can use instead of creating ones every imaging session.  Once the camera has stablized thermally, bias frames and scaled dark frames can be created and saved away.  Whenever the temperature changes by a few degrees however, the scaled dark frames won't match.  Make scaled dark frames at a variety of temperatures and save them in your library.  Flat fields can be created and saved for later assuming you don't change the optical configuration in any way.  The most important scaled dark frame is for the self-guided Starlight Xpress MX cameras.  Because the amplifier is turned on frequently to read out the guide image, a glow in the upper left hand corner results.  I create a master dark frame by creating a meridian combination of 5 - 30 minute exposures.  In order to get the auto-guiding control defeated, select the "Ignore Errors" option in the Self-Guide pane.  This takes a long time - the perfect exercise on a cloudy night.  Make sure your scaled darks are bias subtracted. Flat fields can be selected in the Exposure Options pane, or if you are using a filter wheel, each filter can have its own flat field .

Using Mutltiple Cameras:

Multiple cameras are supported.  This allows for simultaneous auto-guiding and exposing with only one computer.

Supported Devices:

Currently, CCD cameras use the CCD kernel drivers available in a seperate package.  Any CCD camera supported with one of these drivers is available to gccd.  The TrueTech Custom Filter Wheel is directly supported.  I understand it uses the same command set as the SBIG filter wheel, so it might also work.  The Starlight Xpress STAR2000 auto-guider interface is supported.  You don't need a Starlight Xpress camera to use this interface, it just connects to the auto-guider port on most telescope mounts.  The LX200 command set is supported but has only been tested on the Losmandy Gemini system.  Should work fine on all LX200 comatibles.