User Manual

NOTE: This manual is a bit outdated. The updated manual will be released with the next program version.

Index:

Introduction

lxnstack is a software to align and stack astronomical images (both planetary and deep-sky) This program is open source and is distributed under the GNU GPL version 3 license.

The requisites to run this program correctly are (as already reported in the installation instructions):

Load files

Click the menu 'File'->'Open files' or push the '+' button in the 'frames list' section and an open file dialog window will appear. There you can select the files to be loaded. All images must have the same size and the same number for color channels (for example you cannot add an RGB image with a gray-scale one).

You can open a video (for example a video captured with a webcam) by clicking the menu 'File'->'Open video'. If you loaded other images, then each frame of the video must have the same size and the same number of color channels of the previously loaded images. At loading, the video is splitted into frames that are saved in a temporary directory ('/tmp/lxnstack' by default), so assure you have enough space for the operation.

Webcam capture

lxnstack can capture images from a video4linux device (like a webcam). Check the 'Capture' checkbox in the 'frames list' section to enable the video capture mode.

If no capture device has been selected yet ( see Settings ) a dialog window will appear where you can choose an available device.

WARNING: opencv libraries CANNOT HANDLE DEVICES USED BY OTHER PROGRAMS and this can lead to a crash! CLOSE ANY OTHER APPLICATION USING A CAPTURE DEVICES.

If all is ok the capture controls will be enabled and a live preview will be shown.

Press 'Start' button to start capture and 'Stop' button to stop capture.

To set the number of frames to capture, check the 'Max' checkbox and digit the number in the right spinbox.

Press 'Single shot' to take a single shot.

ATTENTION: remember to exit the capture mode by unchecking the 'Capture' checkbox when finished!

Display and zoom images

To show an image you can simply select its name in the 'frames list' and it will be shown in the canvas on the right.

To modify the zoom you can use the 'zoom' controls

Three types of zoom are available:

zoom: none

zoom: fit

zoom: full

Images selection

You can choose the image to use in the average process by checking the checkbox on the left side of each image in the 'frames list'. The images with this checkbox unchecked will not be used.

In this way the unwanted images can be excluded.

Automatic alignment

Two types of alignment are available: the alignment method can be choosen by selecting it in the combobox.

The actual alignment is performed by pushing the 'Align' button in the 'Tools' section.

Alignment by phase correlation

This method does not require any user action, just press the 'Align' button in the 'Tools' section to start the alignment process. If the images are misaligned, try to change the sharpness value.

Alignment by points

Push the '+' button to add an alignment point and the '-' button to remove the currently selected alignment point.

You can detect the best align points pushing the 'Automatic detection'. The points are chosen by their quality (see Settings ).

The current selected alignment point can be moved by clicking on the displayed image with the right mouse button or by using the spinboxes 'X offset' ed 'Y offset'.

All align points must be set on all images in the same position. To do this an automatic positioning feature can be used: you can this process by pushing the 'Set on all images' button. All points will be automatically detected on all images using the current shown image as reference. Only the points that have not been detected yet or that are been moved will be detected/re-detected.

ATTENTION: avoid to place the alignment points near the image borders!

Manual alignment

An alternative manual alignment is available in the 'Manual alignment'. This method is based on a comparison of two images: a reference image can be selected by checking the checkbox on the left side of its name in the list while the image to be aligned can be selected simply clicking its name in the list with the left mouse button.

The difference of the two images will be shown

The lower is the difference the darker will be the image shown. The spinboxes 'X offset' and 'Y offset' can be used to move the image and align it to the reference image.

Darkframe e Flatfield

In these sections the calibration images (master-dark and master-flatfield) can be selected or created.

If you do not want to use calibration images, simply leave blank the two sections.

NOTE: the best practice is to create first the master-dark as average of dark-images (skipping the alignment process), then create the master-flat as average of flat-images (skipping the alignment process) using the master-dark previously created (and scaled to adapt to the dynamic of the flat images) and at the end use both these two files to calibrate the images you want to stack.

The calibration images can be created by adding the single images to the respective lists.

NOTE: If you want, when you load some images the directory in which they are located can be scanned for the presence of the folders 'dark' and 'flat' by enabling the checkbox "Automatically scan for dark and flat frames" in the application settings. The images located in these directories will be added to the respective lists (for example if you load some images from the directory '/home/images', the dark and flat images will be searched in the directories '/home/images/dark' and '/home/images/flat').

If you already made the calibration images (you can make them using this program by loading the dark/flat frames and adding them without aligning), you can check the 'Use master-dark' and/or 'Use master-flatfield' checkboxes and selecting the files clicking the '...' buttons or typing the full url of the files in the respective text box.

Lastly you can choose a factor by which the master files will be multiplied before the calibration takes place (the value is 1 by default).

Average

First of all you need to align the images (if you haven't done it yet) by pushing the 'Align' button in the 'Tools' section. After the align process ends you can add the images using the 'Add' button.

When the operation is ended the resulting average image will be shown.

NOTE: The "exposure controls" tool has been removed because it does not work anymore since the computation is done using floating point numbers instead of integers. Moreover there are several image manupulation programs that provide similar functionalities.

Here are shown a single image and the average of 33 images calibrated with a master-dark and a master-flatfield.

The application settings are available trough the menu 'Options'->'Preferences'

In the 'Settings' section you can choose the application language.

The settings for automatic alignment can be changed in the 'Alignment' section: The 'rectangle size' controls the size of a portion of the image (centered on the alignment point and marked by a yellow rectangle) that is compared with the other images during the automatic positioning process ( see Automatic alignment ). By default this "rectangle" is compared with a limited portion of other images. If the image shift is excessive, the alignment points will not be correctly set: in this case it is useful to check the 'Use whole image to track points'. However also using the entire image method has its negative consequences: it is slower and there are more false positive detections.

The maximum number of alignment points that can be detected automatically and their minimum quality can be changed in the 'Auto detection' box (For more information refer to opencv user manual and search for "goodFeaturesToTrack").

The capture device can be chosen in the 'Capture devices' section. The basic device settings can be also changed. Use the programv4l2-ctl to change other settings.

WARNING: opencv libraries CANNOT HANDLE DEVICES USED BY OTHER PROGRAMS and this can lead to a crash! CLOSE ANY OTHER APPLICATION USING A CAPTURE DEVICES.

ATTENTION: Opencv libraries cannot change the device properties 'video format' (such as 'YUYV') and fps. These feature are been added only for forward compatibility in the hope they will be implemented soon.

Note

This program is distributed AS IS and without any warranty.

The author takes NO RESPONSABILITY for eventual damages.

For more legal information see the GNU GPL version 3 license. You should have received a copy along with this software, if no see the website http://www.gnu.org/licenses

lxnstack © 2013 Maurizio D'Addona