Moravian instruments, Inc., source: https://www.gxccd.com/art?id=609&lang=409, printed: 19.10.2020 23:30:03
|Majority of new features of the Scientific Image Processing System version 3.22 are related to the word Scientific in its name, improving astrometric and photometric processing of images taken for research purposes. Also, the single important bug-fix of this SIPS release concerns unreliable astrometry plate solution of arbitrary rotated images.|
While SIPS implemented code for general image rotation (more precisely not only for rotation, but for arbitrary 2D transformation including scaling and translation), used when stacking multiple mutually shifted and rotated images, this function remained hidden in the Auto Align command of Image Blink or Image Add tools and was not presented in the program GUI to allow user to arbitrary rotate any image. SIPS v3.22 introduces this function into Image Transform tool, which up to now offered only image rotation by multiply of 90 degrees.
As opposed to rotation by multiply of right angle or image mirroring, general rotation cannot exactly preserve values of individual pixels. The operation inevitably performs sub-pixel operations. Still, the transformation is designed to preserve all important image features (e.g. total flux of individual stars) as close to original as possible.
Astrometry reduction in SIPS v3.22 fixed number of issues.
Astrometry in SIPS suffered from one important bug, making plate solution of arbitrary rotated images unreliable. This bug evaded detection because it caused problems only if image orientation was far from multiply of right angle. Even quite significant differences from right-angle orientation (up to approximately 10 or 20 degrees to both sides) were tolerated. Also rotation of the images by 180 degrees after telescope meridian flip brought no problems, providing original image axis were at last approximately in the north-south or east-west directions.
Either way, version 3.22 is capable of proper astrometry solution regardless of image rotation.
All SIPS astrometry manuals and presentations emphasized how important is to set parameters for searching stars to find all, even the very brightest stars within the image. If there are bright stars on image left undetected, e.g. because the used aperture is too small to cover whole bright star image, a set of image brightest stars and a set of brightest stars retrieved from the catalog are different. This can lead to unsuccessful plate solution (this also depends on number of image and catalog stars used for matching etc.).
However, when a NABG camera is used, images of bright star are typically heavily distorted by blooming. Event if such stars are detected, calculated star centroids are way off the real star centers. So, if such stars are detected and used for plate solution, results are always wrong because star pattern seen by SIPS within the image more or less differs from the pattern of stars read from catalog.
SIPS v3.22 solves this problem by adding of one more parameter into Image/catalog matching parameters dialog box, allowing the user to define maximal number of brightest stars in the catalog, which should be skipped.
Please note this is the maximal number of skipped stars, SIPS starts with plate solution for all catalog stars and only if it fails then also for all values up to defined maximum to check if the catalog pattern matches the pattern defined with the processed image. If the used camera is equipped with ABG and also star search parameters are set that all stars are always found, this value should be set to 0. However, if the image contains bloomed bright stars, intentionally not detected by star search algorithms, increase this number to roughly equal to number of such stars (it is not necessary to set this number to exact number of bloomed stars).
Photometry tool in SIPS v3.22 was also significantly enhanced.
New version 3.22 fixes this bug.
Previous versions of the SIPS photometry tool used quite wide displacement tolerances when trying to find VAR/CMP/CHK star on each image. These tolerances were derived from the chosen photometry apertures, which turned to be inappropriate decision. Especially in dense fields and when measuring very weak star, which did not fulfill star detection criteria and was left undetected on some image, wide tolerances lead to marking of nearby, but different star on particular image as proper VAR/CMP/CHK star. Flux of this star was used to calculate respective point of light curve, of course causing an outlier as the measured star was not the proper one.
New SIPS v3.22 abandoned algorithm, determining star position tolerances from used photometric aperture, and replaced them with fixed, much tighter tolerances. This virtually eliminated mis-selection of different stars on images, on which the actual star was not detected.
Mutual binding of individual parts of the Photometry tool and their common reaction to user's actions was always a design goal. For instance a click on a point in the light curve selects corresponding image in the list box, shows the original frame in the image pane and centers variable star etc. Now multiple points in the Light curve pane can be selected (framed) by mouse click and drag (similarly like selection of a part of an image).
All images, corresponding to selected (framed) points in the light curve, are the highlighted in the photometry image list box. Once images are highlighted, they can be e.g. excluded from the light curve using the Deselect highlighted images in list tool in the list tool bar or just by a hot key Ctrl+U.
Many actions within the SIPS Photometry tool can be performed using keyboard hot-keys (for instance Ctrl+F selects the first image, Ctrl+L selects the last image, Ctrl+A includes all images into the light curve, Ctrl+R opens Photometry report dialog box, Ctrl+D selects Field description pane, so the cursor keys can be then used to go to next filed description item etc.). Remembering of these shortcuts cam save lots of mouse clicks and makes processing of photometry series much more effective.
SIPS photometry report (photometry protocol) could contain various comment lines, e.g. with star meta-data etc. (star meta-data contain important information about stars used to generate particular light curve and could be especially important if protocols are further processed e.g. using the SILICUPS software package). SIPS v3.22 adds one more User's comment line, this time left for user to put an information he or she wants.
There already exists a user-defined line marked Comment prefix. If this line is left empty, comment lines appear in the photometry report file as is, without any prefix. All comment lines begin with some keyword (first character is a letter), so comment lines should not be confused with data line, which begins with a number (julian date of a particular point).
Still, if some software package, used to process light curves, require some special character to indicate comment (non-data) line, it can be defined as the Comment prefix. Typically the comment lines begin with character '#', but any other character or character combination can be used.
Please note the Comment prefix character(s) is inserted at the beginning of each comment line to help to distinguish them from data lines. So, the Comment prefix is not intended to keep any specific information.
SIPS is a freeware and can be downloaded from the Download section of this web site.
Do you want to stay informed about new SIPS versions and features? Follow @sipssw on Twitter..