Elbrus plate solver
Upload a picture of the sky and Elbrus tells you where it points: the coordinates of its centre, its orientation and its scale.
In memory of Alfonso Pulido, who wrote Elbrus between 2003 and 2009 and taught many of us not to get lost among the stars. "Better lost than wrongly found."
FITS, JPEG, PNG, BMP, TGA, GIF · up to 25 MB
Know nothing about the picture? Leave everything empty: Elbrus solves it blind, telescope photos and camera-lens shots of constellations or the Milky Way alike. Giving the scale and the approximate position makes it faster and more certain.
Try an example
Photos from Wikimedia Commons, under their authors' free licenses. "Solve" fills in sample parameters; "Solve blind" gives Elbrus nothing but the picture.
Green circles: stars recognised in the catalog. Cyan: some of the triangles that matched. Amber: known objects (OpenNGC, HYG).
Get the Android app
Elbrus on your phone: point it at the sky and it names the stars, constellations and planets as you move, or solve a picture from your gallery. It works offline: the camera-lens star database travels in the app. Android 9 or newer.
Scan the code with your phone's camera, or open this page on the phone. Installing an app from outside Google Play needs "Install unknown apps" for your browser. Elbrus checks for updates once a day and asks before downloading one.
Installed a test build from before 1.4.4? Uninstall it once first: the published builds are signed with a new key.
Check the download
File SHA-256:
Signing certificate SHA-256: 6a6cf8636ce8586d84028de7611a44dc4e90b592043d7121643b7b8e323aa54a
Check them with apksigner verify --print-certs and sha256sum (or Get-FileHash).
In memory of Alfonso
Alfonso was a good friend, and an inspiration when I was refining my way of thinking to solve engineering problems. Thank you, Alfonso, and thank you, Galia, for all the good times, and for your patience, your enthusiasm and your rigour.
Alfonso fue un buen amigo y una inspiración a la hora de depurar mi manera de pensar para resolver problemas de ingeniería. Gracias, Alfonso, y gracias, Galia, por todos los buenos momentos, por vuestra paciencia, vuestro entusiasmo y vuestro formalismo.
José Luis Rey
About Elbrus
Elbrus ("Elemento Localizador por BRUjula Sidérea") is the star locator of the Spanish amateur astronomer Alfonso Pulido, written with Galia Pulido. Observers used it to know where their telescopes pointed, for remote observing and to recentre between exposures. In 2008 the IAU named asteroid 66843 Pulido after him for it. Alfonso died in 2023.
This service runs a C++ port of his program, made by his friend José Luis Rey as a tribute. Elbrus keeps the 5 brightest stars of every small box of the sky and the distances between them, finds the distances it measures in your image, builds polygons of matching stars, checks every side against the catalog, and answers only when one polygon wins clearly. The wide-field mode does the same with triangles over the whole sky. And when nothing is known about a picture, groups of four stars are compared by their shape alone, which does not depend on the scale or the orientation; that proposes a scale and a place, and Alfonso's method then has the last word. Photos taken with a camera lens (a constellation, the Milky Way over a landscape, 10° to 100° across) go to a fourth database: the same shapes over the 41,000 stars brighter than magnitude 8, with the lens's own distortion fitted into the solution.
Alfonso's pages: astrosurf.com/pulgar ·
REST API
Free and open, no key. Fair use: 30 solves per hour per address, one solve at a time, 60 s per solve. Images are not stored.
curl -F image=@photo.jpg https://cs-miniserver.chronoscan.org/api/elbrus/solve curl -F image=@m52.jpg -F scale=1.53 -F ra=351.2 -F dec=61.6 https://cs-miniserver.chronoscan.org/api/elbrus/solve curl -F image=@wide.png -F scale_min=20 -F scale_max=30 https://cs-miniserver.chronoscan.org/api/elbrus/solve curl --data-binary @image.fit "https://cs-miniserver.chronoscan.org/api/elbrus/solve?scale=2.4&ra=83.8&dec=-5.4" curl https://cs-miniserver.chronoscan.org/api/elbrus/info
Parameters (form fields or query string)
| Name | Meaning |
|---|---|
image | The image (multipart field), or the raw request body. |
scale | Arcsec per pixel. Or scale_min + scale_max to scan a range (max ≤ 4 × min). Optional: with no scale at all the image is solved blind. |
ra, dec | Approximate centre in degrees. Needed for fields under ~2°; narrows the wide search. |
window | Search window around ra/dec, degrees (default 5; 30 for wide fields). |
angle, tol | Known image angle (position angle of "up", north through east) and its tolerance; otherwise all angles are tried. |
flip | 1 if the image is mirrored (otherwise both are tried). |
mode | auto (default), 1 narrow-field database, 2 wide-field database, blind ignore the scale and search by star-quad shapes, lens camera-lens database only. |
err, stars, min_sides, min_stars | Advanced: distance error %, stars used, sides or stars needed to accept a solution. |
Answer
solved, failure, blind (found with no hints), blind_matched, database (1, 2, 3 = star quads only, or 4 = camera lens), ra, dec (degrees, ICRS), ra_hms, dec_dms, angle, flipped, scale_x, scale_y (arcsec/px), field_deg, sides, rms_arcsec, matched (x, y, ra, dec, id, resid_arcsec), objects (known objects on the image: name, kind M Messier / D deep sky / S star, x, y, r in pixels, mag), plate (tan_ra, tan_dec and the coefficients P, Q: ξ = P0·px + P1·py + P2, η = Q0·px + Q1·py + Q2 in arcsec, px = x − width/2, py = height/2 − y; camera lenses add k1, k2, r0: px and py are first multiplied by 1 + k1·q + k2·q², q = (px² + py²) / r0²), lens_search (when the camera-lens search ran), image, seconds. Errors answer 400 (bad input), 429 (over the hourly budget), 503 (busy).
Star positions: ESA Gaia DR3. This work has made use of data from the European Space Agency (ESA) mission Gaia (cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC). The star databases derived from it are under CC BY-SA 3.0 IGO. Object names: OpenNGC by Mattia Verga and the HYG database by David Nash, both CC BY-SA 4.0. The camera-lens database uses the HYG stars brighter than magnitude 8.