Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/skyrobot804/node_v1/llms.txt

Use this file to discover all available pages before exploring further.

The Imaging panel brings together four related capabilities: manual camera control for single-frame or test exposures, a FITS file browser for reviewing and analyzing saved frames, a pier camera live view for monitoring the physical mount, and a live stacking module that co-adds incoming sub-frames to reveal faint detail in real time. All four are accessible from a single panel in the dashboard without needing a separate application.

​
Manual Exposure

The Expose form lets you trigger a single camera exposure on demand. Click Capture to send: 
POST /api/camera/expose
{
  "duration": 10.0,
  "binning": 1
}
​
duration
float
required
Exposure duration in seconds. Fractional values are accepted (e.g. 0.5 for a half-second frame).
​
binning
integer
required
Camera pixel binning factor. 1 = 1×1 (full resolution), 2 = 2×2 (quarter resolution, four times the sensitivity), etc.
The default values for these fields are drawn from the node configuration:
Config keyDescription
camera.exposure_durationPre-filled exposure duration when the Expose form loads
camera.binningPre-filled binning when the Expose form loads
Manual exposures are saved to the same watch directory as scheduled exposures and immediately appear in the FITS Browser. They are also logged in the observation history.
When the exposure completes, the resulting FITS file path is returned in the response and the FITS Browser list refreshes automatically.

​
FITS Browser

The FITS Browser panel lists every FITS file in the node’s configured watch directory. Clicking Refresh (or waiting for the auto-refresh interval) sends: 
GET /api/fits/list

// Response (excerpt)
[
  "m42_001.fits",
  "m42_002.fits",
  "ngc891_001.fits"
]
Clicking any filename in the list opens a preview of the image, rendered directly in the browser with an auto-stretch applied for visibility. From the preview view you can:
  • Run Photometry — triggers POST /api/photometry on the selected file, sending it through the aperture photometry pipeline. Results appear in the Photometry status panel.
  • Submit to AAVSO — available after a successful photometry run; sends POST /api/aavso with the measured magnitude and metadata.
  • Open in History — links the file back to its parent observation record in the history modal.
FITS files accumulate quickly during a long run. Use the filter box in the FITS Browser to search by filename prefix (e.g. "m42") and focus on the frames from a specific target.

​
Pier Camera

The pier camera is an optional secondary USB camera—typically a ZWO ASI unit—mounted on the pier or tripod to provide a wide-angle view of the scope and mount during a session. It is useful for verifying that cables are not snagging, the mount is moving freely, and the scope cap is off. Enable the pier cam by setting pier_cam.enabled: true in config.yaml. When enabled, the dashboard displays a live MJPEG-style feed in the Pier Cam panel, fetched from: 
GET /api/pier-cam/snapshot
This endpoint returns a single JPEG frame on each request. The dashboard polls it at the configured frame rate to produce the live view. The pier cam is configured with these config.yaml keys:
Config keyTypeDescription
pier_cam.enabledboolSet to true to activate the pier cam feed
pier_cam.device_indexintUSB device index of the ASI camera (0 = first connected)
pier_cam.exposure_msintSensor exposure time in milliseconds per frame
pier_cam.gainintAnalog gain setting (ASI driver scale, e.g. 0–500)
pier_cam.binintBinning factor applied to the pier cam sensor
pier_cam.target_fpsfloatTarget frame delivery rate for the live view
pier_cam.jpeg_qualityintJPEG compression quality (1–100; 80 is a good default)
The pier cam runs independently of the main imaging camera and does not affect exposures or the FITS pipeline. Its frames are served as JPEGs only and are not saved to disk.
Setting pier_cam.target_fps too high on a Raspberry Pi or low-power node can saturate the CPU and cause the Flask server to become unresponsive. Start with 2–5 fps and increase only if the node has headroom.

​
Live Stacking

Live stacking co-adds incoming sub-frames in real time to accumulate signal and improve the signal-to-noise ratio as the night progresses. Each new frame is aligned to the first frame using RANSAC-based translation estimation, then added to the running stack. Because noise averages out while signal adds coherently, SNR improves proportionally to the square root of the number of stacked frames (√N): stacking 25 frames yields a 5× SNR improvement over a single frame. Start and stop stacking with:
ActionEndpoint
Start live stackingPOST /api/stack/start
Stop live stackingDELETE /api/stack/start
Both endpoints accept an empty JSON body {}. The stacking module is configured via config.yaml:
Config keyTypeDescription
stacking.framesintTotal number of sub-frames to accumulate before stopping automatically (0 = run until stopped manually)
stacking.exposure_sfloatDuration of each sub-frame exposure triggered by the stacker
stacking.preview_everyintUpdate the dashboard preview PNG after every N new frames
As frames arrive, the dashboard displays a preview of the current stack in the Live Stacking panel. The preview is a PNG with auto-stretch applied so faint nebulosity becomes visible long before the full sequence is complete. The preview updates every stacking.preview_every frames, so setting this to 1 gives a live update after every frame at the cost of extra CPU. 
POST /api/stack/start
{}

// The stacker begins exposing and stacking. Progress is streamed
// through GET /api/logs and the preview updates in the dashboard.
RANSAC translation alignment handles small tracking errors and atmospheric drift, but it does not correct for field rotation. For mounts without polar alignment or for long focal lengths, field rotation will degrade stack quality over long sequences. Use an equatorially mounted scope with reasonable polar alignment for best results.
Set stacking.preview_every to 5 or 10 for a good balance between preview freshness and CPU load on embedded hardware. On a desktop machine you can set it to 1 for frame-by-frame updates.
For the full module reference—including the alignment algorithm details, output file formats, and integration with the photometry pipeline—see the Live Stacking guide.

Build docs developers (and LLMs) love

Get started for freeTalk to us