Full Frame Guiding and Focusing

A revolution on auto-guiding and focusing

Our full frame guiding and focusing patent pending technology does not require, nor need, any specific star for guiding and focusing* in real time. Instead any astronomical structures available in the whole (full) guider frame is used.

Want to know more? Click here to access the SkyGuide/SkyGuard SKG documentation (HTML).

Using the whole (full) frame of your guider

The traditional approach to auto-guiding software has relied on centroid calculations of one or a few guide stars using guiding cameras with relatively small chip diagonals. However, Innovations Foresight has developed an innovative technology that utilizes the on-axis guider (ONAG) technology, offering a large field of view (FOV) for the guiding camera without the need for rotation or motion of any optical components. This technology allows for auto-guiding and continuous auto-focusing using the entire guider frame, without the need for manual or automatic selection of guide stars.

By processing the full frame of the guider, Innovations Foresight’s technology improves signal-to-noise ratio (SNR) and reduces auto-guiding errors induced by atmospheric seeing. Advanced digital image and statistical processing techniques extract guiding and real-time focus information from astronomical structures visible in the guider frame. Unlike centroid calculations, this approach does not make assumptions about the shape of the guide star(s) and eliminates the need to manually or automatically find and select guide stars for auto-guiding. Instead, all pixels in the guider frame, or a selected sub-frame, are weighted automatically based on their likelihood of carrying relevant information for auto-guiding and auto-focus purposes using advanced multi-variate statistics.

This full frame approach allows for auto-guiding even under very low SNR conditions, where the signal is close to the noise floor. In such scenarios, traditional centroid-based auto-guiding software would typically lose track of the guide star, even with higher SNR. When used with an ONAG, this full frame approach enables real-time monitoring and control of telescope focus during image capture, ensuring sharp images. The technology is implemented in the SkySurveyor Suite (SKSS), including SkyGuide and SkyGuard, and it also features optimal guiding capabilities that predict the image full width at half maximum (FWHM) value based on the setup, mount performance, and local seeing conditions.

Please note that full frame real-time focusing requires the use of an ONAG, while full frame guiding can be employed with various setups, including guide scopes, off-axis guiders (OAG), self-guided cameras with filter wheels, or ONAG. However, when used with an ONAG, the guider can achieve a large FOV of up to 28mm, supporting the use of large diagonal guider chips such as those found in APS-C cameras.

 

Click on the link below to watch a SkyGuide introduction video:

SkyGuide introduction YouTube video

 

Here is yet another link to a TAIC YouTube video on SkyGuard:

 

SkyGuard Astro-Imaging Channel (TAIC) YouTube video

Theory of operation Demonstration with prototype software

How does it work?

The full frame guiding technology developed by Innovations Foresight utilizes advanced digital image correlation (DIC) techniques to determine the image registration between a reference guider image and the current guider frame. Unlike traditional guiding methods that rely on stars, this technology can utilize any pattern present in the guider image, making it versatile and not limited to star-shaped objects. Given the large field of view provided by an ONAG, there is a high probability of capturing astronomical structures within the guider frame, particularly when guiding on-axis to the target.

The algorithm employed does not assume the presence of stars or any specific shape for the objects in the guider frame. Instead, it models the noise characteristics of the guider image and extracts information from the noise floor. This is accomplished using an advanced multi-variate noise model and processing techniques, which make the system robust against various types of noise, artifacts, hot pixels, image gradients, and other sources of interference.

In the provided example of images taken at an Apollo landing site, there are no stars visible in the sky due to the short exposure time used to capture the picture, similar to daytime conditions. However, the full frame guiding algorithm can still extract relevant information from the available patterns in the guider image, such as the astronauts, the Lunar Excursion Module (LEM), the US flag, and the lunar background. This demonstrates the capability of the technology to guide even in scenarios where traditional star-based guiding methods would not be applicable.

Offset Image (~ 100px in X & Y)
Reference Image

The next two plots show the 2D and 3D DIC function of those two Moon landing images, notice the peak of the correlation function is located at registration values around 100 pixels from the upper left image corner.

DIC 2D, color coded (red peak, blue through)
DIC 3D
Advanced DIC and noise rejection:

The above Moon landing pictures feature a very rich background as well as a good SNR. However in the context of auto-guiding one usually does not have such a nice situation, but more like few stars, or a diffused pattern, among a sea of noise. To illustrate this, below two guider images of a single star offset by 30 pixels in X and Y. Since the SNR is poor they have been encircled in yellow for the reference frame and in green for the current frame for better visibility.

Raw DIC 2D (poor SNR)
raw DIC 3D (poor SNR)

Advanced DIC (ADIC) uses optimal filtering techniques as well as multi-variate statistics for extracting the useful information from the noise floor. Basically each, and every, pixel in the guider frame (or sub-frame if the user have elected to use only a portion of the all guider frame), are weighted and used based to its likelihood level of carrying information, noise pixel are therefore ignored. Below the processed 2D and 3D ADICs using Innovations Foresight proprietary advanced noise reduction and image enhancement resulting from the processing exactly the same two above guider images under poor SNR. Now one can  clearly see the image registration peak at 30 pixel (X and Y), as it should be.

Advanced DIC 2D (same guider frames)
Advanced DIC 3D (same guider frames)
More than one star

When using a device called an ONAG (Off-axis Guider) with a wide field of view or a guide scope in astronomy, it’s common to have multiple stars or astronomical structures within the field of view. This actually provides more information, resulting in better Signal-to-Noise Ratio (SNR) because of the way the entire frame is processed.

Another advantage of having multiple stars in the guider’s field of view is that the effect of atmospheric turbulence, known as “seeing,” is only correlated over a very narrow area called the “isoplanatic patch.” By having information from multiple stars across the guider’s field of view, their effects average out, which is particularly beneficial when using short guider exposures. This can be a limiting factor otherwise.

For instance, consider the following example: Two images demonstrate the performance of aligning 100 different guider frames. Each frame contains the same 40 stars with varying brightness levels (ranging from -4.5 to 0 magnitudes). Some stars may be too dim to be seen but are still utilized by the ADIC algorithm. Each of the 100 frames has its own unique noise pattern and seeing conditions, which vary for each star in the field of view.

The left image shows the reference frame with the brightest star (indicated by a red circle), used as a guide star for a comparison between a traditional centroid-based algorithm (which uses only this star) and a full-frame approach (which utilizes all the stars across the entire frame) for auto-guiding. The right plot displays a scatter plot of the registration errors (100) obtained from both methods. The traditional centroid-based approach is represented by red diamonds, using the guide star alone, while the full-frame technique is represented by green circles.

The plot shows that the full-frame algorithm achieves a much tighter cluster of errors compared to the centroid-based approach, with about a fourth of the standard deviation observed in the centroid cluster. In scenarios with limited seeing conditions and short guider exposure times, this difference becomes even more pronounced, especially when using an ONAG for guiding in the near-infrared (NIR) range.

40 stars under seeing limited conditions – encircled in red is the guide star for centroid auto-guiding
Centroid versus ADIC scatter plot – axes in pixel (100 frames)
ENJOY REAL TIME AUTO-FOCUS WITH AN ONAG

The ONAG presents a unique opportunity to maintain sharpness in your images while the camera’s shutter is open, thanks to its real-time auto-focus capability. Our patented “sharpLock” technology, which is already available with the “FocusLock” (FL) software using a single guide star, can now also be utilized with full-frame global processing using the SkySurveyor Suite (consisting of SkyGuard and SkySurveyor).

With SkySurveyor Suite, users can take advantage of the entire guider frame for real-time auto-focus when using an ONAG. Our full-frame technology is capable of extracting the necessary information, specifically astigmatism, for this task. Alternatively, SKSS also allows users to select a sub-frame and a single or a few stars for real-time auto-focus.

Additionally, there is a convenient register and stack feature that enables the stacking of multiple full or sub-guider frames before processing them for auto-focus. Since focus adjustments generally require a slower correction rate compared to auto-guiding, this approach enhances the Signal-to-Noise Ratio (SNR) and leads to more accurate auto-focus results.

SkySurveyor Suite (SGD and SKS): Full frame Guiding and Focusing (real time)

SkySurveyor Suite