Skip links

MODULE

AutoQuant Deconvolution

Industry-leading Image Restoration

Clear images in an instant

The best of AutoQuant standalone software has been upgraded, lab tested, GPU-accelerated by default, simplified into a single intuitive dialog, and is ready to fit into your imaging workflow.

The AutoQuant Deconvolution Module delivers repeatable, quality results in only a fraction of the time. Restore image fidelity, enhance quality, and reverse distortions introduced during the image acquisition process.

After
Before

Why Deconvolve Fluorescent Images?

Convolution + Noise = a ‘Blurry’ image

AutoQuant Deconvolution greatly improves both the image resolution and contrast, leading to enhanced visualization, better measurements, and more meaningful analysis. To do this, both the convolution and noise are handled to take your images from blurry to beautiful.

convolution

Convolution

When light waves encounter an object – such as the sample of interest, a lens, or even air molecules – they scatter & bend. Within an optical system, such as a microscope, this effect on a light wave is called the point-spread function, commonly abbreviated to PSF.

The PSF applies to every point of light passing through the optical system, in a process called convolution.

Noise

Digital images invariably suffer from the addition of some amount of noise that is introduced through a combination of ambient factors and the materials and electronics present in the digital camera hardware.

Signal-dependent noise can be characterized by a Poisson distribution, while noise arising from the imaging system often follows a Gaussian distribution.

Before & After Gallery

After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before

Intuitive Guided Process

AutoQuant is still the most intuitive deconvolution software in the industry, using a simple yet elegant workflow to direct any user through the necessary steps to achieve repeatable image restoration.

01

Confirm Image Metadata

02

Confirm Channel Metadata

03

Choose Starting PSF

04

Choose Deconvolution Settings

Deconvolve!

Straightforward Point-Spread Function (PSF) Models for Common Microscopy Modalities

Measure the PSF experimentally by collecting an image of a sub-resolution bead. When this small, the bead image represents the PSF of the microscope and appears as a single point of light that has been blurred.

Advantages

Provides a very accurate representation of the PSF

Includes & corrects for aberrations in the imaging system

Can be used with modalities unavailable as Theoretical

Disadvantages

Difficult and time consuming to collect correctly

Can potentially misrepresent the PSF if collected improperly

Widefield EPI Fluorescence

Two Photon (Multi-Photon)

STED

NEW

Confocal EPI Fluorescence

Spinning Disk Confocal

Other

Save time and effort collecting a measured PSF by using optical parameters such as Microscope modality, X, Y, & Z spacing, Objective Lens Numerical Aperture, and Refractive index of the medium to generate an accurate Theoretical model.

Advantages

Saves time over collecting accurate bead images

Not subject to changing variations in collection conditions

Disadvantages

Limited to the modalities supported by the software

Less accurate than a properly collected bead image

Widefield EPI Fluorescence

Two Photon (Multi-Photon)

STED

NEW

Confocal EPI Fluorescence

Spinning Disk Confocal

Other

Generates an initial PSF estimate from an auto-correction of the image data. This method requires only the image itself as an input.

Advantages

Doesn't require bead images or optical parameters

Allows just about any modality to be deconvolved

Disadvantages

Available only for single plane (2D) images/sets

Requires more iterations and is susceptible to over-processing

Widefield EPI Fluorescence

Two Photon (Multi-Photon)

STED

NEW

Confocal EPI Fluorescence

Spinning Disk Confocal

Other

The Deconvolution Cycle

Step 1

Start with a PSF estimate, either calculated from optics or input as a bead image.

Step 2

Create an initial image estimate using the original image, which may be inverse filtered. Convolve the image estimate with the PSF estimate to create a re-blurred estimate.

Step 3

Compare the re-blurred estimate against the original image to determine the adjustment to be made to the image estimate.

Step 4

Update the image estimate and amplify it until it no longer improves. After several iterations, the difference between the original image and the re-blurred estimate becomes minimal and the original image is deconvolved.

Step #1

Start with a PSF estimate, either calculated from optics or input as a bead image.

Step #2

Create an initial image estimate using the original image, which may be inverse filtered. Convolve the image estimate with the PSF estimate to create a re-blurred estimate.

Step #3

Compare the re-blurred estimate against the original image to determine the adjustment to be made to the image estimate.

Step #4

Update the image estimate and amplify it until it no longer improves. After several iterations, the difference between the original image and the re-blurred estimate becomes minimal and the original image is deconvolved.

Time tested and still trustworthy

While integrating the time-tested deconvolution algorithm in the new Image-Pro module, we conducted a series of laboratory tests to ensure the new integration is still trustworthy and that it still delivers reliable, repeatable results, even within a new platform. The whitepaper below details the results and conclusions of those tests.

In short, all of our new tests yielded reassuring outcomes. The AutoQuant Fixed PSF and Adaptive PSF algorithms continue to deliver consistent improvement and reliable data in all cases, especially for 3D data where we observed nearly identical improvement. Indeed, none of our tests showed a significant difference in results. Therefore, the new module has been streamlined to focus on the Fixed PSF algorithm alone for 3D images, making setup faster and generating the same quality data as previous AutoQuant versions.

Deconvolve Images Faster with
Built-in GPU Acceleration

For computationally-intensive applications like deconvolution, being able to process thousands of mathematical calculations at once is preferable to processing a few at a time. In short, you have more bandwidth available when you utilize the GPU rather than the CPU allowing for faster processing of the image.

Image Size

Deconvolution Time

The CPU-GPU Collaboration

Certain portions of the deconvolution process still occur on the CPU, notably the initial setup. Once this is complete, AutoQuant accesses the many GPU processor cores that carry out the actual iterative deconvolution. The final results are then transferred back to the CPU and saved.

CPU-GPU-Collaboration

Step 1: Initial Setup

Allocate memory and initialize math libraries.

Step 2: Transfer Data to GPU

Step 3: Perform Iterations

Convolve image guess with PSF guess. Compare the results with the original data. Update the image and (if bling) PSF guess. Continue to the next iteration.

Step 4: Retrieve Result from GPU

Thousands of Cores

The individual processor cores on a GPU are considerably slower than most CPU processor cores, but what they lack in power, they more than make up for in numbers.

Multiple Cores

The core count on a high-end workstation’s CPU(s) will typically be around 32. In contrast, the core counts reported on even consumer-grade GPUs number into the thousands.

Batch Processing

Use Image-Pro’s native Batch Processing to select a folder of similar images to be deconvolved.

Save valuable time. Set up the batch and walk away.

Deconvolve groups of images from the same experiment without the need to write custom macros.

Choose either a fixed or adaptive PSF and apply it to the whole batch.

Automatically save deconvolved datasets for review and analysis.

AutoQuant

AutoQuant

Image-Pro

Image-Pro

Use it alone or combine it with Image-Pro modules for an even more powerful solution.

3D Viewer

3D Viewer

3D Analysis

3D Analysis

2D Automated Analysis

2D Automated Analysis

How does the new Module stack up against the older Standalone software?

Standalone New Module
Macro language
OME file and metadata support
Theoretical PSF for STED Modality
3D Movie Maker

3D Viewer
Image Adjustments and Correction
Image Import and Set Builder
Spherical Aberration Correction
Batch Processing
GPU Acceleration
Inverse Filter
No/Nearest Neighbors Algorithm
DIC Restoration

* = Optional

Like What You See?

Get Started with Image-Pro for AutoQuant Deconvolution.

This website uses cookies to improve your web experience.