|
eLynx SDK
v3.3.0 C++ image processing API reference |
The all processing have the same goal: reduce noice to improve image signal also call details enhancements.
You can get various techniques explained in :
When you sharpen an image using whatever method you choose, there is a limit to how far you can sharpen that image. If you sharpen too aggressively, you run the risk of creating artifacts. These are false details that don't actually exist in the image. They are created by the overlyl aggressive sharpening process. This applies to any form of sharpening, including high pass filters, unsharp masking, deconvolution, and so on.
See articles : Real Digital Unsharp Masking
A type of blurring that looks more natural. Many other types of blurring, such as simple averaging, do not look as natural as a Guassian blur. The Gaussian blur works by weighting the contributon of surrounding pixels to the blur. The weighting is based on a Gaussian distribution (bell curve). This adds low-frequency data to the blur and is very effective for blurring noisy or background areas in an astronomical image.
See articles : Compositing part1, part2.
Image registration is the process of establishing point-by-point correspondence between two images of a scene. This process is needed in various computer vision applications, such as motion analysis, change detection, object localization and image fusion.
See articles : Registration tutorial,
See articles : Channel Substitution, Synthetic Luminance Channels, LRGB Color Production, Color CCD Imaging with Luminance Layering by Robert Gendler.
DDP is an image processing algorithm which diplays information in a non-linear way. Normally a CCD image is displayed in such a way that the faintest details of an image are visible. However, in doing so very bright portions of an image look very over-exposed.
See articles : Algorithm from Dr. Kunihiko Okano, DDP.
See articles :