The Colour-Science local sharpness enhancement is a totally new approach for sharpening or smoothing images. Until now sharpening was normally a process applied to the whole image with a predefined strength. The problem was that this works well for some images, but others are probably over-sharpened or even worse the noise in the image is strongly increased. In fact many images with bad quality or high jpeg compression would rather need some smoothing then sharpening, because sharpening will only increase the jpeg artefacts or the image noise.

 

LSE is examining first the image and then decides which parts of the image have to be sharpened, smoothed or left as they are. LSE makes use of the Colour-Science image segmentation technology.

 

Sample of LSE sharpened images

 

Selective local sharpening or smoothing

 

 

 

Visualisation of the masks used to calculate the sharpening/smoothing alpha mask

 

 

The sharpening/smoothing alpha mask is constructed by a combination of special detail masks

The alpha mask is in fact the sharpening/smoothing factor. White parts are sharpened, black parts are smoothed and grey parts are leaved unchanged

 

The new LSE is active in image Editor version 2.4.

 

Here we have some images which have been sharpened with normal unsharp masking and with LSE sharpening.

Download LSE sample images (6.7Mb) >>>

 

You can also open these images directly in the Internet explorer. But it is quite slow if you do not have a fast connection.

noisy image with lse sharpening.jpg

noisy image with normal sharpening.jpg

skin smothing with lse sharpening.jpg

skin with normal sharpening.jpg

sky artefact removal with new lse sharpening.jpg

sky artefacts with normal sharpening.jpg

sky noise removal with lse sharpening.jpg

sky noise with normal sharpening.jpg