Content

Written by Esben Elfving   
Thursday, 21 January 2010 08:02

MDKGraphicsEngine 2.6

Features & Description

This article describes the features of the current version of the MDKGraphicsEngine. The intent of this article is to provide the reader information about the features in the current version. It is assumed that the reader has a basic understanding of image conversion, such as familiarity with different file formats and color spaces.


Some features require a specific version of the engine. The engine can be purchased in three different versions: Basic, Standard and Professional. Some features are enabled in the different versions and some has to be purchased separately. Please read the "Version & License" article for additional info on which version a specific feature is enabled in.

Key features

- Support for JPG, PSD, TIFF, PNG, GIF and BMP images
- Native color space workflow
- Extract metadata and file information from image files fast
- Alpha channel and transparency functionality
- Spot colors
- Photoshop and postscript clipping path support
- Extract layer data from Photoshop and TIFF images.
- Embedded CMM color space conversion
- Color space conversion based on either embedded or user specified ICC profiles
- Blend and add watermarks to images
- Sharpen and smooth images

 

File formats

The above mentioned file formats are supported for both reading and writing. In addition the PDF and EPS file formats can be used for writing.  

Metadata extraction and addition

The MDKGraphicsEngine reads and writes IPTC, XMP and EXIF metadata from JPG, TIFF and PSD files.  All metadata can be extracted without reading the actual image data. This allows fast extraction of metadata for search engines and other tools which do not need to read the image data.    

Color space conversion

The engine has an embedded CMM (Little CMS, littlecms.com) which is used as default. But the OS supplied CMM or the Adobe CMM (if installed on host machine) can be used as well. The engine supports Grayscale, RGB and CMYK as native color spaces. Unlike many other imaging SDK’s image data is not converted to RGB during load. This preserves the original color values internally until a color space transformation is done. However Black & White and indexed color spaces are not valid as internal spaces and will be converted during load.  So if an image e.g. has a B&W color space it will be converted to grayscale internally. Indexed color tables are converted to RGB internally.
ICC profiles can either be loaded from the OS or from the source image (embedded). Advanced CMM features like rendering intent, black point compensation and black key preservation is fully supported.

Alpha channels

Except for GIF and JPG images all formats which can be handled by the MDKGraphicsEngine supports alpha channels. It is optional whether to ignore or preserve alpha channels when an image is loaded.  If alpha channels are preserved they can be transferred to output images both as regular alpha channels or as a special transparency channel.

As default alpha channels are ignored when an image is loaded and therefore the alpha channel feature must be enabled first.

 

Spot Colors

TIFF and PSD images support separate channels saved at a spot channel (e.g. Pantone).  The MDKGraphicsEngine professional detects if a channel is saved with spot information and keeps this channel information during a conversion process.

Photoshop and postscript clipping paths

Adobe Photoshoptm (and other imaging products) are able to save clipping paths into JPG, TIFF and PSD images. The MDKGraphicsEngine can read and write these clipping paths from the supported formats. It is also possible to change the default clipping path from an image or remove it completely.  The engine can do an image "cutout" from a specified clipping path with a supplied background color. By saving the "cutout" image to a PNG or GIF image it is possible to set the supplied background color transparent. If an image is saved as PDF or EPS clipping paths is converted to postscript paths before saving.

From version 2.6 clipping paths can be converted to alpha channels as either regular or transparency alpha channels. If the converted image is saved as a PNG, TIF, Gif or PSD image the created alpha channel can be used to define the transparent background and thereby leaving the original image data intact.

Watermarks and blending

It is possible to blend an image onto another one. This is usually done for images which may only be shown with a watermark. As an advanced kind of watermark the engine can blend a text string onto the image with a specified font and color and at any position on the image.

Last Updated on Monday, 04 February 2013 07:33
 
 
Copyright © 2013, Middleware Denmark ApS