Colour-Science i2e intelligent image enhancement software librarys

Colour-Science i²e Intelligent Image Enhancement software libraries

Introduction

The i2e image enhancement library is available for the following environnements:

• Windows 32bit

• Windows 64bit

• Linux 32bit

• Linux 64bit

• MAC 32bit

Please email to mail@colour-science.com if you would like get the SDK (Software Developpment Kit).

It provides all the image processing functions which are needed to enhance images coming from digital cameras or scanners. The DLL's can be used with almost any programming environment (including C/C++, VB, C#, MC++, Delphi, Access, and many more).

The library is constantly updated with new improved functions. The following 2 documents will show you the version history and also planned future improvements:

• i²e image enhancement library version history >>> version history

• i²e image enhancement projects and future improvements >>> future improvements

The current Version of the Library is Version 6. This new versions includes the following improvements:

• Oversaturation of saturated colors is removed.

• Clipping of bright whites is reduced.

The i2e image enhancement library lets the user scale the image enhancement pipeline which is suitable for his application. This way he can configure for example a basic image enhancement pipeline which runs on a MMS cell phone or a more complete enhancement pipeline with all features for a high end film scanner. There are several enhancement stages in the pipeline which can be enabled or disabled.

The enhancement stages and technologies can be divided in two classes. The base processing stages act globally on the whole image. In addition to these global enhancement stages there are some specialized stages which act locally on the image.

Global image enhancement processing stages	Local image enhancement processing stages
ABE - Adaptive Brightness and Contrast Enhancement	SHE - Shadow and Highlight Enhancement
ACE - Adaptive Color Enhancement (including single channel film statistics if used for film scanners)	MCE - Memory Color Enhancement
ICC - ICC input and output profiling	LSE - Local Sharpness Enhancement (for example: remove grain in the sky but sharpen vegetation)
	ENR - Edge preserving anisotropic Noise Reduction
	RER - Automatic Red Eye Reduction using face detection

In addition to these enhancement stages the library allows a statistical analysis of the whole film which is especially appropriate for negative film scanners allowing to scan in a single channel mode.

The i2e library is doing an extensive analysis of the images and the data of this analysis is available also for other purposes then image enhancement. The data can for example be used for image classification.

The following image info is available:

• vegetation in % of total image surface

• sky in % of total image surface

• skin in % of total image surface

• number of faces

• face coordinates of every face rectangle

• total face surface in % of total image surface

• exposure level

• local contrast

• red eye flag

• exif tags

• others ... (we are currently working on this part of the library)