Документ взят из кэша поисковой машины. Адрес оригинального документа : http://graphics.cs.msu.ru/en/science/research/videoqualityincreasing/smartbc
Дата изменения: Sun Apr 10 02:20:14 2016
Дата индексирования: Sun Apr 10 02:20:14 2016
Кодировка: UTF-8
Smart brightness and contrast correction | Graphics and Media Lab

Smart brightness and contrast correction

MSU Graphics & Media Lab (Video Group)

First free professional Brightness & Contrast Filter!


The filter allows to adjust image brightness and contrast with minimum loss of details, also supports automatic mode. Object boundaries, that often create artifacts on automatic contrast improvement, are protected with "Accurate edges" technique. The algorithm can be applied to still images.

Main features

  • The filter doesn't shine bright areas, and doesn't move dark areas to 0 brightness value, keeping details in them.
  • Automatic brightness tuning mode, when contrast is set (useful for long mixed films montage).
  • Automatic brightness & contrast adjust mode (useful for long mixed professionally created films).
  • Color correction feature - usable for hard contrast correction of nearly black-and-white movies.
  • Adjustable intraframe contrast & brightness accuracy.
  • Controllable noise influence in final parameters in automatic parameters calculation.
  • Careful intrascene modification of parameters.
  • Automatic detection of scene replacement (use to prevent brightness flicking during one scene).
  • Supports Job Control and AviSynth.

Examples

We shall compare Smart Brightness and Contrast Filter to Photoshop filter using a portrait of M. Lomonosov as the first example.


Source image

Image processed by Photoshop Image processed by the Filter

Let us zoom in so as to see the differences better and take a close look at areas outlined in red.

 


Source image

Image processed by Photoshop Image processed by the Filter

The left histogram tells that Photoshop cares little about bright and dark pixels, it simply whitens or blacks them out. The right one tells that the filter tends to keep maximum detail and takes good care of bright and dark pixels.

 

And one more similar example with histograms.

Source image Histogram of source image
Image processed by Photoshop Image processed by the Filter
Histogram of image processed by Photoshop Histogram of image processed by the Filter

Frames from sequence MSU.avi:


Original frames
Processed with Brightness & Contrast from Photoshop
Processed with MSU Smart Brightness & Contrast in automatic mode (auto correcting manual settings)

Download

Filter for VirtualDub MSU Smart Brightness and Contrast (55 KB, ZIP)
Unpack this file in the folder "Plugins" in VirtualDub directory to use.

See also

Detailed description of MSU Smart Brightness and Contrast filter

The project team

Project lead:

  • Dr. Dmitriy Vatolin

Researchers:

  • Maksim Mahinya
  • Alexey Moskvin
  • Denis Kubasov

Contacts

e-mail:
video@graphics.cs.msu.ru