Transcript
Stettbacher Signal Processing Neugutstrasse 54 CH-8600 Dübendorf Phone: Fax: E-Mail:
+41 43 299 57 23 +41 43 299 57 25
[email protected]
O-3000 Camera Series White Balancing, Color Adjustment
Version 1.10 2014-09-16
Abstract: This application note will guide the user through the process of color balance adjustment using the ’O-3000 Java Demo App’, available with the camera.
O-3000 Camera Series
Contents 1
Introduction
2
2
Color Balancing
2
3
Typical Values
3
4
Notes
3
1
Introduction
An scene captured by a camera and displayed on a computer screen1 often looks different than the same scene watched directly. The actual lighting situation typically has a strong impact onto the chroma (hue) of the camera image. This is because of the light’s so-called color temperature. An incandescent light source will have another impact on a scene than a fluorescent lamp. The white offset will drift towards red hue in a low color temperature and towards blue hue in a high color temperature light respectively. This behavior is undesired in most cases. In order to correct unnatural chroma, all color cameras provide a color balancing mechanism. Many consumer cameras offer automatic color balancing. In industrial application, manual color balancing is usually preferred. The following chapters describe the procedure for the O-3000 camera series.
2
Color Balancing
The O-3000 Cameras and driver allow for manual color balancing. The ’O-3000 Camera Series Java Demo App’ uses the following procedure to access the camera’s balancing controls: 1
The computer screen can display colors inaccurate if not calibrated.
Page 2
O-3000 Camera Series
• There are four different color gains to adjust, namely red, blue, and two green gains. The adjustment sliders ranges from 0 to 100 percent and can be accessed through ’Settings -> White Balance’. • Because gain amplification is done on sensor itself (bayer color space) please be aware that each set point adjustment in one color range will have an impact on another color range (cross talk). So if one decreases the gain of the blue channel the image automatically reduces blue hue but also increases red hue. • To carry out the white balancing, aim the camera at a surface of a consistent grayish color (white paper may be bleached which gives a fluorescent tinge). • Adjust the sliders in the ’O-3000 Camera Series Java Demo App’ that the image on the screen matches the color of the scene. Start with the sensor’s default values2 . An increase in color gain may lead to an increase in undesired noise. Just amplify color gain as high as needed with the individual sliders. • Under the conditions the balancing process has been completed, you will receive photographic results neutral to the human eye.
3
Typical Values
We achieved best results for different conditions with the parameters listed below. Color gain parameter listed as (red / green relative to red / green relative to blue / blue). Acquisition parameters constantly set to manual exposure mode with 5’000 µs exposure time and 20 % sensitivity. • Indoors, halogen lamp: (12.5 / 12.5 / 12.5 / 12.5) • Indoors, fluorescent lamp: (12.5 / 12.5 / 12.5 / 25) • Outdoors, overcast weather conditions: (12.0 / 12.5 / 12.5 / 12.0) • Outdoors, clear sky, sunny: (12.0 / 12.5 / 12.5 / 10.0)
4
Notes
Please note a few restrictions: 2
Default values for fluorescent light will be set during camera start up procedure (when power up the camera).
Page 3
O-3000 Camera Series
• The process of white balancing is only valid in the same lighting scene. Even slight changes in lighting can cause the picture to have a different tone. • Up to firmware 1.0.9 white balancing is only possible in manual exposure mode. From version 1.0.10 it is possible also in automatic exposure mode. • Make sure the two sliders for green (relative to red and blue) are adjusted simultaneously or with slight differences only in order to get the best results. • The camera keeps its settings until reconnect (power off/on). • Better results can be achieved if there is a more sophisticated demosaicing algorithm. Additional image processing procedures, such as color, aperture and gamma correction, may help as well to increase image color quality.
Page 4
