White Balance, Shading, and Pixel Corrections

JAI cameras offer a variety of correction and compensation features to help produce high-quality, accurate images. Some of the key features include:

• PRNU/DSNU Compensation
• Flat/Color Shading Correction
• White Balance

While these features are highly useful, their function and effects are not always intuitive.

PRNU/DSNU Compensation

PRNU (Photo-Response Non-Uniformity) and DSNU (Dark Signal Non-Uniformity) compensation are used in line-scan cameras to correct small variations in the way individual sensor pixels respond to the same lighting conditions. 

• DSNU (PixelBlackCorrect) calibrates how each pixel responds under completely dark conditions.
• PRNU (PixelGainCorrect) calibrates how each pixel responds to the same intensity of light.
   

Flat Shading Correction

Flat Shading Correction is a block-level, macro adjustment which means that each block consists of a group of pixels, and the correction adjusts output levels to produce uniform brightness across the entire sensor area.

This compensation method is designed to produce a flat, equal response to light. It is generally considered a coarse correction for variations in image brightness, which typically result from optics-related shading issues caused by the lens and/or prism. 

Unlike PRNU/DSNU, Flat Shading does not correct pixel-level variations, but it is highly effective at creating an evenly lit image across large areas. This makes it particularly useful for applications where consistent illumination is essential.

Color Shading Correction

Color Shading Correction is another block-level, macro adjustment. Each block consists of a group of pixels, and the correction works by adjusting Red and Blue channel outputs within the block so their proportions match the Green channel.

After correction, the relative proportions of Red, Green, and Blue channels are identical within each block, though overall levels may still differ. The color correction helps achieve consistent color representation across the image, which is especially valuable for high-precision or industrial imaging applications.

White Balance

White Balance ensures that red, green, and blue pixels respond correctly to the light source. Color cameras usually use a Bayer filter, which places a pattern of red, green, and blue filters over the sensor’s pixels. Because different light sources contain different color mixes, the camera may perceive colors inaccurately if White Balance is not applied. This typically looks like a yellowish color on a background that should be white. 

To correct this, the camera is pointed at a smooth white card or surface (illuminated below the saturation point). Gain is then applied to the pixels so that all pixels reach the same value as the color channel with the highest output, typically green. After correction, the colors in the image should resemble those seen in the real world by the human eye. 

We recommend the following order of image adjustments: 

PRNU/DSNU (pixel-level smoothing for linescan cameras) → Flat Shading (brightness uniformity) → Color Shading (block-level color uniformity) → White Balance (final global color accuracy).