mt9v125 aptina, mt9v125 Datasheet - Page 13

• Manufacturer Part Number: mt9v125
• Description: 1/4-inch System-on-a-chip Soc Vga Ntsc And Pal Cmos Digital Image Sensor
• Manufacturer: aptina

Defect Correction
Color Correction
Color Saturation Control
Automatic White Balance
Auto Exposure
PDF: 09005aef829db8a0/Source: 09005aef829dbfdb
MT9V125_LDS_2.fm - Rev. A 4/07 EN
This device supports 2D defect correction. In 2D defect detection/correction, pixels with
values different from their neighbors by greater than a defined threshold are considered
defects unless near the image boundary. The approach is termed 2D, as pixels on neigh-
boring lines as well as neighboring pixels on the same line are considered in both detec-
tion and correction.
To obtain good color rendition and saturation, it is necessary to compensate for the dif-
ferences between the spectral characteristics of the imager color filter array and the
spectral response of the human eye. This compensation, also known as color separation,
is achieved through linear transformation of the image with a 3 x 3 element color correc-
tion matrix. The optimal values for the color correction coefficients depend on the spec-
tra of the incident illumination and can be programmed by the user.
Both color saturation and sharpness enhancement can be set by the user, or adjusted
automatically by tracking the magnitude of the gains used by the auto exposure algo-
rithm.
The MT9V125 has a built-in automatic white balance (AWB) algorithm designed to com-
pensate for the effects of changing scene illumination the color rendition quality. This
sophisticated algorithm consists of three major sub-modules:
• A measurement engine (ME) performing statistical analysis of the image
• A module selecting the optimal color correction matrix
• A module selecting analog color channel gains in the sensor core
While the default algorithm settings are adequate in most situations, the user can repro-
gram base color correction matrices and limit color channel gains. The AWB does not
attempt to locate the brightest or grayest elements in the image; it performs in-depth
image analysis to differentiate between changes in predominant spectra of illumination
and changes in predominant scene colors. Factory defaults are suitable for most appli-
cations, however, a wide range of algorithm parameters can be overwritten by the user
through the serial interface.
The auto exposure (AE) algorithm performs automatic adjustments to image brightness
by controlling exposure time and analog gains in the sensor core, as well as digital gain
applied to the image. The algorithm relies on the auto exposure measurement engine
that tracks speed and amplitude changes in the overall luminance of selected windows
in the image.
Backlight compensation is achieved by weighting the luminance in the center of the
image higher than the luminance on the periphery. Other algorithm features include:
fast-fluctuating illumination rejection (time averaging), response-speed control, and
controlled sensitivity to small changes.
While the default settings are adequate in most situations, the user can program target
brightness, measurement window, and other parameters as described above. The auto
exposure algorithm enables compensation for a broad range of illumination intensities.
MT9V125: 1/4-Inch VGA SOC Digital Image Sensor
13
Micron Technology, Inc., reserves the right to change products or specifications without notice.
Detailed Architecture Overview
©2007 Micron Technology, Inc. All rights reserved.