| |
Luminance Uniformity   |
Luminance Uniformity Overview
This test is usually performed to check for non-uniform illumination or lens shading. The lens and sensor array will have a non-uniform response. This can result in a number of effects, the most prevalent of which is dark corners. A correction algorithm is implemented in some sensors to improve this characteristic.
Relative illumination is characterized in one of two ways. It is the illumination at any point, with reference to the maximum illumination. Generally, relative illumination is reported as a percent. Relative illumination is the percentage that the two areas have in common. Shading is the percentage that they differ. The test target typically used for this test is a uniform grey target with uniform illumination.
Relative Illumination = (Measured Intensity)/(Maximum Intensity)*100
Shading = 100 – Relative Illumination |
|
 |
Luminance Uniformity Algorithm
The Luminance Uniformity Algorithm first divides the image up into 9 regions. The pixel intensities within each region are then statistically analyzed and the mean, standard deviation, minimum and maximum are reported.
The region sizes are user configurable, with the center region being circular and the outer regions rectangular. The size of each region is controlled through one common ROI percentage panel control. The figure below shows ROI percentage setting of 33%, 25%, and 10%.
|
| ROI Percentage = 33% |
|
ROI Percentage = 25% |
|
ROI Percentage = 10% |
|
 |
|
 |
|
 |
|
The ratio of the mean intensity of each region to the mean intensity of the center region is calculated and reported. These ratios are then compared to acceptance limits to ensure that the expected fall-off is within reason.
The four corner regions are also compared and the standard deviation calculated. This part of the test is performed to ensure even fall-off in each corner and to check that unacceptable image “tilt” is not present.
The optical center of the image is also calculated during this test. The optical center is defined as the intensity weighted center of mass of the image. |
| |
Luminance Uniformity Input Parameters
Name |
Description |
 |
Input Image |
An 8-bit monochrome luminance image is required for this test. |
|
ROI Percentage |
Specifies the % of the entire images X and Y resolution used for each sub-region. |
|
Uniformity Limits |
Separate limits for the maximum % deviation from center are available for the upper left, upper center, upper right, center left, center right, lower left, lower center, and lower right regions. |
|
Optical Center Limits |
Upper and lower limits for both the X and Y image center location are user configurable. |
|
Luminance Uniformity Outputs
Name |
Description |
|
Uniformity Results Table |
A table report output contains the intensity mean, standard deviation, minimum and maximum for each defined region. |
|
Uniformity Region Ratio Results Table |
A table, containing the ratio of each region’s mean intensity to the center mean intensity, is also output. |
|
Optical Center |
The (X, Y) location of the optical center of the image is also output. |
|
|
| |
| |
 |
| |
| |
|
