Electronic Automated Measurement User System (EAMUS™)

Formal Image Quality Control

The image quality control program automatically provides you with some image features that are related to the “distribution”
of gray values within the whole image. These features include

the “homogenous illumination” (shading),
the equalization of the gray value histogram (gray values from intervals of the same size should be represented by the same
number of pixels), and
the minimum <> maximum range of gray values (gray value normalization, or adjusting the gray value range to 0 – 8 bits).

The quality control program works independent from the chosen measurement program. Its results will be automatically presented
in a separated pdf file (-tco.pdf). The above mentioned three image filters will be applied to the original and to the gradient
image. The obtained filter data of your previously submitted images obtained by the same visualization stain serve for computing
the statistical data of your present image. The computed range and the value of your actual image are listed. Running control curves
that include the data of your previously submitted images are attached too. The segmentation threshold is calculated according to
the methods described by Otsu.

The measurement program will provide you with the following results:

Measurement data

image shading (median): median of gray value differences (of all three colour dimensions) between the original and the
filtered image (shading)

image shading (range): range of gray value differences (of all three colour dimensions) between the original and the
filtered image (shading)

image shading (segmentation gray value): gray value differences (of all three colour dimensions) of the segmentation
thresholds between the original and the filtered image (shading)

histogram normalization (median): median of gray value differences (of all three colour dimensions) between the original
and the filtered image (histogram equalization)

histogram normalization (range): range of gray value differences (of all three colour dimensions) between the original
and the filtered image (histogram equalization)

histogram normalization (segmentation gray value): gray value differences (of all three colour dimensions) of segmentation
thresholds between the original and the filtered image (histogram equalization)

gray value normalization (median): median of gray value differences (of all three colour dimensions) between the original
and the filtered image (gray value equalization)

gray value normalization (range): range of gray value differences (of all three colour dimensions) between the original
and the filtered image (gray value equalization)

gray value normalization (segmentation gray value): gray value differences (of all three colour dimensions) of segmentation
thresholds between the original and the filtered image (gray value equalization)

gradient shading (median): application of the gradient filter and computation of the median of gray value differences (of
all three colour dimensions) between the original (gradient) and the (second) filtered image (shading)

gradient shading (range): application of the gradient filter and computation of the range of gray value differences (of all
three colour dimensions) between the original (gradient) and the (second) filtered image (shading)

gradient shading (segmentation): application of the gradient filter and computation of gray value differences of the
segmentation thresholds (of all three colour dimensions) between the original and the filtered image (shading)

gradient histogram norm (median): application of the gradient filter and computation of the median of gray value differences
(of all three colour dimensions) between the original and the filtered image (histogram equalization)

gradient histogram norm (range): application of the gradient filter and computation of the range of gray value differences
(of all three colour dimensions) between the original and the filtered image (histogram equalization)

gradient histogram norm (segmentation): application of the gradient filter and computation of gray value differences of the
segmentation thresholds (of all three colour dimensions) between the original and the filtered image (histogram equalization)

gradient gray value norm (median): application of the gradient filter and computation of the median of gray value differences
(of all three colour dimensions) between the original and the filtered image (gray value equalization)

gradient gray value norm (range): application of the gradient filter and computation of the range of gray value differences
(of all three colour dimensions) between the original and the filtered image (gray value equalization)

gradient gray value norm (segmentation): application of the gradient filter and computation of gray value differences of the
segmentation thresholds (of all three colour dimensions) between the original and the filtered image (gray value equalization)