Mitsubishi P91DW User Manual download pdf (Page 93)

Alpha Innotech User’s Manual, version 4

4-18

Noise Filters

These filters are particularly effective in eliminating random noise contained in an image

and produce less blurring than the smoothing filter described below. This filtering

process uses the values of the pixels contained in the area surrounding a pixel to

determine the new value given to the pixel of interest. The noise filter sorts the pixels in

the neighborhood into ascending order and picks the middle or median pixel value as the

new value for the pixel of interest. 3 levels of noise reduction are available

Smoothing Filters

These filters are very useful for reducing the visual noise present in an image. When a

smoothing filter is applied to an image, rapid changes in intensity are averaged out with

the remaining pixels in the neighborhood, thereby decreasing the high frequency content.

The visual result is a slight smoothing of the image because sharp pixel transitions are

averaged with their surroundings.

Smooth High has the largest effect on the image. Smooth Med has an intermediate

effect. Smooth Low has the most subtle effect on the image.

Edge Filters

These filters use Laplacian enhancement to highlight edges, regardless of direction. All

edge-enhancement operations attenuate the low frequencies of the image. Regions of

constant intensity or linearly increasing intensity become black as a result of these

transformations, and regions of rapidly changing intensity values are highlighted.

Note: The White level may need to be adjusted after using the Edge filters in order

to see the result of this filtering process.

Horizontal Edge Filter

This filter brightens horizontal edges. This can be useful in pinpointing bands on a gel.

The horizontal edge filter (Horz. Edge) enhances image edges by shifting an image

vertically by one pixel and then subtracting the shifted image from the original. In an

area of constant pixel intensity, the subtraction yields black pixel values. At an edge,

which is an area with large changes in intensity, the subtraction yields light-colored pixel

values. The larger the difference in intensities, the lighter the resultant pixels.

1 2 ... 88 89 90 91 92 93 94 95 96 97 98 ... 205 206

No comments

Mitsubishi P91DW User Manual Page 93