contrast

This article covers contrast in a visual context. For additional meanings, see Contrast (disambiguation) .

Contrast ( borrowed from the Italian contrasto 'contrast' to Latin contra 'against' and stare 'standing' ) describes the difference between light and dark areas of an image (it differentiates between light and dark colors). Colloquially, one also speaks of depth of color or "brilliance". The latter designation is misleading because of its deviation from the physical quantity brilliance .

definition

Artificially enhanced contrast: the image looks like a silhouette

Original: high-contrast image

Artificially reduced contrast: the picture looks dull

The contrast is a distinguishing feature for the brightness curve of an image or between two pixels. The range of contrast or dynamics describe the difference in intensity between the brightest and darkest point in an image. In the general case, the contrast is described as a function of the resolution via the modulation transfer function.

The contrast is defined by the maximum and minimum luminance and . ${\displaystyle L_{\text{max}}}$ ${\displaystyle L_{\text{min}}}$

The Weber Contrast (named after Ernst Heinrich Weber ) as: ${\displaystyle K_{\text{w}}}$

${\displaystyle K_{\text{w}}={\frac {{L_{\text{max}}}-{L_{\text{min}}}}{L_{\text{min}}}}}$ With ${\displaystyle 0\leq K_{\text{w}}\leq \infty }$

When the minimum luminance reaches the black level, the contrast is infinite. ${\displaystyle L_{\text{min}}=0}$

The Michelson contrast (named after Albert A. Michelson ) or the modulation as: ${\displaystyle K_{\text{m}}}$

${\displaystyle K_{\text{m}}={\frac {L_{\text{max}}-L_{\text{min}}}{L_{\text{max}}+L_{\text{min}}}}}$ With ${\displaystyle 0\leq K_{\text{m}}\leq 1}$

When the minimum luminance reaches the black level, the contrast is also maximum here, but limited to the standardized value one. ${\displaystyle (L_{\text{min}}=0)}$

In both cases the contrast is zero if the two luminances do not differ. The loss of contrast due to vanishing differences in luminance is called blackout in the dark or whiteout in very bright light conditions . ${\displaystyle (L_{\text{min}}\lesssim L_{\text{max}})}$

The human eye can handle most of the contrast ranges occurring in nature relatively well (the intensities are not perceived linearly, but logarithmically , see Weber-Fechner law ).

When looking at unusual objects in particular, there are physiological and psychological effects on subjective perception . Contrast phenomena can cause optical illusions , but are also involved in the recognizability of the finest line structures . Examples of this are some of the “ Martian canals ” and, in shades of gray, the Mach stripes .

Image display

In analog photography , in addition to the camera lens, it is primarily the film material that determines the range of contrast, and in the case of prints it is also the gradation of the photo paper . In digital photography , it is the performance of the analog-to-digital converter . Newer digital cameras try to expand the dynamic range through a non-linear response behavior (similar to the human eye). Another important factor influencing the contrast range is the selected ISO sensitivity: higher ISO sensitivities usually lead to a lower displayable contrast range. The f-number and exposure time only shift the area, but do not increase the scope.

Photos with a particularly high subject contrast tend to render light parts of the picture even lighter and darker parts of the picture even darker than in the motif shown. The image appears to be rich in contrast to the viewer (i.e. the opposite of "dull"), but shows the viewer fewer details overall, that is, no drawing ( tonal gradation ) is visible in dark (shadows) as well as in light areas (lights) so no details can be seen. If a motif has greater differences in brightness than the digital camera can capture, the camera cannot fully reproduce the tonal range in the image. In this case, it can be better to expose rather tightly in order not to lose the drawing in the lights (e.g. the structure of sunlit clouds). Because eroded lights can hardly be restored, while parts of the image that are too dark can usually still be saved through appropriate post-processing.

In the case of digital post-processing of images, however, a high contrast range is always better, as the contrast can subsequently be increased within wide limits, but can only be reduced again to a very limited extent.

See also : this description of contrast in the sharpness display.

One way of influencing the contrast of digital photos is to take a series of exposures in conjunction with image processing using HDR software.

When evaluating lenses, contrast plays a decisive role. The modulation transfer function describes the course of the contrast, which decreases with increasing spatial frequency and thus also limits the resolution.

From the painting techniques of so-called are seven color contrasts known to frame contrast images.

Imaging procedures (medicine)

An increase in the contrast is often necessary in medical imaging . In many cases, it is done by injecting a contrast agent into the bloodstream before the recordings , for example in angiography or magnetic resonance tomography . The temporal distribution of the contrast agent - i.e. the blood flow to the respective organ or connective tissue - can be checked by means of image series . Another less finely tunable nature of the contrast agent takes place in the diagnostic radiology of the gastrointestinal tract (such as by ingestion of barium sulfate suspensions) or in the computer tomography by iodine-containing contrast medium.

Web links

• Modulation transfer function
• Contrast in the Lexicon of Optics on Spektrum.de

Individual evidence

1. ↑ Markus Bautsch: Modulation , Wikibools Digital Imaging Methods , Chapter General Image Properties , accessed on March 8, 2014
2. ↑ Bernd Leuschner: MTF measurement (PDF; 136 kB), Laboratory for Device Technology, Optics and Sensor Technology, Beuth University of Technology Berlin
3. ↑ See also ISO 12233: 2000 - 3.19 "modulation" and DIN ISO 12231: 2017 - 3.110 "modulation"