Image processor

from Wikipedia, the free encyclopedia
This article or section needs to be revised. Details should be given on the talk page. Please help to improve it , and then remove this flag.

This article or the following section is not adequately provided with supporting documents ( e.g. individual evidence ). Information without sufficient evidence could be removed soon. Please help Wikipedia by researching the information and including good evidence.

The image processor - also known as the image processing engine , image processing system or image processing system - is one of the most important components of video and digital cameras . In addition to the lens and the image sensor , it plays an essential role in obtaining the digital image . The image processor is a combination of hardware ( processors ) and software ( algorithms ).

An important task is z. B. the processing of the sensor data: The photodiodes of most image sensors only recognize shades of gray. To obtain color information, the individual pixels with color filters for z. B. red, green and blue ( RGB ) or CYMK . RGB filters are mostly arranged according to the Bayer matrix , ie there are one red, one blue and two green pixels. Since each photodiode only records one of three color information items, only these red, blue and green pixels would be visible without an image processor. One of the tasks of the image processor is thus the Bayer conversion, colloquially called "debayering" or demosaicing . The image processor also does the non-uniformity correction , i.e. the elimination of the fixed pattern noise , which is caused by different sensitivities of the sensor pixels . Further tasks are color correction: For example, it calculates the correct color and brightness values ​​for each individual pixel using the data supplied by the image sensor on the chrominance ( hue and color saturation ) and luminance ( brightness ) of the individual pixels. The better the algorithms used, the more natural the colors and the more balanced the contrast .

This process of image data processing is extremely complex and involves many different processes. Its success depends essentially on the "intelligence" of the algorithms used. The most important properties of the digital image on which the image processor has a decisive influence are:

Color representation

The image processor evaluates the color and brightness values ​​of each individual pixel and compares the information with that of the neighboring pixels. A complex algorithm calculates the correct color and brightness of the respective pixel. At the same time, the image processor analyzes the entire image to determine the correct contrast distribution. By adjusting the gamma value (increasing or decreasing the contrast range of the midtones of the image), fine color gradients - for example in human skin or the blue of the sky - are displayed more realistically.

Noise reduction

As noise disturbances are called, which occur in all electronic circuits. Image noise appears in digital photos in the form of irregularly occurring pixels that differ in color and / or brightness from their surroundings. The effect is increased by higher ambient temperatures, longer exposure times or higher ISO settings.

At high ISO values ​​(= high sensitivity of the sensor), the electronic signal in the image sensor is increased and thus the noise is amplified at the same time, which reduces the signal-to-noise ratio . The image processor tries to separate image and interference signals from each other and thus suppress the noise. This is particularly difficult in image areas with fine detailed structures. If these are mistakenly regarded and treated as noise by the image processor, they lose their drawing.

Smooth and sharp edges

After the color and brightness values have been interpolated for each pixel , the image processor softens the image in order to compensate for any color deviations in individual pixels. In order to still get a sharp and detailed image, edges and contours are then sharpened. The quality of the result depends on how well the image processor recognizes edges and reproduces them smoothly and without over-sharpening.

speed

The processing speed of the image processor is becoming increasingly important, particularly in view of the steadily increasing number of megapixels in digital cameras. In order to rule out waiting times in the workflow as far as possible, they must also be constantly optimized in terms of their speed. Image processors in modern cameras consist of a combination of ASICs and FPGAs as well as DSPs.

Manufacturer

Here are a few well-known manufacturers of cameras with their own processors:

Web links

Individual evidence

  1. ↑ Bayer filter :: bayer filter :: ITWissen.info. IT knowledge, accessed on July 21, 2020 .