noise filter

As a noise filter is a feature that can filter out specific frequencies from a data base such as an image or a data stream in order to eliminate or to process them. In the signal processing, they are mapped as mathematical functions and technical devices implemented in an electronic filter. Corresponding filter functions can also be in areas of mechanics, eg acoustics or hydraulic occur, where they are often called noise filter.

Areas of application

Optics

The effect of optical sensors and sensitivity can be improved if, prior to the sensors optical filter are switched, scatter or absorb certain spectral components. The filters are tuned to the area to be measured. For IR sensors, for example, Used filters that transmit only a small narrow frequency range of the light. All other effects of light, such as light reflections from objects or unwanted portions of the sunlight that during image acquisition were as scattered light to the sensor and would be manifested as noise are ignored by it.

Acoustics

In the field of microphone technology certain soft materials are used to insulate sound above the measurement frequency so that it can not have a disturbing effect next acoustic lenses. These are, for example, from 20 kHz to 25 kHz active when man can no longer hear. Reached these sound waves in the microphone capsule, it would cause artifacts.

Electronics

In the electronic audio equipment noise filter is used to improve the signal quality as well as the subjective listening pleasure. The noise reduction and the elimination of noise signal can be distinguished from each other. In the simple noise suppression using a " denoiser " the signal below a certain level is masked out, so strictly speaking there is no filtering. A real Rauschelemination is only possible with complex algorithms in signal processors. A simple noise filtering is possible by cutting off frequencies above an arbitrary frequency used by a low-pass filter.

Image processing

In image processing, noise filters play an important role. You eliminate unwanted disturbances after rescaling and random technical influences such as interference and allow an analysis and processing of the image. By estimating and processing of the noise statements about details and objects can be made into an image.

Example of image enhancement by means of digital signal processing by superimposing a plurality of source images, the processing of a noise image, and mixing in a sharper picture. Despite under-exposure of the dark source images, which has led to a high amount of noise in the picture, is still an acceptable result can be achieved.

Brightened source image 2 of 10 with a lot of noise

Averaged source images as the sum lightened image with less noise

Prozessiertes denoised image as a subpicture

Average final image with optically less noise than 2

Examples

• A surface acoustic wave filter
• Low-pass