Video camera

Video camera (Sony Betacam SP)

A video camera is a device for taking pictures in the form of electrical signals. In contrast to the film camera , which is based on photographic technology, the electronically stored image signals can be made immediately visible again as images.

A microphone is integrated in consumer and prosumer models to record the sound ; professional cameras (e.g. for broadcasting) have connection sockets instead . Different specialized microphones are connected to this depending on the recording situation. In a broader sense, digital cameras are also referred to as video cameras . Many smartphones can also record videos. A video camera with an integrated video recorder is called a camcorder .

Imager

CCD chip

CCDs represented an important step towards the downsizing of video cameras. In the meantime they clearly surpass the quality of iconoscopes . In the case of particularly high-quality cameras, the image sensors are cooled and thus enable significantly lower-noise images. Cheap cameras do not have an aperture , but regulate the exposure via the charging time.

The larger the area of the image sensor used , the more light can be "captured". This increases the sensitivity of the chip used to light and reduces the image noise that occurs particularly in poor lighting conditions. Common sizes are ¹ ⁄ ₆ ″ , ¹ ⁄ ₄ ″, ¹ ⁄ _3.4 ″, ¹ ⁄ ₃ "or ¹ ⁄ ₂ " (1 ″ = 2.54 cm). The resolution of the image converter does not necessarily say anything about the resolution actually offered. The pixel information on the housing mostly refers to the pure photo function. The number used for video recordings is usually not emphasized so clearly.

Some of the specified pixels are often used for the digital image stabilizer. The only interesting thing here is the net amount of pixels that is actually used for filming. Usually these values are in the megapixel range or just below. 800,000 pixels is a common value for PAL cameras. In any case, these can only save 720 × 576 pixels (European PAL standard). With high-quality cameras and professional devices for film production, such as HDCAM , three ² ⁄ ₃ ″ sensors are used, each with a resolution of well over 2,000,000 pixels.

CMOS sensors

This type of image sensor, also known as an active pixel sensor , uses CMOS technology. In the beginning, these sensors were mainly found in very cheap cameras. However, after the further development that has taken place in the meantime, they are also used for demanding observation tasks and for image processing. The CMOS chip records the image continuously, so it can be read out at any time. The number of images per second depends on how high the pixel frequency and the number of pixels of the image section read are, but is higher than with the CCD chip. The functionality of individual pixels can be programmed and read out individually or in groups. The dynamics (the range between the weakest and the strongest still perfectly recorded signal) of the CMOS chip is significantly higher than that of the CCD chip, extreme lighting situations (e.g. car headlights faded in at night in an unlit tunnel) can be achieved with a precision that has not been achieved before being represented. The so-called smear effect hardly or not at all occurs with CMOS sensors. In addition, the CMOS technology is characterized by low power consumption and high image transmission rates (up to 300 kb / s compared to 100 kb / s with CCD technology). Despite these considerable advantages, CMOS technology is not superior to CCD technology in every respect.

Nipkow disk

The first video cameras were based on the so-called Nipkow disk . This is a round disc with about 30 holes arranged in a spiral. The image is projected onto a rectangular area of the pane. The path of the holes in this area causes the image to be scanned line by line: a hole always travels in one direction to disappear at the edge. Then the next hole appears a little further down and scans the next line. All the light passing through provides the video signal and is recorded by a fast photo receiver ( photo cell ).

Iconoscope

The iconoscope (Greek) is a television tube that was invented by Vladimir K. Zworykin in 1923 and that contains a layer of microscopic photocells that are scanned by an electron beam. It replaced mechanical scanning processes. Iconoscope technology was later improved and used until the 1990s.

Vidicon

The Vidicon system also works with a cathode ray tube and is still used today in special applications (medicine, places exposed to radiation).

More information on image scanning electron tubes can be found in the article Image pickup tube .

Entry into the film business

Professional video camera at a rock concert

From the year 2000, camcorders aimed at film production became available, the most important group being the devices according to the HDCAM standard. As a system, they cost five to six-figure sums and have been used by several directors, producers and cameramen. These digital cinema cameras differ enormously from their counterparts for TV production and private users.

CMOS sensors with over twelve megapixels are now used for digital cinema cameras, and the sensors are S-35 mm film frame size. Examples are ARRI D-20, Dalsa or RED .

Color separation

To generate a colored video signal, you need three color components (red, green, blue). The method is also known from color theory as RGB or additive color mixing .

Separation into 3 monochrome images

The most obvious way to achieve color separation is to use three image recorders for one color each (“three chippers”) and to couple them via an optical system that also does the color separation with filters . In the age of the Vidicons, the procedure was common. Today's 3-chip cameras distribute the light falling through the lens via prisms or splitter mirrors onto three monochrome CCD chips. Color filters ensure that one chip picks up the green parts, the other two each red or blue. This process leads to very high quality images and is therefore used in the professional field.

Color separation at the pixel level

A different method is used, particularly with inexpensive cameras (“one-chipper”). The image pickup chip alternately carries color filters of the respective color in front of each pixel, so that different adjacent pixels absorb different color components. Electronics generate a color video signal from this. The higher number of pixels required (2 × green, 1 × red, 1 × blue per pixel) usually leads to a poorer resolution; in any case, however, the color representation is not as precise as with 3-chip cameras. One image recorder that uses this method is the Bayer sensor .

3-chip MiniDV camera Panasonic "NV-GS500" (with separate wide-angle attachment)

Sequential color separation

Before that, they experimented with sequential color separation. A rotating color filter alternately filtered out the three color components. If the same filter rotates synchronously at the receiver in front of a glowing white display tube, a color image is created there for the eye. These attempts were soon stopped. In the space to such color filters still used, since many frequency ranges are needed.

Comparison of 1-chip and 3-chip CCD processes

According to the PAL format, 720 × 576 pixels are displayed on the television screen. In order to save the complete information in the camcorder, the chip would have to have at least 414,720 × 3 pixels. Since the pixels in PAL are elongated, 768 × 576 of the always square CCD / CMOS pixels are actually required, i.e. a total of 442,368 × 3 pixels, which are then converted to 720 × 576. For 16: 9 recording, 1024 × 576 = 589,824 × 3 pixels are required, which are also converted into 720 × 576 pixels, but anamorphically compressed. (Note: Since image sensors usually have an aspect ratio of 4: 3, 16: 9-capable cameras actually have to have image sensors that are 1024 (horizontal resolution of 16: 9 PAL) × 768 ( ³ ⁄ ₄ of 1024) = 786,432 pixels. If the sensor has fewer pixels, one can assume that a real 16: 9 picture is not generated, but a section of the 4: 3 picture is scaled (blow up). The number must be multiplied by three , because a pixel cannot perceive the complete color information, but only differences in brightness. With a 3-chip model, the colors are broken down into red, green and blue ( RGB ) using a prism and distributed over the three chips. Without taking a digital image stabilizer into account, this number of pixels would be sufficient to reproduce all the required image information on a television screen.

With a 1-chip model, image storage is done differently. Since each CCD pixel perceives only differences in brightness, a color filter ( Bayer filter ) with either green, red or blue is set in front of each pixel . The DV signal is recorded in a ratio of 4: 2: 0 ( YCbCr color model). Y is the luma component, which only stores the brightness. U and V denote the color difference components (chroma). This means that the brightness is saved for each pixel and only one color value is saved for four pixels together. Since the human eye reacts much more sensitively to differences in brightness than to differences in color, this reduction can be carried out without major losses.

Mathematically, this results in a pixel set of 720 × 576 × ³ ⁄ ₂ = 622,080. It has to be multiplied by three, since three pixels are required for each color. The reduction (4: 2: 0) in the YCbCr color model again saves half (division by 2).

If a digital image stabilizer is used, the required number of pixels can be increased by 60% or more.

The comparison shows that a 1-chip model can meanwhile produce the same qualities as a 3-chip model. With the 1-chip model, the rather complex optical image division is saved. With ever higher resolution sensors - in 2004, chips with more than 8 million pixels were not uncommon in digital photography - 1-chippers can certainly compete with 3-chip arrangements.

With the same chip size, a higher resolution is always associated with increased image noise. In unfavorable lighting conditions, the situation can arise that a lower-resolution camera with a larger chip delivers a better image than a high-resolution camera with a smaller chip. A rule of thumb for buying is "take the largest imager you can get". In 2005 this is ¹ ⁄ ₃ ″ to ² ⁄ ₃ ″ in the consumer to prosumer area .

Image stabilizer

Electronically: A converter chip is used here, which compensates for annoying jerks by moving a smaller image section (reading window) on the large chip area. The net chip resolution, however, is significantly smaller than the total chip resolution often exclusively specified by the manufacturer.
Optically: The optical image stabilizer is preferable to digital stabilization because the full resolution of the image converter is available for the recording. Here, the compensation takes place through mechanically (e.g. with magnetic fields) moving "floating" lenses, which counteract the unwanted movements under sensor control. Optical image stabilizers used to be only found in the upper price segment, but since the 2010s there have been high-performance optical image stabilizers in very inexpensive digital cameras and camcorders. The sensor shift technology is also part of the optical stabilization. In this case, no lens, but the sensor located in the camera body is moved to compensate for the camera movements. This means that (almost) any lens can be used, especially with system cameras, because no compensating lens is required within the lens.
Combined

lens

The lens used is the most important component to achieve good results. Even the best electronics will not be able to compensate for what is already “spoiled” in the lens (for example blurring, shadowing of the edges, distortion, chromatic aberrations ). The lens not only has to project the required resolution and sharpness onto the image converter, but at the same time must also generate very little or no scattered light. This is achieved through the coating of the lenses, recognizable by a bluish or - more rarely - brownish shimmer of the front lens. Zoom lenses are used for optical zoom ; there are practically no more video cameras without optical zoom. A digital zoom is just an electronic enlargement of the image sent by the lens to the imager. However, the pixel structures are enlarged in the same way as the actual motif. The resolution is reduced depending on the magnification up to the point where the digitally zoomed image is unusable. In contrast, the optical zoom always offers the same resolution with different zoom factors . Because wide-angle lenses are more elaborately designed than other lenses, the wide-angle and macro range of video cameras is usually only very small. Although this can be remedied by using additional lenses, the image quality is fundamentally worse. Usable wide-angle attachments are very expensive. The - if available - automatic focus control can be impaired by attachments on the lens.

Designs

There are many different types of camcorders, which are determined by the purpose and price. Here are the most important distinguishing features:

Carrying and posture design: A distinction is made here between shoulder cameras that are carried on the shoulder and hand cameras that are held in front of the body. Shoulder cameras have a formation on the underside for resting on the shoulder and a corresponding focus on the shoulder. The shutter release is usually located on an attachment in the front right, with which the camera is stabilized with the right hand. The viewfinder is attached to the side. Handheld cameras are usually more compact and lighter, with the viewfinder usually at the back of the camera and the shutter release in different places, based on the manufacturer's assumptions about how the camera should be held.

Recorder attachment: In the majority of camcorders, the camera head (with recording electronics and control unit) and the recorder part form an integral unit. In the case of more sophisticated cameras, however, there are models in which the recorder is attached to the head, and which can therefore be used to record on different media by exchanging the recorder.

Change in handheld cameras: The more recent development (as of 06/2012) has led to even system digital cameras, bridge cameras and even “pocket” or “traveler cameras” (often with “super zoom capabilities”) becoming the "Video cameras" can be counted. AVCHD-lite can handle even relatively cheap cameras, whereas higher quality models with FullHD up to 1920 × 1080 pixels with 50p can come up with them.

Electrical and software interfaces

There are different levels of preprocessing of the image signal from a video camera. The serial output is common to the analog and digital data formats.

Analog output

Line-by-line scanning and output of the brightness information of all (black / white) or the three individual colors as well as the synchronous signals (line and image synchronous signal):

Transmission of all information on one line (mixed signal or composite video , PAL- coded)
Output of the three color intensities and the synchronous signals on separate lines

Digital edition

Serial digital data interfaces for video cameras are z. B. FireWire (IEEE1394, with Sony: iLink) or USB , as well as for the transmission of uncompressed data via SDI and HD-SDI or of compressed data SDTI . The data can also be played on a memory card , which can then be removed and read out.

There are adapters for converting analog video signals into digital data formats.

Software interfaces for a video signal on the computer are used to display, further process or save the image sequences (videos), examples: SANE , TWAIN .

Web links

Commons : Video cameras - collection of pictures, videos and audio files

Wiktionary: video camera - explanations of meanings, word origins, synonyms, translations