Some history on video.

The first camera tubes were invented in the 1920's and were called the Iconoscope, a later version was called the vidicon tube. Up until the CCD chip became popular the vidicon tube was the standard way to convert an image to an electronic signal. The vidicon tube was a vacuum tube that had a photosensitive screen inside of it that your image was focused onto. An electron beam scanned this screen (usually made of a phosphorus coating) and converted any part of the screen that was illuminated to an electrical signal one point (called a pixel) at a time. This point by point image was then reassembled by your tv tube (cathode ray tube, CRT) that performed the reverse operation by scanning the front of the CRT with an electron beam and excited one pixel at a time to match the transmitted picture.

After World War II the United States set up a committee called the National Television System Committee (NTSC). Their job was to set a standard that all video signals must be generated in so that tv sets, cameras and tv transmissions could be mass produced and still be compatible with each other. They picked the specifications already developed by Radio Corporation of America (RCA). I will discuss the NTSC standard later.

The original cameras were only black and white. Color cameras were developed by Columbia Broadcasting System (CBS, the fine sponsor of Late Night with David Letterman!). So the committee (NTSC) reconvene again in 1953 to establish the NTSC standard for color video. Color video was accomplished by using three vidicon tubes with red, green and blue filters in front of them. The color CRT likewise has a red, green and blue electron gun that then recreates the image sent electronically by the vidicon tube.

Video technology today.

Todays video cameras use an imaging device called a CCD, short for Charge-Coupled Device. These imaging devices efficiently convert up to 60% of the photons that strike it to a signal.

CCD chip

The CCD, Charge-Coupled Device, is an array of pockets (called pixels). It is an integrated circuit silicon chip in an array of squares on average with 600 square pixels by 800 square pixels to a side. The illustration below shows an array of 8 horizontal pixels and 3 vertical pixels. In a camera an image will be brought to focus on the array. The photons from the light source (the image) will strike the array and cause a charge to build up in individual pixels that is directly proportional to the number of photons that strike it. In other words, the brighter the object the more photons that will strike a pixel and the higher the charge will build up in the pixel.

                                  ######## Row 3
                                  ######## Row 2
                  Electronics <---######## Row 1
                               

There are different types of CCD chips and thus different methods of extracting the image from the array, but here is a typical method of retrieving the image. After the time for the exposure is up, the pixels will start passing their charges down one row at a time. For instance, Row 1 will pass its charge out its side to the electronics. Row 2 will pass its charge down to Row 1 which will then pass the charge out of its side. While this is going on Row 3 will pass its charge to Row 2 which will then pass it to Row 1 which will then pass it to the electronics.

Each one of the pixels has an X,Y position and an intensity. The electronics in the video camera takes all of that and inserts it in the NTSC standard video signal which comes out of the camera and can either be displayed directly on a monitor or recorded by a VCR. The TV tube reconstructs the image by putting a bright spot in the same X,Y position with the same intensity as the CCD array saw. In an integrating CCD camera these pixels are sent digitally down a connection to a computer.

The NTSC standard.

• Video cameras generate 30 pictures or frames each second (usually abbreviated as fps). This means that one frame represents 1/30'th or 0.033 seconds in time resolution.
• However, the time the camera spends actually taking each picture (what photographers usually call the integration time) is anywhere from 1/60'th of a second to 1/10,000'th of a second. It uses the rest of that 0.033 seconds per frame to convert its picture into a NTSC standard video signal.
• The CCD chip used in video cameras is more desirable than film in some ways. The CCD has a higher dynamic range, meaning it can distinguish between more subtler changes in color than film. The typical CCD can see the difference between up to 1 million different shades of gray! It is also more sensitive than film is in a wider range of the visible and invisible spectrum of light.

• CCD's have no memory like the phosphorus vidicon tubes had. This is because after one image is integrated the wells of each pixels are completely emptied before the next image begins integrating. This prevents streaking and burn in like the old vidicons were susceptible to.

• Video cameras usually have some type of electronics built in to either adjust the gain (or sensitivity) of the camera (we call this autogain) or the length of the integration time of each frame (we call this autoshutter) in accordance to how bright the object you are viewing is.

• Video cameras have a rating system designed to tell you how sensitive it is to light. You will find this rating in the camera specifications usually called the minimum scene illumination and it is usually rated in lux. The lower the lux rating the less light it needs to be able to display an image. The typical home color camcorder has a low light rating of 2.0 lux.