{"id":7048,"title":"Principles of Electronic Image Processing - Switch, Mix, Compare","kind":"biblio","url":"https://www.videohistoryproject.org/principles-electronic-image-processing-switch-mix-compare","version":"64aaaebb6696718d379f4601333c5bc60deac3a725b81c41f386a52bccaee652","modified":"2026-09-13T21:51:53.8870279+00:00","credits":[{"role":"Author","name":"Sherry Miller Hocking"}],"authorKeys":["Sherry Miller Hocking"],"subjects":["Tool Texts"],"metadata":[{"label":"ETC archive","value":"ETC1045","url":null}],"citations":{"page":"\u0022Principles of Electronic Image Processing -  Switch, Mix, Compare.\u0022 Video History Project, Experimental Television Center, 2026. https://www.videohistoryproject.org/principles-electronic-image-processing-switch-mix-compare.","publication":"Sherry Miller Hocking. Principles of Electronic Image Processing - Switch, Mix, Compare. 1978-80.","risUrl":"https://www.videohistoryproject.org/citation/7048.ris"},"links":[{"label":"Scanned document","url":"https://archive.org/details/ETC1045","relationship":"linked document; contents not retrieved","contentType":"application/pdf"}],"sections":[{"name":"body","source":"VHP record body","generated":false,"characters":24156},{"name":"abstract","source":"recorded publication abstract","generated":false,"characters":84},{"name":"summary","source":"existing VHP generated summary","generated":true,"characters":594},{"name":"archive_summary","source":"existing ETC archive summaries: ETC1045","generated":false,"characters":84}],"content":{"section":"body","source":"VHP record body","generated":false,"text":"An imaging system can be thought of in terms of three functions: image or signal generators, signal processors and signal controllers. Image, or more accurately signal, generators produce optically-based signals from cameras or non-optically based signals from oscillators. Image processors are devices which alter a signal from these sources in different ways; each processing module provides a means by which a single parameter or set of parameters of an image can be changed. Processing modules within a system are often arranged so that the output of one, a keyer, for example, can serve as an input for a second, for example a colorizer. Signal processors act on the signal after its initial generation and before its recording or display. The image or signal controllers often act on the processors; for example, a control voltage in the shape of a square wave may switch color changes in a colorizer from red to green. Further, some processing devices, such as certain colorizers and computer-based systems, are more accurately signal/image generators since they operate without external inputs. Switching Switching is a basic kind of signal processing from which other imaging techniques such as mixing and keying, or signal comparing, are derived. Switchers were the first commercially developed processing or \u0022special effects\u0022 systems, appearing prior to the mid-1950s. The motivation for their development lay in the desire to switch or \u0022cut\u0022 from one camera to another. Switching then can be defined as the serial presentation of image and/or sound derived from 2 or more sources and is achieved by the sequential changing from one input signal to another. As we will see, the term \u0022serial\u0022 can be misleading; as the rate of switching is increased beyond a certain frequency, we no longer see one image following another. Instead the image we see might contain two or more source images simultaneously. A switcher is a device which turns a signal on or off. It completes a circuit, allowing the signal to flow through a circuit from the input to the output. If the switch is on, the circuit is complete; if the switch is off, the circuit is broken, and the signal cannot travel from the input to the output. Switchers, then, have 2 or more inputs; the input signals, or images, may come from cameras or from oscillators or any other sync-compatible source. The switches in a switcher allow you to choose which of these inputs will be allowed to pass to the output, for example a monitor, at any one time. At its simplest level, a switcher can employ a mechanical switch, such as a toggle switch; by manually changing the position of the switch, the signal of one of the 2 inputs is allowed to pass because the circuit path is complete. In Figure 1, the switch position of camera 2 allows that signal to go to the monitor. A series of push-button switches works in a similar way; only one can be pushed at a time, and when it is pushed a circuit is completed between the input and output. In the push-button system of Figure 1, when button 1 is pushed, the signal from camera 1 goes to the monitor and the signal from camera 2 does not. These mechanical switching systems illustrate a 2 input/1 output switcher. With a larger number of inputs, a multi-position switch might replace the toggle or on/off type; these have a number of on positions and eliminate the need to turn off an input before turning a second one on. Many switchers also have more than one output; an example might be a multiple monitor matrix where each monitor displays a different camera image. As the number of inputs and outputs increases, these mechanical switching systems quickly become cumbersome. For example, if you have 10 inputs and 5 outputs and want any input to be available at any output, to be able to display any camera on any monitor, then you need 50 switches. In general there are two kinds of switchers. In a \u0022 broadcast \u0022 type, there are any number of inputs and certain of them are permanently connected to a single specific output. Each bus is a series of inputs; in Figure 2 there are 5 inputs on each of the two busses, A and B, while one output is available. On each bus you can select one input; for example, on bus A you choose input 5 and on bus B you choose input 2. If the switcher is a push-button type, you can switch between 5 and 2 or any other combination of inputs by pushing the appropriate buttons. Only one signal comes to the output at any one time. If a fader is present before the output, you are then able to preselect a signal image on the A bus, one on the B bus and then fade between A and B; while the image from the B bus is then on the monitor you are able to select another image on the A bus and when you fade back to A you have a third image on the monitor. In the \u0022distribution\u0022 type switcher, any input can be sent to any output. Because of the large number of cross-points, this system gets bulky and the problem is one of control. Manual switching of this large number of points would be time-consuming and confusing. Most image processing systems using distribution switching use a second system to electronically control these switching points. The transition or timing of the switching is important. If you have to turn the first signal off before you turn the second on, as would be the case with a simple switcher using two position toggle switches, it is referred to as a gap switch. This type of transition will result in a blank screen at the point of switching as can be seen in the signal represented in Figure 4. If the first is turned off after the second is turned on, it is referred to as a lap type or \u0022make before break\u0022 switcher. The point in time at which the switching occurs is also important and is directly related to the sync or timing of the incoming signal. Historically switchers began with simple switches; each camera was running on internal sync so there was no single common timing relationship among the camera inputs. Running on internal sync meant that each camera was beginning and ending its scanning process at a different time. Without a common external sync source insuring that all cameras began to scan their separate images at the same time, the output image would break up at the switching point. The switch could occur when camera 1 was 2/3 through the scanning process while camera 2 had just begun. The development of externally syncable vertical interval switchers prevented this picture disturbance. All cameras were externally locked to a common sync source, all beginning and ending their scans at the same time; with this it was possible to define and locate switch transitions at a point in time within the signal, called the vertical interval, which occurred at precisely the same instant for all camera inputs and which would be invisible since there is no picture information carried during the interval. Thus the problem of picture disturbance at the switching point was eliminated. Up to this point we have talked about cameras or oscillators as switcher inputs; a single input can also be the output of another device, for example, an SEG or mixer, which themselves contain multiple inputs. When the outputs do not loop back and themselves become inputs to that output, the switcher is referred to as a \u0022single re-entry\u0022 type. An example, shown in Figure 5, indicates 3 cameras as inputs to an SEG the output of which goes to a mixer, and that output is displayed. A \u0022double re-entry\u0022 type indicates that the output of one device, fed to a mixer, is used as an input to that device; that input, then, potentially contains mixed or layered images of several cameras as a single image which can then be mixed again with the other single camera images. Sequencing A sequencer may be thought of as a specialized kind of switcher. The order of the switching can be repeated and is often programmable so that the series can be repeated indefinitely. The order is changeable so that the switching is not necessarily consecutive from input 1 through input 5, for example, but may be pre-programmed so that you can jump from one to another in any order and repeat that order. Sequencers also often contain a clock which controls the speed of the sequence. A signal pulse from the clock turns the first input off and then on when the level of the signal voltage reaches a certain point; the clock contains an oscillator the frequency of which can be changed, thus changing the speed of the sequencer. Sometimes sequencers have a specialized input for accepting a signal from an external clock. An external oscillator puts out a pulse which controls the speed of the sequencer. This type of device is referred to as voltage controllable because it takes in a signal from another device and uses it to control a parameter of image processing, in this case the rate of sequencing. Sequencers can be banked or cascaded, that is, arranged in a configuration where the output or clock of the first triggers the second. Switching thus began as a technique for cutting among a number of video images and thus allowed a number of points of view to be presented sequentially. Switching and sequencing are integral to video because they function within the timing limits of video imaging. As techniques of image processing they are frequently used with \u0022real time\u0022 images. In contrast, editing presumes that the images are recorded. If the switcher is able to accept videotape recorders as inputs, this sense of \u0022real time\u0022 sequence may be false, since \u0022recorded time\u0022 events are intercut with live events. Mixing Mixers were developed in the 1950s as special cases of switchers. As with switchers, mixers used 2 inputs, but the images were combined in an additive way rather than an either/or way. In a mixer, the input image signals are added together and then, in effect, divide or scaled down so that the resulting signal output is within the prescribed limitations of signal amplitude. Mixing happens in \u0022real time\u0022 so that the output is the instantaneous sum of all the instantaneous amplitudes for each point along each horizontal line of the inputs. Because the 2 inputs are synced to an external source, they begin to scan at the same time; as scanning occurs, each signal contains a series of instantaneous voltages or amplitude levels corresponding to the gray values in the scenes. The voltage levels for each point along the horizontal line are then added together, producing an output signal containing information from both inputs. As the lines are built up into fields and frames, the resulting image is a mix of both source images. Figure 6 illustrates the process. Mixers thus provide the means for the development of superimposition. Mixers often contain slide pots which attenuate or vary the \u0022amount\u0022 of the input signals which are added together. Fading then is a variation on mixing and switching. Rather than an abrupt or instantaneous cut or switch from one signal to another, fading allows a gradual increase or decrease in signal levels, resulting in an image containing elements of 2 inputs with one becoming increasingly predominant (fade in) while the other either remains (superimposition) or gradually disappears (fade out). With the slide pots, the input signal voltage presently on is decreased in amplitude, while the off input signal voltage is increase in amplitude. In a \u0022lap dissolve\u0022, both images are superimposed for a time while one is decreased and the other increased; in a \u0022fade in/out\u0022 the first is gradually reduced to zero amplitude (no image) before the second is gradually increased, so no superimposition results. So far we have talked about switching which occurs during the vertical interval of the signal and at a rate which is slow enough for the perceptual system to register two or more discrete images being sequenced in time. Obviously video is capable of switching at rates much faster than these. Superimposition will appear if the rate of switching is faster th","offset":0,"nextOffset":12000,"totalCharacters":24156,"complete":false},"rightsUrl":"https://www.videohistoryproject.org/terms-service"}