.

.
Dies ist eine Tabelle aus Deutschland. Hier werden die nominellen Zeilen-Normen spezifiziert. Von diesen "nominellen" 525 oder 625 Zeilen werden sowieso nicht alle übertragen, sondern nur die sichtbaren Zeilen. Und dann ist das immer noch kein Maß für die effektive Resolution, die Auflösung in "Linien", die am Ende wirklich sichtbar und unterscheidbar im professionellen Studio-Monitor bzw. im Heim-Fernseher zu sehen sind.
.

The television system used for almost all broadcast and closed circuit television in the United States is the system described by the National Television System Committee (NTSC). This standard is described in EIA standards RS-170 for broadcast and RS-330 for closed circuit industrial applications. For the purpose of the following discussion, the requirements are the same.

The standard RS-170 system consists of two 262.5 line fields each 1/60 second long (1/59.939 second for color), interlaced 2:1 to form one complete frame every 1/30 second. Approximately 23% of each frame is devoted to blanking time for the display. This time is used to transfer charge from one register to another in a CCD.

The RS-170 vertical blanking interval is .075 ± .005 of the field interval yielding 241.5 to 244.125 (243 nominal) active display lines per field.

Hier wird zur weiteren verwirrung von "active display lines" gesprochen.

This places a requirement of a minimum of 242 lines to be generated by the imager for each field. The maximum number of lines in each field is 262 to enable all of the cells for that field to be read out before the next field starts.

One row of cells is required for each line in the display. The RS-170 sync pulses cause the display to interlace each field with the preceding field. This causes a display with approximately 486 active lines per frame. The same 243 line positions generated on one field may be repeated on the next field giving a 486 active line display, with the resolution of 243 lines, or a new interlaced set of 243 elements may be supplied by the sensor (each field) for a total of 486 lines of vertical resolution per frame.
.
Die Übersetuung ist gar nicht mehr trivial, weil hier von Feldern anstelle von zum Beispiel Frames oder Einzelbildern gesprochen wird.
.

The RS-170 standards call for a horizontal line frequency of 15,750 lines/s (15,734 lines/s for color). The line time is therefore 63.49 microseconds. The nominal horizontal blanking interval is 17% of the horizontal line time leaving 52.7 microseconds active display time.

In closed circuit applications, the horizontal video bandwidth is unlimited. In monochrome and color broadcast, the luminance bandwidth is 4.2 MHz. Filling this bandwidth requires approximately 450 cells.

A composite color signal has the color information modulated on a 3.58 MHz subcarrier. Unless the video monitor or receiver uses an expensive comb filter to remove the chroma information from the luminance signal (not commercially used), the color subcarrier beats with the luminance signal causing an annoying edge creep.

It is therefore common practice to limit the luminance video bandwidth to 3 MHz. Most present-day video tape recorders are also limited to 3 MHz video bandwidths for luminance information.

This leads to a practical minimum number of cells for the 3 MHz bandwidth of 320 cells. The corresponding video data rate is 6 MHz. Table I summarizes the 525-line system requirements.
.

There are several factors affecting the choice of picture format.

Table II lists the most popular formats used in television, movie film, and still photography. Most movie film and tv formats have the 4:3 aspect ratio which is required for the standard tv system.

There has been a general trend towards smaller formats through the years in all three media. This has been the result of improvements in the resolution of lenses and photosurfaces as well as the need "for more compact and lightweight equipment.
.

The most popular lenses are the "C-mount" variety. There are two basic subdivisions of C-mount lenses.

These are the designs for the 16mm format of one inch vidicons and 16mm movie film. Any of the 16mm vidicon lenses can however be used for the smaller formats.

The 11-12mm formats seem very attractive because of their small size and ready availability of inexpensive interchangeable lenses. Lenses for 8mm and Super 8mm have low MTF's in the tv line ranges of interest and the small format requires extremely small CCD cell dimensions — beyond the state of the art for fabrication and device operation considerations to achieve standard tv resolution (240 television lines per picture height).

Hier ist die weitere - aber zur Unterscheidung wichtige - Benennung der "lines" erwähnt, bei uns sind das die so benannten "Linien" im Vergleich mit den "Zeilen".

An additional drawback of very small formats is the inability to achieve selective focus special effects with normal fields of view due to the inherent very large depth of focus.
.

Table I — Summary of 525-line television system requirements (RS-170).
.
Vertical Parameters
.

• Frame Time (1/30 Second)
• 2 Interlaced Fields (1/60 Second Each)
• Total Number of Lines Per Field (262.5)
• Vertical Blanking Interval (.075 ± .005 V)
• Active Scan Lines Per Field (Nominal/243)

.
Required Vertical Cell Count:
.

• 242-262 Each Field (Repeated or Interlaced)

.
Horizontal Parameters
.

• Luminance Electrical Bandwidth (4.2 MHz)
• Color Receiver and VTR Practice (3 MHz Bandwidth)
• Required Cell Count:
• 4.2 MHz (Approx. 450 Cells) 3 MHz (Approx. 320 Cells)

.
Format Parameters
.

• Picture Aspect Ratio (4:3)

.

Table II — Popular image formats
.

.

A vertical frame-transfer CCD imager processed for backside illumination that has good blue response may be fabricated into a new type of high-sensitivity CCD imager. This device is the CCD version of the Silicon Intensifier Target (SIT) Camera Tube *6).

The SIT-CCD imager consists of a photoemissive cathode, electron-optical focusing section, and CCD imager in an evacuated envelope. The photoelectron image is injected into the silicon at high energy, forming many secondary hole-electron pairs. This current gain greatly increases the sensitivity of a CCD by raising the output signal above background signal variation limitations. Interline-transfer CCD imagers cannot use this method of low-light-level sensitivity enhancement.
.

The most basic limitation on limiting resolution is the number of cells contained in the imager. (Heute sprechen wir von der Zahl der Pixel.) In the horizontal direction, the number of cells is determined by the number of channels formed by the channel stop diffusions (320 for the 512 x 320 device).

In the vertical, the number of resolution elements is determined by the number of different storage cell configurations formed by the transfer electrodes in the image area (256 actual cells electronically interlaced 2:1 to achieve 512 sample positions in the vertical direction).

Resolution is usually defined in terms of Television Lines / Picture Height (tvl / ph). In the horizontal direction, this number is 3/4 of the horizontal cell count or 240 tvl/ph due to the 4:3 aspect ratio.

In the vertical direction, only 486 resolution samples are displayed out of the 512 possible, and the finite line structure in the display is often quoted as reducing the effective useful resolution to approximately 350 tvl/ph due to the kell factor even though actual resolution may be seen to the 486 sample limit with proper pattern phasing. In the horizontal, the video is smoothed into a continuous waveform and the finite cell (moire) effects are not as severe.

Another factor affecting picture sharpness is the reduction of MTF (sine wave response) due to resolution loss mechanisms in the image formation and readout processes.

These losses are tabulated in Table III for a 1.2 mil cell dimension. The lens is the first contributor to the loss of resolution in image generation.

The lateral diffusion of electrons in the imager before collection is the second mechanism. It is a "Sech"-function of the optical spatial frequency and field-free diffusion length in the bulk. The numbers shown are representative of typical geometries.

The third loss is the Sin X/X loss due to the finite cell sampling process. All of these effects are also present in silicon vidicons *1).

The resolution losses in the readout process are given by the equation in Table III. "Ne" is the transfer loss and "fe" is the frequency of one cell. This resolution loss mechanism is analogous to the resolution loss of a scanning electron beam in a silicon vidicon.

The overall MTF is the product of the image formation MTF and the readout MTF. The squarewave response CI F is larger than the sine-wave response MTF. The actual measured responses depend on the phasing of the light pattern with respect to the cell pattern. If the black-and-white bars are lined up with the cells in an in-phase condition, the response is greatest.

If the bars are lined up with half a white bar and half a black bar contained in each cell, the response at the cell limit will be zero.
.

In the vertical frame transfer design, the picture is integrated for one field (16.66 milliseconds). The picture is then read out completely each field. There are therefore no smearing effects due to lag during panning except for resolution loss due to image motion during exposure as in any movie film camera.

Some other types of solid-state sensors integrate for a full frame (33.3 milliseconds) and have twice the panning resolution loss because of the longer exposure time.
.

CCD's exhibit unique problems of charge containment when overloaded with light. Charge is fairly well confined from spreading sideways by the channel stop diffusions. However, excess charge can spill up and down the channel.

A special mode of operation has been developed to greatly reduce the spreading of charge down the channel under overload conditions. This low-blooming mode is accomplished by maaintaining the phase fingers adjacent to a charge integrating phase finger at a voltage that maintains the surface under these fingers in light accumulation. This forms a temporary extension of the channel stop around each charge intergration collecting site minimizing charge spread down the channel.
.

A CCD imager has been developed that will offer an attractive alternative to camera tubes for many 525-line television applications.

It is capable of supplying the full resolution of color tv (240 tvl/ph). The features of CCD imagers that are superior to camera tubes are signal-to-noise ratio, freedom from lag, and absence of microphonics (soetwas kommt nur in Aufnahmeröhren vor) . The small light weight, low-power consumption, and precision image characteristics are additional benefits. Table IV summarizes the performance parameters for this developmental 512 x 320 CCD imager.
.

• Anmerkung : Hier nochmal der Hinweis, daß die Bezeichnung "240 tvl/ph" - also der Zusatz "TV-Lines" !! bei uns die Anzahl der Linien sind, die die echte Auflösung spezifizieren.

.

.

• 1. Rodgers, III. R.L., "Beam Scanned Silicon Targets for Camera Tubes," Paper presented at IEEE Intercon, New York. NY, March I973.
• 2. Weimer, P.K., Pike, W.S., Kovac, M.G., AND Shallcross, F. V., "The Design and Operation of Charge-Coupled Image Sensors," Paper presented at IEEE Solid State Circuit Conference, Philadelphia, Pa., February 1973.
• 3. Kovac, M.F., Shallcross, F.V., Pike, W.S., and Weimer, P. K.., "Design, Fabrication, and Performance of a 128X 160 Element Charge Coupled Image Sensor," Paper presented at CCD Applications Conference, San Diego, California, September 1973.
• 4. Tompsett, M.F., Amelio, G.F., Bertram, W.S., Buckley, R.R., McNamara, W.J., Mikkelsen, Jr., J.C., and Sealer, D.A., "Charge-Coupled Imaging Devices: Experimental Results," IEEE Trans. Electron Devices, Vol. EIV^ pp.992-996, November 1971.
• 5. Sequin, C.H., "Interlacing in Charge-Coupled Imaging Devices, " IEEE Trans. Electron Devices, Vol. ED-20, pp. 535-541, June 1973.
• 6. Rodgers, 3rd, R.L., et ah "Silicon Intensifier Target Camera Tube." Paper presented at IEEE International Solid State Circuits Conference, Philadelphia, Pa.

.