CA1087295A - Television picture size altering apparatus - Google Patents

Abstract

Abstract of the Disclosure The size of a color television picture field is changed without changing the size of the raster. This is achieved in an apparatus that includes means responsive to a field of television video signals for sampling same and providing a predetermined number of samples of television video signals per line and a predetermined number of lines per field and means responsive to the predetermined number of samples per line and lines per field for providing a full field of television video signals. Size reduction is accomplished by causing the first mentioned means to skip samples provided at the output thereof in groups of N, where N is the number of samples in a color subcarrier cycle, and to skip line pairs of the field.

Description

~0~7Z9S RCA 70,667 This invention relates to apparatus for altering the size of a tclevision picture and, more particularly, to an apparatus whicll includcs means for samplin~ incoming video signals and providing sccond television video signals constructed from samples of the incoming television video signals.
Recently, video synchronizers have been introduced.
These synchronizers have been introduced in order to synchronize broadcast sources to the local station reference generator. A video synchronizer is an electronic unit that samples the analog input, converts it to a digital format, stores the digital data, and operates on the digital data to deliver a desired analog output which is constructed from the sampled video. It is designed to automatically lock a non-synchronous broadcast signal to a reference generator ana thus allow fully synchronous treatment of the incomillg video for mixing with station programs. The non-synchronous signal is digitized and stored in a memory ! The data is clocked out of the memory at a rate locked to the reference sync generator (usually the local station). This synchronizer isolates the input and output video lines and the output is fully synchronous in vertical, horizontal and color phases with the reference.
; 25 Tliese video synchronizers make possible many special effects for relatively low additional cost.
App]icant's invention herein relates to the special effect of expansion or compression using apparatus like these video synchronizers by adding or eliminating samples.
Applicant has found, however, that picture compression or

-2-, . .

1087Z95 RCA 70,667 1 expansion by indiscriminately adding or eliminating samples or lines is unworkable for reproducing color since such adding or eliminating disrupts the subcarrier, and randomly adding or eliminating lines disrupts the normal color sequence required for sequential lines in the NTSC
television system.
In accordance with a preferred embodiment of the present invention, an apparatus for altering the size of a television picture without changing the size of the raster is provided. This apparatus includes means responsive to a field of television signals for sampling same and providing a predetermined number of samples of a television video signal per line and a predetermined number of lines per field, and means responsive to the predetermined number of samples per line and lines per field for providing a full field of television video signals. The size of the television picture is changed by means coupled to the first means for causing the first means to alter the number of samples provided at the output thereof.
FIGURE l is a system block diagram of an embodiment of the present invention.
FIGURE 2 presents waveforms useful in illustrating the skipping of whole cycles to produce picture compression without color loss.
FIGURE 3 presents waveforms useful in illustrating the skipping of alternate line pairs of a field to produce vertical picture compression without color loss.
i FIGURE 4 is a block diagram of a system for i reducing both the width and the height of a television j ; 30 - 3 -s .
,, ., RCA 70,667 1 picturc fiel~l by 2 to 1.

Referring to FIGURI. 1, there is illustrated a system for altering the size of a television picture using, for example, a video synchronizer 11. This synchronizer may be for example RCA type TFS-121 video synchronizer sold by RCA Broadcast Systems, Camdell, New Jersey. The video synchronizer 11 include~ an A to D (analog to digital) convcr~er 13, a memory 15, a 1) to A (digital to analog) lo convertcr 17, a controllablc write clock generator 19, a memory control 21 and a controllable read clock generator 23. The picture video is applied at terminal 25 and is coupled to an A to D converter 13. The controllable write clock generator 19 generates strobing or sampling pulses to the A to D converter 13 for example at a 14.3 MHz rate.
In a typical system, 768 samples of the video are taken Oll each horizontal Iinc of thc telcvision picture. Based on the relativc alllplitudc of thcse signals a digital code for each sample is stored in the memory 15. This code may be for example an 8-bit digital code. The video is sampled four times the subcarrier cycle (3.57 MHz) in order to store and faithfully reproduce the color subcarrier signal. At the same time, these clock pulses from the write clock generator 19 cause the memory control 21 to provide a write address in the mcmory via multiple leads 27 for each sample. In additioll, memory control 21 senses the end of a line (for example by counting 768 samples from clock 1~) and provides that address to the memory 15.
The video is read out of the memory 15 by applying clock pulses from controllable read clock generator 23 to the 1087Z95 RCA 70,667 I D to /~ convertcr 17 and memory control 21. The read clock - pulses apl)lied to the memory control Zl cause the memory control 21 to supply appropriate read addresses via multiple leads 29 to the memory 15. Upon receipt of the clock pulses applied to the D to A converter 17, video appears at the output thereof that corresponds to the 8-bit coded video at the input. In the typical synchronizer system 11, tlle video out of the D to A converter 17 is applied to video processor 20 where locally generated sync, burst and blanking are applied. In the system describe~ herein, field compression or expansion is pro-vided by field compressor/exI)ander 31 coupled, for example, to the memory control 21, to write clock generator 19 and to read clock generator 23.
Field compression is achieved, for example, by ~ sending inhibit signals from compressor/expander 31 to : the controllable write clock generator 19 via lead 18 inhibiting pulses therefrom and causing some of the video not to be converted to digital form and applied to the memory 15. At thc saltle time inhibit pulses are sent fr~m compressor/expander 31 ~ia lead 16 to the memory control 21 for causing the omitting of the addressing during the sample skipping periods. This reduces the number of samples and hence the amount of video stored in the memory 15. For example, assume that every other sample of the digitized picture was eliminated by failure to send a .,; .
clock pulse from write cloclc generator 19. When the read clock generator Z3 addresses the memory control 21 at the normal rate, the picture width would come out in half the time and there would be a left-to-right compression of .

~ A 7~,667 10~729~

2 to 1. Similarly, an inhibitiIlg signal from the fieldcompressor/exl)alldcr 31 may be applied for example to memory control 21 via lcad 1~ to cause the memory control 21 to providc addresses for only every other line. At ~5 thc same timc an inhibiting signal may be applied from -compressor/exI)an~er 31 to thc controllable write clock generator l'~ via lead 18 to cause the write clock generator 19 to apply strobing pulses to the A to D converter 13 only duritlg every other 1 illC and thereby to cease writing I0 every othcr line of thc fic1d into tlle memory. When the read clock generator 23 applies signals at the normal rate, the resultin~ output picture is vertically compressed 2 to l.
In accordance with the system described in FIGURI' l, the compressioIl or expansion can be achieved by writing into the memory at thc normal rate but'by controlling the controllable read clock generator 23 via lead 30 so as to skip samples in the memory (compression) or to repeat samples in the melllory (expansion). For example, field compression is achieved by providing an inhibit signal via lead 32 to cause the memory control 21 to skip addresses in the memory. Field expansion can be provided for example by providing a second signal from the field compressor/expander 31 to be applied to memory 25 control 21 via lead 32 to cause the memory control 21 to rcpeat a~drcsses and thereby repeat the scanning of portions of the memory 15 by providing repeated samples ~ and lines to be converted to video.
f ~ The approach to picture compression by eliminating or skipping every other sample of a digitized picture or - ~

RCA 70,007 1(~87Z95 tllc ~upllrcssillg of the ~ritin~ into the memory of every Othel' linc iS .1 problem if thc vi~eo transmitted includes color inforl~ tioll. Ihis apl)roacll would cut the sample ratc in hal f and l~encc rendcr the system incapable of reliably rel)roducing the color subcarrier. ~.ven if this dificll]ty was circumvented, it is not possible to achieve vertical comprcssion by omitting alternate lines since the phase of the color subcarricr alternates on each successive line of the ficld. Thc phase of the color burst and chroma on each successive linc of a field is 180~ out of phase with respect to color burst and chroma of the preceding line. Ior examl)lc, the pllase of the color burst and chroma of the sccond line of a field is 180 out of phase with the color burst and chroma of the first line of that field. If every other line were omitted, this subcarricr on thc remaining lines would be of identical phase line-to-line and hence the reproduced picture would be devoid of color.
These problems arc solved hcrein by eliminating or repeating the number of sanlples per line in groups of N, where N is the number of samples per subcarrier cycle of ; a color television signal, and by changing the height of the picture by eliminating or repeating adjacent pairs of lines in a given field. For example, in the arran~ement 2S shown in }~IGURF. l where it is desired to reduce the width of the picture by 2 to 1, the write clock sampling rate would be reduced 2 to 1 by skipping every other cycle of the color subcarrier. In the sysl:em described above, there are for example 768 samples per line and the number of samples per cycle of the subcarrier frequency I~CA 70,667 ~0~37295 is four. ~a]lll)lcs in this case are skipped in groups of four. Io rcdllce the width l)y 2 to l, four sampling pulses would hc providcd by the ~rite clo~k generator l9 at the normal ratc iith the next four sampling pulses inhibited to thereby write a full color subcarrier cycle into the memory and then skip a whole cycle. The result is that the color is preserved but the picture is horizontally compressed 2 to l as shown in FIGURl' 2. Waveform A of FIGIJRI. 2 shows six wholc cyclcs of the color subcarrier sigllal (cycles 33 thru 38 - 24 samples). Waveform B of I IGIJRl 2 shows the skippillg of whole cycles of the color subcarrier (cycles 34 and 36 - 8 samples) to produce color compression without color loss. The "x" and "-" on waveform A indicate the normal sampling points. The "-"
indicates the samples skipped to compress. Note that four `~ samples are skipped when eacl- cycle is skipped.
In the system where the vertical height is also to be rcduced 2 to 1, the skippin~ of the lines is modified to write in a field in the sequence store - store, skip - skip, storc - store. Since half of the vertical elements are omitted, the picture is vertically compressed 2 to l. 'I'he skip configuration as chosen is the key to vertical compression. Since adjacent pairs of lines show the proper color phase alternation, the lines are ~5 written in pairs and skipped in pairs, thus preserving the vital line-to-linecolor phase alternation as shown in FIGIJRI. 3. Waveforms 41 and 42 Oll picture field lines A
and l3 illustrate the phase of the subcarrier wave on the adjacent lines of a field. Similarly, waveforms 45, 46, 47 and 48 illustrate the phase of the subcarrier at the '. '' KCA 70,667 1~ !37Z~5 hc~in~ of lincs G, ~, 1, and I, respcctively, of thefic1d. ~s llotc~ l)y thc store - store, skip - skip mcthod, linc~ G and ~ arc removcd and since the subcarrier on line l. is 180 out of phase with the subcarrier of line B, a reproduced picture is provided in which the proper phase alternation is maintained.
The above techni4ucs are extendable to picture expansion by repcatillg groups of cycles and pairs of lines in a fie]d wllerein the grollpillg within a line is done in terms of N sallll)lcx, wherc N is e~lual to the number of samples in a subcarricr cycle and the lines are repeated in adjacent pairs. For example, whcrc it is desirable to expand 2 to 1, the four samples or full subcarrier cycle would he read from the memory and then repeated from the memory before gOillg to a new group of four samples.
Also, in this arrangcmcnt to cxpand the vertical height by 2 to 1, the lines would bc repeated in pairs. This field expansion in thc arrangeIllcnt of FIGlIR~ 1 would be controlled hy a signal from the compression/expander 31 via lead 32 which causes the mcmory control 21 to repeat samplc and line addresses to repcat samples in groups of four and to repeat lines of a field in pairs.
In the typical system described above, where there are 768 samples per active video line, dropping four samples or one cycle of subcarrier at some time during the active-line write-in reduces the readout picture to 764/768 ths of its normal horizontal size.
~ Further, by causing the apparatus to provide evenly ¦ distributed drops of four samples will cause horizontal compression as the groups of samples dropped are slowly g ' ,, .
~ ' .

~CA 70,667 incrcased in nunlber. ~onversely, a single repeat of four samples (on the read side~ ~ill increase the picture wi~tl~ to 172/768 ths of its norlllal size; more repeats smoothly increasing in number will smoothly expand the picture horizontally. ~11 of these repeats or dropping of salllplcs are to bc done on a subcarrier-cycle basis so as to reproducc the subcarrier cycle. Similarly, the height of the picture can be changed by adding or sub-tracting pairs of lines as describcd above to thereby -~
afford a vnriable size change in the vertical direction.
laken together, these two mechanisms will provide zoom, comprcssioll or aspect ratio modificatioll.
Referring to ~IGURF 4, a system is illustrated for compression of the width and height by a ratio of 2 to 1.
The video input signals to be comprcssed are applied to input terminal 51 of for example a video synchronizer, such as XCA TFS-121 Video Synchronizer. The videosignals at terminal 51 are applied to an A to D tanalog to digital) converter 53. rhe ~ to 1) converter is strobed by clock pulses applied to terminal 55 and the sampled analog video signal is converte~ to for example an 8-bit parallel digital code. lach 8-bit word is coupled via an 8-wire lead 54 to the memory 57 where it is placed in location according to an address provided via read/write switch 59 when it is in the write or store position "W". When read/write switch 59 is in the read or "R" position, with the same address from the read address generator 95, the same 8-bit word is transferred out to the D to A (digital to analog) converter 61 via 8-wire lead 60 where it is

3 converted from the digital code back to analog and applied : - 1 0 -~0~7Z95 R~A 70,667 to the vi~leo output ~2.
l~hether or not the ~pparatus compresses is determined by thc position of ganged switches 63, 64 and 65.
When tllese switches are in the "C" position, the apparatus is compressing and when it is in the "N" position it is normal. All of the switchcs 63, 64 and 65 are ganged together aIl~l they are shown in FIGU~r 4 in the compressed or "C" position. Consider, first, ilowever what happens in the "N" or normal position. In the normal position, o 768 c]ock pulses from a clock generator 71 are provided to encode the active portion of each television line. If the blanking interval is also encoded, 910 pulses are required per line. These clock pulses (at 14.3 MHz rate) are applied to AND gate 73. I~hen switch 63 is in the "N" position, a constant high level signal is provided to AND gate 73 enabling the pulses from clock generator 71 to , be applied to the picture-element (PIXIL) counter 75 and to AND gate 77 which is coupled to the strobe terminal 55 of A
to 1) converter 53. When switch 65 is in the "N" position, ; 20 an enabling signal level (hig}l) is provided via lead 76 to ANI) gate 77 causing each of the clock pulses provided from AND gate 73 to be applied as strobe pulses to the A
to D converter 53. The picture-element counter 75 is a digital counter capable of counting to 768. This counter may be for example a 10-bit digital counter. This counter provides via lO-wire lead 79,a lO-bit code word which is :.
the address for the video sampled at that instant to the "W"
or write terminal of read/write switch 59. A decoder 81 is coupled to the counter 75 and provides an output to terminal "N" of switch 64. This decoder 81 senses the ~, 10872~5 RCA 70 667 1 reception o~ 768 pulses to coullter 75. Since there are 768 samples per active piCtulC line this decoder 81 senses the termination of an active picture line. With the switch 64 in the N positioll the decoder 81 provides a reset pulse to the picture element counter 75 and also provides a signal to AN~ gate 83 via bus 82. As stated previously when the switch 65 is in the N position, a high enabling level is provi~ed to ~ND gate 83 enabling the pulse from decoder 81 to be applied via lead 82 to line counter 85. Line counter 85 provides a line address to the W' or write inl)ut tcrminal of read/write switch 59 via lead 78. This colmter 85 may be for example a 9-bit counter and lead 78 inay be a 9-wire lead to provide the `- necessary number of bits to identify each line. There is also a multi-wire (l9-wirel lead 58 between read/write ~; switch 59 and memory 57. The video is read at the normal rate when switch 59 is in thc R' or read position and read address signals from rea~ address generator 95 are applied via leacl 80 ~19 wires) to 'R" terminal of read/
write switch 59 and read clock puls~s arc applied to the digital to analog converter 61 from generator 97. The read address generator 95 addresses the memory 57 with for example a l9-bit code which code in the normal state for the same sample in the same line is identical with that placed by the picturc-element counter 75 and the line countcr 85 to pick up the same 8-bit digital data at that location. Read/write switch 59 electronically switches automatically between read ("R") and write ("W') positions ~ .
in a msnner such that reading and writin~ does not occur at the same location at the same time with reading "

- : : , ~. . .
' " ' ~ . '~ ', ' .

1~87295 I~C~ 70,6~7 occurring only after writin~. When the switch 59 is in thc read position, the 8-bit code is read out of the memory 57 to tile D to ~ con~erter 61 via 8-wire lead 60. rhc 8-bit code is converted back to analog in response to the strobe pulse at the converter 61 provided by generator 97. The read clock generator for normal reading is set at 14.3 M~lz.
When switches G3, 64 and 65 are thrown into the "C" or compressed position, the constant high enabling signal levels presented from switchcs 63 and 65 are removed and the decoder 81 is disconnected. A decoder 99 is connected betwcen picture-element counter 75 and terminal "C" of switcll 64. This decoder 99 detects the sampling of 384 clock pulses received at counter 75 from the generator 71. Coupled between the output of generator 71 and the "C" terminal of switch 63 is a divide-by-four circuit 101 and a flip-flop circuit 103. Every four pulses received from the clock generator 71 causes the flip-flop 103 to change state. If flip-flop 103 is initially in a high enabling level output state, a high enabling level is applied to terminal "C" of switch 63 for the first four clock pulses and a "low" levcl (inhibit level) to terminal "C" for the next four clock pulses. The first four clock pulses are thereby enabled to pass through AND gate 73 to the picture-element counter 75 and AND
gate 77. When the second group of four pulses occurs, however, a low inhibit level is present at terminal "C"
of switch 63 and at AND gate 73 which causes these clock pulses to be inhibited from being applied thru AND gate 73 to the picture-element counter 75 and AND gate 77. After ^13-RCA 70,~67 ~087Z~5 1 thc second ~et of four pulses, a high enabling level is again preselltcd at the "C" termina:l of switch 63 and these clock pulses are again enabled to pass through AND
gate 73. Since durin~ the second set of four clock pulses, there are no pulses applied to ANI) gate 77, there are no ~trobe pulses to tcrminal 55 of A to ]) converter 53. The video ~ill be cut off during the absence of the clock pulses since che A to ~ conYerter 53 will not convert the second four video samples. Consequently, this second set of four video samples will not be applied to : the memory 57. Similarly during this blanking period, there will be no changes of addresses from the picture-element counter 75 since no pulses will be received from the .. ANI) gate 73. When the picture-element counter 75 has . 15 counte~ to 384 pulses, the decoder 99 will apply a high enab~ing level to terminal "C" o switch 64. This will provide a high or reset pulse via lead bus 82 to the . coullter 75 an~ a hi.gh enabling signal to AND ~ate 83.
Since the number of picture elements encoded dur.ing a : 20 horizontal line has beel- cut in half, a horizontal line has beem complete~ on the reception of 384 pulses applied to the picture-element counter 75.
Vertical compression is provided by the circuit .- comprising flip-flops 105 and 107. Flip-flop 105 respond to each high level signal Oll the load bus 82 by chan~ing . s~ate. l`ach time a high level signal is provided at output of flip-flop 105, this high is applied via lead 109 to flip-flop 107 causing it to change state. Consequently, :. the output of flip-flop 107 changes states on the receipt 3 of every other pulse applied from the load bus 82 .There-.

.. . .
.. .

1087Z95 ~CA 70,667 ore, the output from flip-flop 107 rcmains high with the first two pulses applied on the load bus 82 and then provides a low for the nc~t two pu1ses on the load bus 82.
Therefore, a high enabling level is provided for the first two lines and a low or inhibitil-g level is provided for the second two lines to "C" terminal of switch 65. Terminal "C" of switch 65 is coupled to both AND gates 83 and 77 via Icad 7~. At ANl) ~atc 83, thc line rate clock pulses from tl-c bus 82 arc cnabled for tlic first two lines but are intcrruptcd for cvcry othcr.pair of lines. At the same time, a high is prov.ided Oll lead 7~ so that gate 77 is also enabled for the first two lines and interrupted for the next two lines to cut off the strobe pulses to the A to D converter 53 for two succcssive lines. Consequently, in the above system, the memory 57 will record four successive samples and it will skip four successive samples and will do this for an entire line. It will .... . .
also write two succcssive :lines and skip two successive : ]ines. When the norma:l read addrcsses from address 95 are ... .
fed to the read/wri.te switch 5~ via lead 80, the memory : will produce at its output a compressed picture in which only one-quarter of its total area is utilized. However, the apparatus ShOWII has preserved the proper subcarrier relationship on a left - right and a line - line basis so that a normal NT~SC subcarrier sequence is read out of the ~,''. memory.
;. The above described techniques for compression : or CXpallS.iOII may be used in the SECAM and PAL systems wlth some modifications to accommodate their specific require-ments. In the SECAM system, the two color difference . -15-, -, .
, , 10~7295 RCA 70,667 1 sig]l;lls are transmitte~1 alternately on successive lines of a field. Tllcrefore, thc ~crtical compression arrangement taught above for the NTSC system applies directly to the Sr.CAM system, since prcscrving and dropping lines in succcs~ivc pairs preserves tllc alternate line color diffcrence arrangement of S1CAM. In the PAL system, the wi~th compression metho~ taught above (full cycle skip) may be applie~ directly. I`hc grouping of lines for omiss:ion or retention must bc ~one in a manner which recognizes the phasing of the lincs in the PAL system.

~ 15 ,~:,'' .
' :

. .

',' . .
:, 25 ., ` . .

' . ;. . ~ . '- . ' , :

Claims (16)

WHAT IS CLAIMED IS:

1. Apparatus for altering the size of a television picture without changing the size of the raster, comprising:
sampling means responsive to incoming television video signals for sampling same and providing a predetermined number of samples of the television video signal per line and a predetermined number of lines per field, providing means coupled to said sampling means and responsive to said predetermined number of samples per line and lines per field for providing a full field of sampled television video signals, and altering means coupled to at least one of said sampling means and said providing means for altering the number of samples per line or lines per field provided at the output thereof to thereby change the size of the television picture formed by said sampled video signals.

2. The combination of Claim 1 wherein said altering means alters the number of lines per field provided to thereby change the vertical height of the television picture.

3. The apparatus of Claim 1 wherein said altering means alters the number of samples per line provided to thereby change the width of the television picture.

4. The apparatus of Claim 1, wherein said altering means alters the number of samples per line by skipping samples in groups of N, where N is the number of samples in a color subcarrier cycle.

5. The apparatus of Claim 4 wherein N is four.

6. The apparatus of Claim 2 wherein said altering means alters the number of lines per field by skipping line pairs of the field.

7. The apparatus of Claim 1, 2 or 3 wherein said altering means alters the number of samples per line by repeating samples in groups of N, where N is the number of samples in a color subcarrier cycle.

8. The apparatus of Claim 1, 2 or 3 wherein said altering means alters the number of lines per field by repeating line pairs of the field.

9. The apparatus claimed in Claim 1, wherein said altering means includes means for causing said sampling means to provide every other group of N samples at the output thereof, where N is the number of samples in a color subcarrier cycle.

10. The apparatus of Claim 9 including means coupled to said sampling means for skipping alternate line pairs of a field.

11. The apparatus of Claim 9 or 10 wherein N is four.

12. The apparatus of Claim 1 wherein said sampling means includes an analog to digital converter and a write clock pulse generator, said analog to digital converter being responsive to strobing pulses from said generator for converting samples of said incoming video to digital data.

13. The apparatus of Claim 12 further including a memory coupled to said analog to digital converter for storing said digital data.

14. The apparatus of Claim 12 wherein said providing means includes a digital to analog converter and a read clock generator with said digital to analog converter coupled to said memory and responsive to read clock pulses for converting said digital data signals for providing analog samples of said video at the read clock rate.

15. The apparatus of Claim 13 wherein said altering means includes means for interrupting the pulses provided by the write clock generator by causing the write clock generator to skip clock pulses provided at the output thereof in groups of N, where N is the number of samples in a color subcarrier cycle.

16. The apparatus of Claim 14 wherein said altering means includes means coupled to said sampling means for inhibiting pulses from the write clock generator for line pairs of the field to thereby change the vertical height of the television picture.