GB2243047A - A digital composite video encoder - Google Patents

Abstract

Input digital luma and chroma video signals encoded in accordance with CCIR Recs. 601/656 are encoded into a composite analogue output signal of desired format, e.g. PAL or NTSC. A phase calculator 27 determines the subcarrier phase at each position along each line, and this phase value together with the chrominance values are used to address lookup tables 26. The outputs of the lookup tables are applied to DACs 29 and combined and filtered 30. The resultant is further combined with the output of a DAC 22 receiving the luminance signal and with information from a sync. pulse input 44 in a combining and filtering circuit 31. The lockup tables 26 are reprogrammable and other parts of the encoder suitably controllable to allow a variety of composite output formats to be generated. As most of the circuitry operates in the digital domain, the need for setting-up adjustments and the problems of component ageing are minimised. <IMAGE>

Description

A DIGITAL COMPOSITE VIDEO ENCODER Background of the Invention This invention relates to a digital composite video encoder which receives digital component video signals and converts them to analogue composite video standards, for instance PAL or NTSC.

Even when working with digital input signals it is the usual practice to convert to analogue form and then use analogue encoding schemes thrpughout. This requires a substantial amount of careful setting up, and results in an encoder that is prone to drifting as the components heat up or age.

There is a particular need for encoders to work with digital signals encoded in accordance with CCIR Recommendations 601 and 656.

Summary of the Invention The invention is defined in the appended claims 1 and 9 to which reference should now be made. Advantageous features of the invention are set out in the subclaims.

In a preferred encoder embodying the invention, to be described in more detail below, input digital luma and chroma video signals encoded in accordance with CCIR Recommendations 601 and 656 are encoded into a composite analogue output signal of desired format, e.g. PAL or NTSC. A phase calculator calculates the subcarrier phase at each position along each line, by counting lines and pixels and making use of pre-programmed lookup tables. This phase value together with the chrominance sample values are then used to address lookup tables, the outputs of which are converted to digital form, and combined and filtered. The resultant is then further combined with the analogue luminance signal and with synchronisation information, and filtered to the required video band.

The lookup tables are reprogrammable and other parts of the encoder suitably controllable by a host processor.

As will be shown the preferred encoder thus provides a mechanism for converting CCIR 601/656 digital component video signals into any one of a number of composite video standards. The multi-standard digital composite encoder performs as much as possible at the digital stage and converts to analogue as late as possible, with the minimum number of adjustments and therefore with a minimum of drift. Those adjustments that are required are performed using digital control from the host system processor.

The mechanism for the conversions is provided by a number of lookup tables which calculate the phase of a reference subcarrier at the beginning of each line, and the relative subcarrier phase of each pixel compared to the beginning of each line, and the relationship between calculated phase chroma values and the expected subcarrier output phase and amplitude. The fact that these lookup tables are programmable means that this function can be set up under software control at any time for any standard (e.g. PAL or NTSC), and many non-standard systems.

Brief DescriPtion of the Drawings The invention will be described in more detail by way of example, with reference to the accompanying drawings, in which: Figure 1 is a block schematic diagram of a preferred PAL/NTSC encoder embodying the invention for use with digital signals in accordance with CCIR Recommendations 601 and 656; Figure 2 is a block diagram of the phase calculator of Figure 1; and Figure 3 is a detailed circuit diagram of the encoder of Figure 1, Figures 3A to 3D being assembled as shown in Figure 3E so as to form a complete Figure 3.

Detailed DescriPtion of the Preferred Embodiment The preferred encoder 20 will first be described with reference to Figure 1. The digital coder may be considered as having four main blocks. The first of these is a phase calculator 27, which is shown in more detail in Figure 2, and which calculates the reference colour subcarrier phase at each pixel on the screen.

The other three blocks are the luminance channel 10, the two colour difference channels 11 and 12, and a final block 13 of analogue filtering and combining, and including an output stage.

The input digital signal in accordance with CCIR Recommendations 601 and 656 is received as a luma or luminance component at an input 32 and a chroma or chominance input 34.

There are two chroma samples associated with alternate luma samples, the chroma samples respectively relating to the red and blue chroma signals. The signals appearing in the blanking periods are removed from the luma and chroma signals prior to application to the circuit of Figure 1.

The luminance signal received at input 32 is passed through a variable-length delay 21 to compensate for delays through the chroma channels. The delay 21 takes the form of a FIFO (first in first out) register, and is controlled by a delay control circuit 36. This is a well-known technique. The output of the delay 21 is then applied to a digital-to-analogue converter (DAC) 22 capable of operating at video speed. The converter 22 latches the digital signal and converts the delayed luminance data into analogue form.

The gain of the converter is controlled by a gain control circuit 23 to allow a variation of plus or minus ten percent to allow for the tolerance of the converter. Conveniently the control 23 comprises a second, slower digital-to-analogue converter, which sinks the converter reference current.

The chroma signal received at input 34 comprises alternate red and blue samples time multiplexed onto a single eight-bit data path. These are distributed by a demultiplexer 38 to two latches 24,25 for the blue and red samples respectively. The two chrominance channels 11,12 are basically identical. A look-up table 26 is connected with some of its address inputs connected to the output of the respective latch 24,25, and the rest of its address inputs connected to the output of a phase calculator circuit 27 described in more detail below. The outputs of the lookup table is applied to a latch 28 and thence to a digital-to-analogue converter 29.

The outputs of the two chrominance channels 11,12 are combined in a chroma combining and filtering circuit 30, and the output of circuit 30 is combined with the output of the luminance channel 10 in a chroma and luma combiner and filtering circuit 31, the output 40 of which constitutes the output of the encoder.

In addition, the encoder includes table reprogramming control circuitry 42 connected to the two lookup tables 26 to reprogram the tables in the event that the output signals are to be generated to a different output standard.

The output signal comprises the luma signal as a base-band signal and the chroma signal modulated in phase quadrature onto a suppressed sub-carrier in the upper part of the luminance spectrum.

Thus much is common to NTSC and PAL signals in all their variants, though the bandwidths and subcarrier frequencies differ, and also PAL has a subcarrier phase inversion on alternating lines, while NTSC uses different reference axes for the subcarrier phases. It should be noted at this point however that the encoder is capable of encoding with other or non-standard formats, by appropriate modification of the lookup tables and filtering circuits. For example it could be used to generate a SECAM signal, in which the red and blue chroma component signals are transmitted alternately line-by-line.

We have appreciated that, for any given desired output standard, the two sets of chroma samples can be directly encoded, provided that the phase of the subcarrier at each sample position, or pixel, is known; and thus that the encoding can be done in the digital domain by the simple use of a lookup table.

The information concerning the subcarrier phase is generated by the phase calculator 27, shown in more detail in Figure 2. The phase calculator comprises a counter 50 for counting lines and outputting the line number, and a counter 52 for counting pixels (or samples) along each line and outputting a count representing the number of pixels along the line. The pixel count is applied to a lookup table 54 which stores the change of phase along the line from the start of the line, and outputs a value representing the phase to an adder 56. A lookup table 58 is connected to the line counter 50 and stores and outputs the phase of the first pixel along the line. This is applied to the other input of the adder 56, which also receives a possible base offset phase at an input 62. The adder 56, which is a modulo-256 adder, provides on an output 60 a value indicative of the instantaneous relative phase of each pixel.

For a different output standard, the contents of the lookup tables 54,58 are changed to give the subcarrier phase appropriate for that standard.

The phase calculator 27 calculates the relative subcarrier phase for each pixel as an 8-bit number where the range 0 to 255 corresponds to the range 0 to 360" of phase. The programmable base phase offset, a programmable constant, can be changed by software at any time to change the phase of the composite video output subcarrier with respect to the start of the frame.

Reverting to Figure 1, the incoming chroma samples are latched (24,25) to provide a constantly available digital value of the current chrominance level. During normal active picture time this value is fed into the lookup table 26 containing the appropriate chroma amplitudes for a particular colour value and for the particular subcarrier phase as determined by the phase calculator 27. The lookup table 26 is stored in static random access memory (SRAM) having 65,536 locations each of 8 bits, and which can be programmed at any time by control circuitry 42 under user control to accommodate any subcarrier frequency and composite video standard. The SRAM has to have a suitably low access time, e.g. 35ns. The output of the lookup table 26 is an 8-bit number representing the amplitude of U (or V) for a PAL output or I (or Q) for an NTSC output.The two chrominance channels are basically identical, the only difference between the phase relationship of the two, being ninety degress apart at subcarrier frequency and, in the case of PAL, having one-hundred-and-eighty-degree switching in the blue colour difference path.

The output of the lookup table 26 can be shown to be mathematically equivalent to the value produced by an ideal analogue composite encoder, sampled at the pixel rate. As the CCIR 601/656 pixel rate is 13.5 MHz and is more than twice the subcarrier rate for all standards, it is clear from Nyquist's theorem that a sine wave output filtered from this will contain all the encoded information.

The resultant chrominance amplitudes are then latched in latches 28 and fed at the luma pixel rate to a pair of digitalto-analogue converters 29 capable of operating at video rates. In a PAL system, one converter 29 is used for each of the U and V components, while in an NTSC system, one converter 29 is used for each of the I and 9 components. The outputs of the DACs are then combined and filtered in circuit 30 to produce a modulated subcarrier output. The final combining and filtering circuit 31 receives the modulated chroma output of circuit 30 and the output of the luma channel 10, and also synchronization information received at a terminal 44. The circuit 31 includes the final low-pass filter, after which the signal is buffered through a high speed operational amplifier to provide a 75 ohm composite video output.

During the inactive picture period, a fixed value is used at the appropriate time to generate the burst pulse. This value is supplied by a host processor through a latch.

Figure 3 is a more detailed circuit diagram of the encoder of Figure 1. A full description of Figure 3 is not necessary in view of the above detailed description of Figure 1. It will also be noted that circuit identification numbers are included on Figure 3. Items identified by reference numerals on Figure 1 are identified by the same reference numerals in Figure 3. The phase calculator 27 and demultiplexer 38 are not shown on Figure 3.

As shown in Figure 3, the luma signal at input 32 passes through FIFO register 21 controlled by a delay signal at input 102 and thence to a video DAC 22, the gain of which is controlled by DAC 23 (Figure 3D) over line FS4. The output of DAC 22 is applied to a video low-pass filter 104.

The blue and red chroma signals from demultiplexer 38 are received at inputs 106,108 and each applied through intermediate latches 110 to the latches 24,25 respectively. The output of phase calculator 27 is received at terminal 60 and applied to a latch 112. During the subcarrier burst interval, the host processor generates a computer subcarrier phase signal (compoutSc ) which is applied to terminal 114 and used instead of the signal at terminal 60, which is zero during these intervals. An adder 116 couples terminals 60,114 to latch 112.

The address inputs of the lookup tables 26 receive the outputs of the latches 24,25 respectively and each receive the phase information from latch 112. These provide a complete address for each lookup table, which provides an output through the respective latch 29 to the video DAC 29. The DACs 29 receive respectively the gain control signals FS5 and FS6 from the DAC 23 of Figure 3D.

As shown in Figure 3, two sets of combining and filtering circuits 30 are provided, the outputs of which are selectively grounded by relay contacts 118a, 118b, operated by a relay solenoid 118, so that only one is used at a time, depending upon whether a PAL output or an NTSC output is required. An adder 120 combines the outputs of the two DACs 29 and applies the resultant to a U,V bandpass filter 122 having a 1.3 MHz bandwidth. This is used for the PAL signal. For NTSC, two adders 124,126 are used which generate respectively: m.Cr + n.Cb and p.Cr + q.Cb where Cr and Cb are the outputs of the DACs in the red and blue chroma channels respectively, and m = 1.03748 n = -0.47844 p = 0.67296 q = 0.72652.

This then skews the effective subcarrier phase by about 33O as required for NTSC. The output of adder 124 is applied to an I bandpass filter 128 having a 1.3 MHz bandwidth and also providing an appropriate compensating delay, while the output of adder 126 is applied to a Q bandpass filter 130 having a narrower bandwidth of 0.4 MHz as required by the NTSC system.

The selected output, either from filter 122 or from both filters 128,130 is combined with the luma output of filter 104, and is mixed with synchronization information from a sync. generator 132. This is controlled by the host processor at an input 134 and by user control at an input 136 indicating whether the output standard has 625 or 525 lines per picture. The composite signal is then low-pass filtered to 5.8 MHz in filter 31 before passing through each of three high speed operational amplifiers 138 to provide buffering to the outputs 40. Only one output need be provided though in this example six are shown.

The parts of the circuit not so far described are concerned with programming the various lookup tables, particularly the lookup tables 26. Those skilled in the art will be fully aware of the steps necessary to program a static RAM, and any suitable method may be used which is convenient in the particular environment in which the RAMs are to operate. The particular arangement of registers, latches and other circuits used does not form part of this invention. However, Figure 3 contains the necessary component identification numbers etc. used in the particular example shown therein.

Claims (9)

1. A digital composite video encoder, comprising: a luma input for receiving a digital luminance signal; a luminance digital-to-analogue converter coupled to the luma input to convert the digital luminance signal into an analogue luminance signal; chroma input means for receiving digital chrominance signals representing two colour difference signals respectively; phase calculator means for calculating colour subcarrier phase for each pixel of the composite video; preprogrammed lookup table means for each of the colour difference signals connected with some of its address inputs coupled to receive the respective chrominance signal and the rest of its address inputs coupled to the output of the phase calculator means for providing chroma outputs to form a modulated composite chrominance output;; chrominance digital-to-analogue converter means coupled to the outputs of the lookup table means; and analogue combining and filtering means for combining analogue luminance and chroma signals and adapted to output a composite video signal therefrom.

2. An encoder according to claim 1, including a compensating delay in the luminance path between the luma input and the combining and filtering means.

3. An encoder according to claim 1 or 2, in which the chroma input means comprises a single chroma input and a demultiplexer connected to the input.

4. An encoder according to claim 1, 2 or 3,in which the lookup table means comprises a static RAM, and including means for reprogramming the RAM.

5. An encoder according to any preceding claim, in which the phase calculator comprises a line counter, a pixel counter for counting the pixels along a line, a phase-in-line lookup table coupled to the output of the pixel counter, a phase- of-line-start lookup table coupled to the output of the line counter, and combining means for combining the output of the phase-in-line lookup table and the phase-of-line-start lookup table.

6. An encoder according to claim 5, in which the lookup tables in the phase calculator means are reprogrammable.

7. An encoder according to any preceding claim, in which the combining and filtering means comprises a first combining and filtering circuit coupled to receive the analogue chrominance signals and to provide bandpass filtering thereto and a second combining and filtering circuit coupled to receive the output of the first combining and filtering circuit and the luminance signal and to provide low-pass filtering thereto.

8. An encoder according to claim 7, in which the first combining and filtering circuit comprises two sets of combining circuits and filter circuits of which one set is selectively usable at a time.

9. A digital chrominance signal encoder, comprising: chroma input means for receiving digital chrominance signals representing two colour difference signals respectively; phase calculator means for calculating colour subcarrier phase for each pixel of the composite chrominance output; preprogrammed lookup table means for each of the colour difference signals connected with some of its address inputs coupled to receive the respective chrominance signal and the rest of its address inputs coupled to the output of the phase calculator means for providing chroma outputs to form a modulated composite chrominance output; digital-to-analogue converter means coupled to the outputs of the lookup table means; and analogue combining and filtering means for combining and band pass filtering the analogue chrominance signals.