GB2031687A - Television standards conversion - Google Patents

Abstract

A video standards converter in which digital persistence is effected by including first and second picture stores (22). First and second coefficient blocks (300, 302) are connected to the first and second stores respectively and the outputs of the blocks added in an adder (301). Coefficients are selected for use in each block to determine the portion of information passed on to the adder. These coefficients may be varied cyclically and/or in dependence on any picture movement to provide variation in the degree of persistence applied. <IMAGE>

Description

SPECIFICATION Television standards conversion The invention relates to a system suitable for use as a video standards converter.

Digital standards converters provide the capability to change picture information from one standard to another e.g. PAL to NTSC.

The present invention is concerned with providing an enhanced standards conversion system.

According to the invention there is provided a video standards converter including a digital persistence arrangement comprising first and second picture information storage means; first coefficient means connected to the output of the first storage means for modifying the output thereof by a selected coefficient; second coefficient means connected to the output of the second storage means for modifying the output thereof by a selected coefficient; and adder means for adding the outputs from the first and second coefficient means.

Further according to the invention there is provided a method of providing digital persistence in a standards converter comprising storing picture information in a first store; storing information from the first store in a second store and adding selected portions of the information from each store.

The invention will now be described by way of example with reference to the accompanying drawings, in which: FIGURE 1 shows an earlier standards converter arrangement, FIGURE 2 shows movement of an object through various fields, FIGURE 3 shows an arrangement for effecting cyclic persistence variation, FIGURE 4 shows one arrangement for providing the enhanced persistence system of the invention, FIGURES 5 and 6 show alternative arrangements for one portion of the system of Figure 4.

Figure 1 shows an arrangement suitable for use as a standards converter, which arrangement is described in detail in British Patent Application 42751/76 (U.S. application 841 519).

Incoming video data is applied to an input processor 20. The incoming video will be in digital form, first having been converted from analogue form as necessary. The input processor 20 effects the synthesis of picture points from adjacent picture point information from at least one field. The number of picture points synthesised can be controlled to be greater orsmallerthan the original number of picture points for a particular field or frame thus effecting expansion of compression. For example, PAL to NTSC standards conversion will require taking an incoming line standard of 625 lines per frame and converting this to an outgoing standard of 525 lines per frame. Thus the number of lines is reduced and is comparable with a fixed amount of compression.

The way in which the picture points are synthesised within processor 20 is described in more detail in the aforementioned patent applications.

The data from processor 20 is received by modifier 210 where the data is modified by a coefficient K11 before receipt by adder 211. This data is added to previously stored data which has first been modified by coefficient K12 in modifier 212. The output of adder 211 passes via modifier 213 which modifies the data by coefficient K13 before storage in the store 22.

The reasons for this modification will now be explained.

In addition to the change in the number of lines for a frame the number of fields per second will change.

For European PAL the field rate is 50 fields/sec and NTSC used in USA is 60 fields/sec. The reduction in the number of lines is effected by the input processor as already mentioned and the increase in field rate can be accommodated due to the asynchronous nature of the frame store which allows different write in and read out rates. A proportion of the data from the old frame is combined with a proportion of the new frame using the 'leaking' integrator system i.e. coefficients K1 1, 12, 13 to produce smooth movement (movement interpolation) at different frame rates.

In the digital standards converter it is required to modify the coefficients k11, k12 and k13 on a cyclic basis. The effect of using a cyclic variation may be utilised to provide 'movement interpolation'. The digital standards converter requirement alters depending upon the type of scene under examination. Movement interpolation minimises distortion due to movement since standards conversion necessitates conversion from say 525 lines60 fields per second (NTSC) to 625 lines50 fields per second (PAL).

If an incoming picture having an incoming field rate of 60 fields per second carries a moving image, then successive pictures provide a general increment of the image moving from left to right on the screen. The outgoing pictures concerned with the outgoing fields are at 50 per second. In an ideal movement interpolator an image of the moving object provided on the outgoing picture would need to be provided in a position which is not depicted on the incoming fields.

Such a moving object is shown in Figure 2. Successive pictures at 60 Hz field rate are shown at Figure 2(a), (b), (c) and (d) and successive pictures at 50 Hz field rate are shown at (e), (f) and (g). In this example field (a) has been chosen to be in time coin cidencewith (e); likewise (d) with (g) and in that time frame the circular object has moved completely across the field of view (FOV) so in (g) to (c) it has moved 1/3 the FOV and 2/3 respectively butjn the 50 Hz system (f) requires the object to have moved FOV. Clearly no object is available in (a), (b), (c) or (d) showing it ata FOV.

A reasonable simulation of movement interpolation may be implemented by utilising components from several fields in various proportions to artifi cally delude the viewer into believing he is seeing true movement interpolation.

The system of movement interpolation used in the above system makes use of three differenttechni quest portray movement in the most acceptable subjective manner. The system is variable and adaptive. Techniques have been developed which enable the adaption system to be semi-automatic.

The three techniques utilised are categorised below: a) Field sequence interchange b) Digital persistence c) Digital cyclic persistence variation The three techniques are now described.

The incoming picture field sequence may be considered to start at a known position in time with respect to the outgoing picture field sequence. As the pictures arrive sequentially the error which occurs in the outgoing picture field sequence increases over a period of twelve incoming fields and ten outgoing firlds in a 60 field to 50 field converter.

The assumption has been made in the above paragraph that the two different types of field (odd field and even field) cannot be interchanged in the sequence. The resultant discontinuous movement is clearly visible to the observer.

If the field sequence is changed so that an odd field is being displayed as an even field the amplitude of the discontinuity can be reduced by a factor of two.

The input processor 20 using adaptive volume manipulation described in the above patent applications is capable of accurately changing an odd field to an even field or an even field to an odd field without introducing distortion of the fixed and moving image. In such a standards converter use is made of the adaptive volume manipulation to undertake a field sequence variation thus reducing the amplitude of the visible movement discontinuities.

An improvement to the subjective effect of movement discontinuity is provided by using digital persistence. A normal television system has a small amount of residual memory due to the effect of phosphor decay. A controlled persistence is introduced in the standards converter by utilising the frame store memory together with coefficients which control the persistence. Thus coefficients k1 1, k12 and k13 dictate what proportion of the previously stored picture is added to the new picture available from the incoming data stream before restoring in the memory. It is possible to arrange the system with only two ports but more typically three will be utilised. Port 1 is an input port allowing data to be entered into the framestore. Port 2 is an output port which may be considered synchronous with port 1 allowing information to be extracted from the memory.Port3 is an asynchronous output port which allows the input and output system to be run at different speeds.

In the basic digital persistence system coefficients k1 1, k12 and k13 are fixed. Typical values are shown below: k11 = 0.625 k12 = 0.375 k13 = 1.0 The movement portrayal utilising digital persistence brings the subjective effect closer to the ideal.

However, different viewers could choose different values for the coefficients which suits them best.

Variation ofkll, k12 and k13 by a control function which may be accessible to the viewer would provide a means of selecting such values.

The effect of movement in a standards converter gives rise to a cyclic discontinuity. Major improvements to the subjective effect may be provided if the digital persistence is varied in a similar cyclic pattern. A typical cycle for k1 1 is shown below: Field kil 1 0.5 2 0.625 3 0.75 4 0.875 5 1.0 The cycle repeats every 5 fields at the output. Once again subjective viewers may choose different values of k's to suit their own viewing requirements and provision may be made to alter the value of k under the control of the viewer.

A way in which the cycle can be effected is shown in Figure 3. A cyclic persistence control 220 incorporating address counters receives incoming field pulses and outgoing field pulses and provides an address as a result of the comparison. The address provided is used to look up k1 1, k12 and k13 in a read only memory (ROM) 221. The ROM has a number of coefficients stored which may be accessed by the address. In addition the ROM has an input for varying the persistence control. The persistence control variation input selects a group of coefficients stored in the ROM. If a single ROM is not large enough to accommodate all the coefficients needed a number of ROM's may be simultaneously addressed and selected by the persistence control variation input.

The operation and control of a ROM is well known for digital processing.

The selected coefficients at the ROM output for k1 1, k12 and k13 are received by modifiers 210,212 and 213. The multiplying function of the modifier may be provided by using a random access memory for example.

As a refinement to improve the system flexibility, one or more suitable programmed digital microporcessors could be used to calculate the required coefficients.

In addition, they can be used to determine the address locations fixed in store and to calculate the interaction of the controls for compression or expansion, digital persistence and other processing functions with the store locations and hardware coefficient look-up tables.

In the preceding descriptions, digital persistence and digital cyclic persistence were provided within a single store using the system within block 100 of Figure 1 for example.

The present invention is concerned with providing an improved system suitable for standards conver sion as now described.

In the present system an alternative form of persistence characteristic may be obtained by utilising more than one store. The method using one store for convenience will be called a recursive filter: the present method using more than one store combines both recursive and non-recursive filtering and has been found to give improved results.

When the coefficient K11 is set to 0.5, the lag causes an undesirable visual effect on movement.

By connecting two similar systems in series, it becomes possible to add proportions of the two pictures together in a non-recursive method as shown in the Figure 4 embodiment.

The system includes two stores 22 each with associated digital persistence components 210, 211, 212,213. Thus dual systems each corresponding to that of block 100 of Figure 1 are included. In addition the output of port 3 of store P is received by coefficient unit 200 for coefficient KP and store Q output is received by unit 302 for coefficient KQ. Their outputs are added in digital adder 301 to provide the video data system output. The operation of coefficients KP and KQcan be effected using the multipliers and look-up system described earlier.

The systems shown in Figure 4 are typically set to have K1 1 at 0.875. The addition of the two pictures takes place in an adder 301 arranged to take a proportion KP and KQ from the two stores. KP and KQ are then cyclically varied in order to simulate the effect of movement interpolation. Typical values of KP and KO are shown below for a 60 to 50 conversion.

Field KP KQ 1 0.5 0.5 2 0.375 0.625 3 0.25 0.75 4 0.125 0.875 5 0.0 1.0 For convenience, two identical systems have been shown in series for ease of explanation. The requirements of store Q are not as stringent as those of store P, which is required to operate in a nonsynchronous manner between input and output.

Further, it is not necessary to utilise the recursive filter elements around store Q, which need only act as a simple delay store.

The addition of stored pictures through nonrecursive filters provides a method of producing noise reduction in television signals. Although this is not a prime objective of the above non-recursive system of the standards converter, it does take place and is considered a useful feature.

If the standards converter is required to change the number of lines/frame then a processor along the lines of block 20 will be required but this is not directly concerned with digital persistence.

In addition to the cyclic variation in the various coefficients described in relation to Figure 4 the coefficients can also be varied to control the degree of persistence in dependence on the degree of any movement detected by including a movement detector which can additionally control the coefficient selected from the ROM for example and as described in British Patent Application 42751/76. Although Figures 1 and 4 shows stores with associated elements for coefficients K11, K12 and K13 these coefficient arrangements could be modified to provide only one variable coefficient (K11) as shown in the systems of Figures 5 and 6. A subtractor 230 is now included and as shown in Figure 5 the processed data is received by the subtractor together with the output from port 2 of the store.

This output also passes to adder 211. In the Figure 6 arrangement the store output from port 2 only passes directly to the subtractor. The incoming video is now received directly by the adder 211.

In practice the system could be further simplified with reduced flexibility so as to comprise units 300 (KP) and 302 (KQ) and their associated adder 301, and stores P and Q. Alternatively to such a simplified system it would be possible to include only one coefficient unit for feedback along the lines of Figure 5 or 6 for only one of the stores P and Q, as already suggested concerning Figure 4.

Although the system described has shown separate storage and processing elements, the system could be constructed in sectors each with associated processing to provide a distributed system along the lines dislosed in U.S. patent applications 841,519 and 015,678.

Claims (16)

1. A video standards converter including a digital persistence arrangement comprising first and second picture information storage means; first coefficient means connected to the output of the first storage means for modifying the output thereof by a selected coefficient; second coefficient means connected to the output of the second storage means for modifying the output thereof by a selected coefficient; and adder means for adding the outputs from the first and second coefficient means.

2. A converter as claimed in claim 1, including third coefficient means connected between the output and the input of the first storge means for selectively providing feedback of at least a portion of the information held in the first storage means to provide additional persistence control.

3. A converter as claimed in claim 2, including fourth coefficient means connected between the output and the input of the second storage means for selectively providing feedback of at least a portion of the information held in the second storage means to provide additional persistence control.

4. A converter as claimed in claim 2, wherein the third coefficient means comprises an adder and at least one multiplier.

5. A converter as claimed in claim 3, wherein the fourth coefficient means comprises an adder and at least one multiplier.

6. A converter as claimed in claim 4 or 5, including a subtractor connected to the multiplier.

7. A converter as claimed in any one of claims 1 to 6, wherein the selected coefficient for one or more of the coefficient means is varied cyclically field by field.

8. A converter as claimed in any one of claims 1 to 7, wherein movement detector means are provided to control the coefficient selected for one or more of the coefficient means in dependence on any picture movement detected.

9. A converter as claimed in any one of claims 1 to 8, wherein the first and second storage means each comprise a solid state frame store.

10. A method of providing digital persistence in a standards converter comprising storing picture information in a first store, storing information from the first store in a second store and adding selected portions of the information from each store.

11. A method as claimed in claim 10, including feeding back at least a portion of the information held in the first store.

12. A method as claimed in claim 11, including feeding back at least a portion of the information held in the second store.

13. A method as claimed in claim 10,11 or 12 including cyclically varying the portion selected.

14. A method as claimed in any one of claims 10 to 13, including detecting any picture movement and varying the portion selected in dependence on any movement detected.

15. A video standards converter having a digital persistence arrangement substantially as described herein and with reference to the accompanying drawings.

16. A method of providing digital persistence in a standards converter substantially as described.