JPH04323988A - Sampling rate conversion system - Google Patents

Abstract

PURPOSE:To attain frame rate inverse conversion without deterioration by using a regular matrix arithmetic operation in the case of sampling rate conversion at a sender side so as to convert a data of a low rate input signal into a high rate output signal. CONSTITUTION:Three primary color signal series R, G, B of 24 frames/sec are outputted in the sequential scanning in matching with a film frame from an image pickup device 1. The R, G, B signals are given to a YIQ conversion circuit 2, in which the signals are converted into a luminance signal Y and color difference signals I, Q by matrix calculation. A frame rate conversion circuit 3 converts the signals into a jumping scanning signal series of 30 frames/ sec so as to be in matching with a television signal to be sent and an NTSC encoder 4 encodes the Y, I, Q signals into the transmission signal.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sampling rate conversion method, and more particularly to a sampling rate conversion method suitable for use in converting a frame rate from a movie film or the like to a television signal.

0002

[Prior Art] The number of frames per second for television film is generally 24 or 25 (CCIR Recommendation 265-
5). In addition, ISO which specifies movie cameras
24 and 25 frames per second are also accepted as the number of frames per second for television film (ISO23, IS
O25).

[0003] When converting a film image at 24 frames per second to an NTSC television signal (30 TV frames/second = 60 TV fields/second), the first film frame is generally divided into two television fields, and the next film frame is divided into two television fields. A method of distributing the frames into three fields and then distributing them in the same manner as 2, 3, 2, 3, etc. (
It is known that frame rate conversion between film and television is performed using the 2-3 pulldown method (hereinafter abbreviated as 2-3 pulldown method) (see, for example, Japanese Patent Laid-Open No. 64-49388).

[0004] Utilizing the field repetition property of a television signal converted from a film image (hereinafter abbreviated as a telecine signal) as described above, high-quality signal processing such as sequential scanning can be realized (for example, , JP-A-2
(Refer to Publication No. 199969).

To achieve this, it is necessary for the receiving side to identify whether the transmitted signal is a general television signal or a telecine signal. As a method for identifying only on the receiving side without transmitting a special identification signal from the transmitting side, there is, for example, a method described in Japanese Patent Application Laid-Open No. 2-199970.

[0006]

[Problem to be Solved by the Invention] Telecine signals converted using the 2-3 pulldown method have two identical frames of film.
Since the period of one field or three fields is continuous, image quality deterioration (stroboscopic interference) occurs in which the movement of the moving image portion becomes unnatural. For this reason, telecine devices that perform frame rate conversion by frame interpolation on the transmitting side to reduce the above-mentioned interference are becoming popular. However, with the conventional technology described in the above-mentioned Japanese Patent Application Laid-Open No. 2-19970, it is impossible to identify telecine signals that have undergone signal processing in the frame direction, so high-quality signal processing using the properties of telecine signals is not possible. The drawback was that it couldn't be done.

[0007] The object of the present invention is that even if the telecine signal has been subjected to frame rate conversion processing such as frame interpolation,
It is an object of the present invention to provide a sampling rate conversion method that enables detection of a telecine signal on the receiving side and inverse frame rate conversion without deterioration.

[0008]

SUMMARY OF THE INVENTION Scanning processing of television signals is vertical and temporal sampling. To achieve the above purpose, frame rate conversion (
When performing sampling rate conversion (sampling rate conversion), regular matrix operations are used to convert data from a low rate input signal to a high rate output signal. On the receiving side, inverse frame rate conversion is performed using inverse matrix calculation.

[0009]

[Operation] The principle of operation of the present invention will be explained with reference to the drawings. As shown in FIG. 2, an example will be given in which a 24 frame/second film image is converted into a 60 field/second television image on the transmitting side. At this time, for the sake of simplicity, the conversion to the first field indicated by the thick solid line arrow and the conversion to the second field indicated by the thin broken line arrow will be explained separately.

Let us take as an example the case of frame interpolation using linear interpolation. The frame series of the input film image is A, B,
C, D, E... and the transmission signal sequence of the first TV field is u1, v1, w1, x1, y1, z1..., frame rate conversion by linear interpolation is performed by matrix calculation as shown in Figure 3(a). can be expressed. Also, the second TV
Assuming that the field transmission signals are w2, x2, y2, z2, . . . , they are expressed as shown in FIG. 3(b), and the calculation is the same as that of the matrix shown in FIG. 4(a), although the phase is shifted. Therefore, only the first TV field will be described below.

The matrix shown in FIG. 3A does not have a regular square matrix and cannot be inversely calculated because the input signal has a low frame rate and the output signal has a high frame rate. Therefore, as in the matrix calculation shown in FIG. 5(a), the matrix calculation is rewritten by overlapping the input film frame C with C1 and C2. This only changes the way the matrix calculation is viewed, and there is no change in the actual signal processing. The matrix in Fig. 5(a) is regular, and the matrix in Fig. 5(b) is regular.
), the inverse operation becomes possible. Therefore, from the transmission signal whose frame rate has been converted by frame interpolation,
The original film frame image can be completely reproduced. Either of the reproduced frames C1 and C2 is
Alternatively, the average value of both may be used as frame C. The same applies to the second TV field. This frame rate inverse conversion method is shown in FIG.

In order to perform the frame rate inverse conversion, it is necessary to detect that the image is a telecine image and to detect the frame phase on the receiving side. When the matrix calculation shown in FIG. 5(b) is performed, frames C1 and C2 can be reproduced independently from the transmitted signals u1 to z1. However, since C1 and C2 are originally the same signal, these two signals can be compared and if they match, it can be identified as the transmission signal of the present invention. Furthermore, since this coincidence occurs once every five TV frames, it is possible to detect the frame phase from this phase.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Examples of the present invention will be described below with reference to the drawings.

FIG. 1 shows a block diagram of a transmitting side according to a first embodiment of the present invention. The imaging device 1 (telecine device) outputs a three-primary color signal series R, G, and B at 24 frames/second in sequential scanning that matches the film frame. These R, G, and B signals are converted into a luminance signal Y and color difference signals I and Q by a predetermined matrix calculation in a YIQ conversion circuit 2. A frame rate conversion circuit 3, which will be described later, converts it into a 30 frame/second interlaced scanning signal sequence to match the television signal to be transmitted, and an NTSC encoder 4 converts the Y, I, and Q signals in the same way as in the current television system. Encode it and use it as a transmission signal.

The operation of the frame rate conversion circuit 3 described above will be explained with reference to FIG. Here, interpolation processing is performed between film frames by interpolation using coefficients as shown in the figure, for example, to convert into a TV field. If this is expressed by matrix calculation, it can be written as shown in FIG. This matrix operation is performed by overlapping the input film frame C to form C1 and C2, as shown in FIG. 5(a).
) can be rewritten as a regular matrix operation.

FIG. 4 shows a block diagram of the receiving side of the first embodiment of the present invention. Commonly used NT
The SC decoder 5 separates the Y, I, and Q signals from the transmission signal to form an interlaced scanning signal sequence of 30 frames/second. A frame rate inverse conversion circuit 6, which will be described later, converts the signal into a 24 frame/second progressive scanning signal sequence, and a later-described scan conversion circuit 7 converts it into a 60 frame/second progressive scanning signal sequence. The RGB conversion circuit 8 performs a predetermined matrix calculation to convert the signal into the original three primary color signal series, and the image is reproduced on the monitor 9.

The operation of the above-mentioned frame rate inverse conversion circuit 6 will be explained with reference to FIG. On the sending side, Figure 5(a)
Since frame rate conversion is performed by the matrix calculation shown in FIG. 5B, the receiving side performs frame rate inverse conversion by the inverse matrix calculation shown in FIG. 5(b) to reproduce the original 24 frames/second signal sequence. FIG. 6 shows this frame rate inversion method.

FIG. 7 shows a configuration diagram and an operation explanatory diagram of an embodiment of the above-described scan conversion circuit 7.

This is constituted by a memory circuit 10 and a memory control circuit 11, and achieves desired scan conversion by controlling write and read operations of the memory circuit 10. For example, as shown in the figure, 1
A 2-3 pulldown image can be obtained by performing a memory write operation every film frame (1/24 second) and a memory read operation every 1 TV field (1 TV frame of progressive scanning = 1/60 second). Can be done. In order to reduce stroboscopic interference, interpolation processing may be performed from previous and subsequent frames. In this case, it is the same as an image obtained by sequentially scanning an interlaced scanning transmission signal using interframe interpolation (frame average interpolation, etc.), but interframe interpolation can always be performed regardless of whether there is movement in the image. , differs from the conventional method in that there is no deterioration in vertical resolution.

FIG. 8 shows a block diagram of the receiving side using the first embodiment of the present invention. This is an example of a receiving side configuration when a general television signal and a telecine signal are transmitted in a mixed manner. In the figure, an NTSC decoder 5 separates a luminance signal Y and color difference signals I and Q from the transmission signal. These signals are input to the above-mentioned frame rate inverse conversion circuit 6, and are then input to the above-mentioned scan conversion circuit 7.
A signal sequence 100 in a progressive scanning format at 60 frames/second is generated. On the other hand, the output signal of the NTSC decoder 5 is also input to the motion adaptive scan conversion circuit 12 . The motion adaptive scan conversion circuit 12 is a circuit commonly used in clear vision receivers, etc., and detects the movement of an image and sequentially scans the image by switching between interframe interpolation and intrafield interpolation according to the amount of movement. This circuit generates a signal sequence 101 in a progressive scanning format at 60 frames/second. Further, a telecine detection circuit 14 (to be described later) detects whether the transmission signal is a general television signal or a telecine signal, and a switch 13 is used to select a signal series 100 and a signal series 101, which are input to the RGB conversion circuit 8. . The three primary color signal series R, G, and B obtained here are input to the monitor 9 to obtain a reproduced image.

FIG. 9 shows an embodiment of the telecine detection circuit 14 described above. Among the input signals, the low frequency components (components below 2 MHz) on which the modulated color signal is not multiplexed are processed by the one frame delay circuit 15 for one TV frame period (=5
25 horizontal scanning period = 1/30 seconds), u1
A signal sequence of ~z1 is obtained. Coefficient calculation circuits 16 and 17
Using , the signals u1, v1, w1 and the signals x1, y1, z1 are multiplied by predetermined coefficients and then added to obtain signals c1 and c2. Comparing circuit 1 compares these two signals.
8, if the input signal is a telecine signal, they match once every 5 TV frames. This is detected by the 5-frame period & phase detection circuit 19, and output as a detection result indicating whether it is a telecine signal or not.

FIG. 10 shows a second frame rate conversion method of the present invention. This is an example of a case where the first frame rate conversion coefficients are different. The method described above converts 2 frames of film into 5 fields of television.
- Similar to 3 pulldown, but uses the average value of 2 film frames for the center field of the 5 field period. In this case, the matrix used for frame rate conversion on the transmitting side becomes a unit matrix as shown in FIG. 11, and the matrix used for inverse frame rate conversion on the receiving side also becomes the unit matrix shown in FIG.
It becomes an identity matrix like . At this time, the transmitted signal u
If we detect whether the average value of 1 and w1 is equal to v1,
It is possible to identify whether the signal is a telecine signal or not.

[0023] Furthermore, a plurality of matrices, such as the first receiving side matrix used in the first frame rate inversion method of the present invention and the second receiving side matrix used in the second frame rate inversion method, Prepare the calculation coefficients on the receiving side,
By applying them simultaneously or sequentially to the transmitted signal, it is also possible to identify which matrix transform coefficients were used on the transmitting side.

FIG. 13 shows an example of a method for converting the frame rate of a telecine signal converted from a 25 frame/second film image. In this case, 5 frames of film are converted into 12 TV fields (=6 TV frames). To do this, for example, as shown in the figure, rate conversion can be performed by making each film frame correspond to 2, 3, 2, 3, 2 fields of television. On the receiving side, it is possible to detect that the frame difference becomes zero only when 3 TV fields are created from 1 film frame, so it is possible to detect that this occurs at a predetermined period (repetition every 5 and 7 TV fields). identify

FIG. 14 shows the present invention in the NTSC system and PAL system.
A fourth embodiment will be described in which the method is applied to scan conversion. The effective number of scanning lines in the NTSC system is approximately 480 (240 per field), while the effective number of scanning lines in the PAL system is approximately 576 (288 per field). Therefore, if both are expressed as a simple integer ratio, it becomes 5:6. Now, let us consider the case where an NTSC television signal is first converted to PAL and then converted back to NTSC. At this time, either 240 lines - 288 lines - 240 lines are converted using the present invention for each field, or 480 lines - 57 lines are converted to a sequential scanning format using motion adaptation processing etc.
It is sufficient to perform 6-480 line conversion. FIG. 14 shows only the first field when scan conversion is performed for each field.

FIG. 15 shows an example of matrix calculation used in the scan conversion shown in FIG. 14. In the same figure (a), the NTS
The scanning line series A to F of the C-system television signal are input, and the scanning line series t to Z of the PAL-system television signal are input.
An example of a matrix operation for transmitting scan conversion to generate . In this, the scanning line of the PAL signal is generated by linear interpolation, and the signal ((C
+D)/2) is duplicated. On the receiving side, the original NTSC scanning line series A to F can be reproduced by the matrix calculation shown in FIG. By checking whether they match or not, it is possible on the receiving side to identify whether the transmission signal is a transmission signal of this method or not.

Although the present invention has been described using frame rate conversion and scan conversion as examples, the present invention can be applied to general sampling rate conversion in addition to the above.

[0028]

[Effects of the Invention] According to the present invention, when transmitting a low sampling rate signal through a high sampling rate transmission path, even if sampling rate conversion is performed by filter processing such as linear interpolation on the transmitting side, the receiving side Since complete inverse conversion characteristics can be achieved, the original signal sequence can be reproduced without deterioration. In particular, if the present invention is applied to television broadcasting of telecine signals, automatic detection of telecine signals, progressive scanning without deterioration in vertical resolution, etc. can be realized, and the effect of improving image quality is very large.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a transmitting side according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram of a first frame rate conversion method of the present invention.

FIG. 3 is an explanatory diagram of matrix calculation used in the first frame rate conversion of the present invention.

FIG. 4 is a block diagram of the receiving side of the first embodiment of the present invention.

FIG. 5 is an explanatory diagram of matrix calculation used in the first frame rate inversion of the present invention.

FIG. 6 is an explanatory diagram of the first frame rate inversion method of the present invention.

FIG. 7 is a more detailed embodiment of the circuit shown in FIG. 4;

FIG. 8 is an application example of the first embodiment of the present invention.

FIG. 9 is a more detailed embodiment of the circuit shown in FIG. 8;

FIG. 10 is an explanatory diagram of a second frame rate conversion method of the present invention.

FIG. 11 is an explanatory diagram of matrix calculation used in the second frame rate conversion of the present invention.

FIG. 12 is an explanatory diagram of matrix calculation used in the second frame rate inversion of the present invention.

FIG. 13 is an explanatory diagram of a third frame rate conversion method of the present invention.

FIG. 14 shows a fourth embodiment of the present invention.

FIG. 15 is an explanatory diagram of matrix calculation used in the fourth embodiment of the present invention.

[Explanation of symbols]

1... Imaging device; 2... YIQ conversion circuit; 3... Frame rate conversion circuit; 4... NTSC encoder; 5... NTSC decoder; 6... Frame rate inverse conversion circuit; 7... Scan conversion circuit; 8... RGB conversion circuit; 9... Monitor; 10...Memory circuit; 11...Memory control circuit; 12...Motion adaptive scan conversion circuit; 13...Switcher; 14...Telecine detection circuit; 1
5...1 frame delay circuit; 16, 17...coefficient calculation circuit;
18...Comparison circuit; 19...Period & phase detection circuit.

Claims (1)

[Claims] Claim 1: Converting a first signal sequence with a low sampling rate to a second signal sequence with a high sampling rate using a regular matrix operation, and converting the signal sequence from the second rate to the first rate using an inverse matrix operation. A sampling rate conversion method that performs inverse sampling rate conversion.