JP2893744B2 - Motion detection circuit - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to, for example, MUSE (Multiple Sub-Nyquist-Sampl
The present invention relates to a motion detection circuit suitable for use in a high-definition signal decoder of an ing Encoding system.

[Summary of the Invention]

The present invention relates to a television signal motion detecting circuit which uses an n-frame image signal such as a MUSE high-definition signal to form an image of one screen. A first memory for providing a period delay, a difference means for calculating a difference between an output signal of the first memory and the input television signal, and a delay of n frame periods for the output signal of the difference means A second memory is provided, and when the output signal of the second memory satisfies a predetermined condition, processing is performed by regarding the output signal of the difference means as a signal corresponding to a still image. This prevents a still image portion from being erroneously determined as a moving image portion not only for a signal but also for a high-frequency signal.

[Conventional technology]

At present, it is considered practical to perform so-called high-definition broadcasting using a broadcasting satellite. In this case, in order to transmit the Hi-Vision signal as it is, a signal bandwidth of about 20 to 25 MHz is required. On the other hand, the Hi-Vision signal is satisfactorily FM-modulated and transmitted on one channel of a broadcasting satellite (27 MHz bandwidth). To do so, the signal bandwidth of the HDTV signal must be 9 MHz or less. Therefore, the MUSE system has been developed for compressing the bandwidth of a Hi-Vision signal without losing the image quality and broadcasting it on one satellite channel.

According to the MUSE method, the luminance signal Y and the chrominance signal C are time-division multiplexed, and the time division multiplexed signal (TCI signal) is subjected to sub-sampling (decimation) of a field offset, a frame offset or a line offset, and a signal band The width is compressed to 8.1MHz. As a method of this band compression
In the MUSE method, so-called three-dimensional subsampling (thinning-out) is used. That is, as shown in FIG. 3, the HDTV signal output from the MUSE encoder has a phase difference of 180 ° between the sample point of the first frame (third frame) and the sample point of the second frame (fourth frame). The first frame and the second frame (or the third frame and the fourth frame) constitute complete image information for one screen. Therefore, in the MUSE method, complete image information for one screen is composed of signals for two frames (four fields).

However, a complete image for one screen is composed of the signals for two frames only when the original image is a still image, and when the original image is a moving image, a signal for one screen is obtained from the signals for the two frames. When the image is configured, there is a possibility that a double image or the like may occur in the reproduced image. Therefore, in the MUSE encoder, a still image band compression circuit and a moving image band compression circuit are provided as is well known, and the motion amount of the original image is detected for each pixel, and the motion amount is determined according to the detected motion amount. The output signals of the band compression circuits of the still image system and the moving image system are mixed at the mixing ratio and supplied to the transmission unit. In this case, on the encoder side, motion detection can be easily performed using the one-frame difference signal between the current frame and the previous frame before sub-sampling, whereas, as shown in FIG. Since there is no signal at the same sample point one frame before, motion detection becomes difficult.

MUSE restores the HDTV signal of about 20MHz bandwidth from the output signal (MUSE signal) of the MUSE encoder described above.
It is a decoder. FIG. 4 shows a main part of a conventional MUSE decoder. In FIG. 4, (1) is an input terminal,
This input terminal (1) has 1
MUSE signal MS _0, which is an analog / digital converter with 6.2MHz is supplied. This MUSE signal MS ₀ is a motion detection circuit (2),
The still image interpolator (3) is supplied to the still image interpolator (3) and the moving image interpolator (4). The still image interpolator (3) supports one frame of image by inter-frame interpolation and inter-field interpolation from signals of two frames. The moving image interpolating circuit (4) forms a signal corresponding to an image for one field from the signal for one field by intra-field interpolation, and supplies the signal to the mixing circuit (5). 5). In the mixing circuit (5), the interpolation circuit (3) has a mixing ratio corresponding to the motion detection signal KT supplied from the motion detection circuit (2).
And the output signals of (4) are mixed and supplied to a time division multiplexing (TCI) decoder (6).

In the motion detection circuit (2), MUSE signal MS ₀ is supplied to the input terminal, one input terminal of the one input terminal and the subtraction circuit subtracting circuit (8) (9) of the frame memory (7), a frame memory which is the output signal (7) 1-frame delayed signal MS ₁ is supplied to the other input terminal of the input terminal and the subtracting circuit of the frame memory (10) (8), two frames is an output signal of the frame memory (10) delayed signal MS ₂ is supplied to the other input terminal of the subtracting circuit (9), the output signal of the subtracting circuit (9) through the absolute value circuit (11) 2-frame difference signal delta ₂ as aND gate (12) It is supplied to one input terminal. The output signal of the subtraction circuit (8) is a low-pass filter circuit (13) having a cutoff frequency of 4 MHz,
Is converted absolute value circuit (14) and the frame memory (15) to delay one frame difference signal delta ₁ through the other of the AND gate delay 1 frame difference signal delta ₁ via the inverter circuit (16) (12) Is supplied to the input terminal of.

In this case, as described earlier with reference to FIG.
Since it is the signal before two frames exists signals of the same sample point is no signal of the same sample point before 1 frame, 2 frame difference signal delta ₂ is a signal corresponding to the amount of movement, delay 1 frame difference signal delta ₁ is completely is not a signal corresponding to the amount of movement. However, if the low-frequency component is extracted even if the sample points do not match, the sample points can be regarded as substantially matching. In the case of the MUSE signal, one frame is used for the low-frequency signal of 4 MHz or less. The difference between them is determined accurately. Since the Figure 4 example low-pass filter circuit (13) is provided, the delay 1 frame difference signal delta ₁ is a signal corresponding to the exact amount of motion for the following low-frequency signal 4MHz I have.

Further, the output signal K ₀ is a field memory of the AND gate (12) (17), (18), is converted is delayed by each one field delayed signal K _1, K _2, K ₃ (19), the The output signal K ₀ and the delay signals K ₁ , K ₂ , K ₃ are respectively supplied to an arithmetic circuit (20), and the arithmetic circuit (20) generates a motion detection signal KT by, for example, a logical OR operation. Although the circuit described above is the signal for simplicity is represented as a 1-bit digital signal, actually MUSE signal MS ₀ is for example 8 bits,
The motion detection signal KT is, for example, a 4-bit digital signal.

In the example of FIG. 4, the MUSE supplied to the input terminal (1)
The operation when the signal MS ₀ is a low-frequency signal of 4 MHz or less will be described. Assuming that the signal MS ₀ is a Hi-Vision signal relating to an object moving at a constant speed, the signal MS ₀ is on a certain horizontal scanning line at the current time t. Consider the signal as a function of position x.

In this case, the MUSE signal MS ₀ , one frame delay signal MS ₁ and 2
Frame delayed signal MS ₂ becomes as shown in respectively Figure 5 A, B and C, the delay 1 frame difference signal _{Δ 1 (| MS 2 -MS 1} |) and 2-frame difference signal Δ _{2 (|} MS ₂ -MS ₀ |) Is the fifth
The result is as shown in FIGS. Therefore, the output signal K ₀ of the fourth diagram of an AND gate (12) becomes as shown in FIG. 5 F for a logical product of the signal delta ₂ and the signal delta _1.

The output signal K ₀ Importantly regard to delays 1 frame difference signal delta ₁ by a high-level position P ₁ pulse (2
1) is that only a high level pulse position P ₂ is erased is left. This means that if an object is at position P ₁
When moved to a position P ₂ from the position P ₁ is determined with only the moving image portion position P ₂ means that determines the still image portion.

Similarly, the delay signal K ₁ ~K ₃ comprising delaying one by one field its output signal K ₀ becomes as shown in respectively Figure 5 G to I, is output from the final motion detecting circuit (2) The motion detection signal KT becomes as shown in FIG. 5J.
, The area (22) is determined to be a moving image part and the area (2)
3) is accurately determined as a still image portion. The reason why the motion detection signal KT is synthesized by using the signals K _{0 to} K ₃ for four fields is that the still image interpolation circuit (3) of the MUSE decoder uses one signal from signals for two frames (for four fields). Since a complete image for the screen is formed, even if the current field is determined to be a still image part, the part determined to be a moving image part in the field one to three fields before must be determined to be a moving image part. It is.

[Problems to be solved by the invention]

As described above, for the low band signal of 4 MHz or less, the example of FIG. 4 can accurately detect the motion. However,
To delay 1 frame difference signal delta ₁ in the conventional motion detecting circuit (2) does not respond with respect to high-frequency signal exceeding 4MHz, with respect to high-frequency signal exceeding the 4MHz determines that the moving image portion by mistake still image portion There was a risk of inconvenience.

That is, in the fourth illustrated example, each unit signal when the MUSE signal MS ₀ supplied to the input terminal (1) is a high-frequency signal exceeding the 4MHz becomes as shown in Figure 6, the output of the Figure 6 F as is apparent from the signal K _0, objects even when moving from the position P ₁ to position P _2, it may remain pulse (24) pixel unrestrained at position P _1. Similarly, the delay signals K _{1 to} K ₃ (FIG. 6G
To I), the pulses two frames before remain, respectively, so that the finally obtained motion detection signal KT (FIG. 6J)
In addition to the original moving image portion (25), a high-level "1" pulse is generated also in the original still image portion (26), which is determined as a moving image portion. As described above, when the part which is originally a still image is determined to be a moving image part, the output signal of the moving image interpolation circuit (4) is selected in the mixing circuit (5) in FIG. There is a disadvantage that the obtained image is partially blurred.

As the motion detection circuit, various systems have been proposed for the normal NTSC-M system and the like (for example, see JP-A-63-90988).
Japanese Patent Application Laid-Open No. Sho 62-175093, Japanese Utility Model Application Laid-Open No. Sho 62-203583, and Japanese Patent Application Laid-Open No. Sho 61-70888), and no other suitable system has been proposed for the MUSE type Hi-Vision signal.

In view of the foregoing, the present invention provides a motion detection circuit for a television signal that composes an image of one screen using an image signal of a plurality of frames, such as a high-definition signal of the MUSE system. It is another object of the present invention to prevent a still image portion from being erroneously determined as a moving image portion even for a high frequency signal.

[Means for solving the problem]

A motion detection circuit according to the present invention, as shown in FIG. 1, for example, detects a motion of a television signal by using an image signal of n frames (n is an integer of 2 or more) to form an image of one screen. in the circuit, a first memory for delaying the n frame period in the television signal MS ₀ input (28), an output signal MS ₂ and the input television signal MS of the first memory (28) A difference means (29) for obtaining a difference from _0, and an output signal (Δ ₂ ,
Delta _4, etc.) is provided a second memory for delaying the n-frame period (35), the output signal of the second memory (35) delta
₃ satisfies a predetermined condition (for example, when a predetermined level is exceeded), the output signal (Δ ₂ , Δ _4, etc.) of the difference means (29) is regarded as a signal corresponding to a still image, and processing is performed. (For example, the output signal is forcibly set to zero level).

[Action]

According to the present invention, a primitive signal corresponding to the motion is first obtained by the difference means (29). If the primitive signal in the example where the object is moved from the position P ₁ to position P _2, the portion P ₁ originally not only the position P ₂ became a still image portion
Is also determined to be a moving image portion. Therefore, in the present invention a signal corresponding to the still image primitive signal of the difference means (29) at that location P ₁ when the output signal delta ₃ of the second memory (35) satisfies the predetermined condition As a result, it is possible to prevent a still image portion from being erroneously determined as a moving image portion.

In this case, since the present invention does not use a low-pass filter circuit, a moving image portion can be accurately detected not only for a low-band signal but also for a high-band signal.

〔Example〕

Hereinafter, an embodiment of the motion detection circuit according to the present invention will be described with reference to FIGS. 1 and 2. FIG. In this example, the present invention is applied to a motion detection circuit of a MUSE decoder (that is, in the case of n = 2 of the present invention), and FIGS. 1 and 2 correspond to FIGS. Portions and signals are denoted by the same reference numerals, and detailed description thereof will be omitted.

FIG. 1 shows the motion detection circuit of the present embodiment. In FIG. 1, (27) shows an input buffer, and FM is demodulated by a receiving unit (not shown) through the input buffer (27).
M _1- bit MUSE converted from analog to digital at MHz
The signal is supplied to an input port of a two-frame memory (28) for giving a delay of two frame periods and one input port of a subtraction circuit (29), and a two-frame delayed signal MS as an output signal of the two-frame memory (28) ₂ was supplied to the other input port of the subtracting circuit (29) supplies the m ₂ bits of the output signal of the subtracting circuit (29) (MS ₀ -MS ₂₎ to the absolute value circuit (30). This absolute value circuit (30) has its output signal (MS ₀ -M
S ₂₎ to after conversion into an absolute value, the absolute value is taken out the top m ₃ bits of the second frame difference signal Δ _{2 (=} | MS ₀ -MS
₂ |), and supplies the two-frame difference signal delta ₂ to the sensitivity setting unit (31). This sensitivity setting device (31) has its m ₃
It converts the bit of the difference signal delta ₂ of 4 bits to the 2-frame difference signal delta _4. In this example, for example, m ₁ = 8, m ₂ =
9, m ₃ = 6 can be set. As a method of converting m ₃ bits of the difference signal delta ₂ to 4-bit differential signal delta _4, in addition to for example may be performed logarithmic compression by a method such as taking out the upper four bits of the m ₃ bits.

Supplying a signal of each bit of the difference signal delta ₄ obtained at its sensitivity setter (31) to one input terminal of each AND gate (32A) ~ (32D), the digital comparison of the difference signal delta ₄ simultaneously The digital comparator (33) supplies 4-bit reference data corresponding to a predetermined threshold to the inverting input port of the digital comparator (33). That is, the digital comparator (33) is 4-bit two-frame difference signal delta ₄ is the differential signal delta ₄ becomes the high level "1" when it exceeds the threshold is low when: the threshold "0" Is converted into a 1-bit two-frame differential signal, and the one-bit two-frame differential signal is supplied to a two-frame memory (35) for giving a delay of two frame periods. Delay of one bit at the output port of the second frame memory (35) 2-frame difference signal delta ₃
And gate (32A) through the inverter circuit (36)
To (32D) are commonly supplied to the other input terminals.

They AND gate (32A) ~ field into 4-bit output signal K ₀ occurring at the output terminal of the (32D) memory (17),
(18), successively applying one field period of the delay time, and supplies the delayed signal K ₁ ~K ₃ output signal K ₀ and 4 bits of the 4 bits in each arithmetic circuit (20) in (19), The arithmetic circuit (20) generates a 4-bit motion detection signal KT by, for example, performing a logical OR operation on each bit of the signals K _{0 to} K ₃ , and outputs the motion detection signal KT to the mixing circuit (FIG. 4). 5)
To supply.

As described above with reference to FIG. 3, in the MUSE system HDTV signal, only the motion detection signal for the low-frequency signal can be obtained because no signal of the same sample point exists in one frame. Since signals at the same sample point exist, a motion detection signal can be accurately obtained not only for a low-frequency signal but also for a high-frequency signal. That is, precisely the motion information for any of the high-frequency signal 2 frame difference signal delta ₄ of 4 bits output from the sensitivity setting unit of the first illustrated example (31) is greater than the following low-frequency signals and 4 MHz 4 MHz Contains.
Thus, the precise movement for any delay 2 frame difference signal delta ₃ also low frequency signal and the high signal made by assigning the delay time of two frame periods compresses the 4-bit signal delta ₄ to 1 bit Contains information.

In the example of FIG. 1, it is supplied to an input buffer (27).
The operation in the case where the MUSE signal MS ₀ is a high-vision signal relating to an object moving at a constant speed will be described with reference to FIG. 2. The value of each signal is displayed as a 1-bit digital signal. , Each signal on a horizontal scan line at the current time t is displayed as a function of the position x. Further, in this embodiment the MUSE signal MS ₀ operates in common be any of the low frequency signal and the high signal.

In this case, the MUSE signal MS ₀ , the two-frame delay signal MS ₂ , the delayed two-frame difference signal Δ ₃ (= | MS ₄ -MS ₂ |, MS ₄ is a four-frame delay signal) and the two-frame difference signal (= | MS ₂ -MS ₀
|) Becomes as shown in respectively Figure 2 to D, the signal K ₀ is a logical product of the signal delta ₄ and the signal ▲ ▼ becomes as shown in Figure 2 E. The important thing about this output signal K ₀ is that the delay 2
Frame difference signal delta ₃ inverted signal ▲ ▼ pulse of the high level "1" position P ₁ by the action (37) is that has been erased. Therefore, only the pulse of the high level "1" position P ₂ is left, the delay signal K ₁ ~K ₃ is unnecessary pulses of the previous two frames as shown respectively Figure 2 F~H is erased, The finally obtained motion detection signal KT (Fig. 2I)
, A high-level "1" signal is formed only in the moving image portion (38), and an unnecessary high-level "1" signal is not formed in the still image portion (39).

As described above, the output signal K ₀ at the position P ₁ is forcibly set to the zero level “0” corresponding to the complete still image portion by using the inverted signal ▼ of the delayed two-frame difference signal Δ ₃ , for example. object (object to the present time t is present at the position P _2.) when moving from the position P ₁ to position P _2, the position P ₁ to determine the position P ₂ and the moving image portion is determined as the still image portion That's why.
On the other hand, if only the two-frame difference signal Δ ₄ (FIG. 2D) is used, not only the position P ₂ but also the position P ₁ is determined to be a moving image portion, so that the position is determined using the delayed two-frame difference signal Δ _3. the P ₁ and forms as it is determined that the still image portion. In this sense, the delay 2 frame difference signal delta ₃ can also be considered as inhibiting signal of the two-frame difference signal delta _4.

As described above, according to this example, the two-frame difference signal Δ ₄
It is precisely the moving portion detected by using a delay 2 frame difference signal delta ₃ and, in particular, the input signal is 4M
There is an advantage that a still image portion is not erroneously determined as a moving image portion even with a high-frequency signal exceeding Hz. Therefore, blurring or the like of the image reproduced by the subsequent processing circuit can be reduced.

Further, according to this example, since the sensitivity setting device (31) and the level setting device (34) are provided, the detection sensitivity of the moving image portion can be variously adjusted, and the still image portion is erroneously determined as the moving image portion. There is an advantage that various kinds of sensitivity can be adjusted in order to suppress this.

In the above embodiment, the present invention is applied when n = 2. However, the present invention is not limited to this, and generally n frames (n = 2, 3, 4, 5,...) It is apparent that the present invention can be applied to a motion detection circuit for a television signal in which a complete image for one screen is constituted by using the above image signal.

As described above, the present invention is not limited to the above-described embodiment, and may adopt various configurations without departing from the spirit of the present invention.

〔The invention's effect〕

According to the present invention, since only a difference signal between n frames is used, there is an advantage that a still image portion is not erroneously determined as a moving image portion not only for a low band signal but also for a high band signal.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a motion detecting circuit according to one embodiment of the present invention, FIG. 2 is a signal waveform diagram for explaining each operation of the example of FIG. 1, and FIG. 3 is a line for explaining a MUSE transmission system. Figure, 4th
FIGS. 5 and 6 show the configuration of a conventional MUSE decoder.
Each of the figures is a signal waveform diagram for explaining the operation of the example shown in FIG. (28) is a two-frame memory, (29) is a subtraction circuit, and (30)
Is an absolute value circuit, (31) is a sensitivity setting device, (33) is a digital comparator, (35) is a two-frame memory, and (36) is an inverter circuit.

Claims (1)

(57) [Claims] 1. A television signal motion detecting circuit which forms an image of one screen by using image signals of n frames (n is an integer of 2 or more). A first memory for providing a delay of a frame period, a difference means for obtaining a difference between an output signal of the first memory and the input television signal, and comparing the output signal of the difference means with a predetermined level Comparison means; a second memory for delaying the comparison output of the comparison means by n frame periods; and a prohibition means for prohibiting the output of the difference means based on the output of the second memory. A motion detection circuit characterized by the following.