JPH06319129A - Time space picture filter - Google Patents

Abstract

PURPOSE:To obtain the time space picture filter in which unnatural disturbance such a double image is not caused by controlling a characteristic of a spatial filter with a compression ratio of a picture to execute spatial frequency limit or emphasis for a picture signal. CONSTITUTION:A frame memory 10 delays an input picture signal by one frame period and gives a signal of an existing frame and a signal of one preceding frame to a motion vector detection circuit 20. The motion vector detection circuit 20 adopts the so-called block matching method to obtain a local moving quantity of a picture. A coefficient generating circuit 30 controls coefficients k0-k3 based on a compression ratio of the coding of a picture signal and a horizontal component vx of the moving quantity of the motion vector detection circuit 20. The coefficients k0-d3 are given to variable coefficient circuits 40-43 forming an adaptive horizontal filter 200 and the adaptive horizontal filter 200 controls a high frequency component of the picture signal so that it is stepwise attenuated as the coding compression ratio gets lower and the motion quantity gets larger.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a spatio-temporal image filter for digitized image data, and more specifically, in the pre-stage of image encoding, depending on the image compression rate or the image motion magnitude, The present invention relates to a three-dimensional band limiting filter that removes a frequency region that is not visually important in advance.

[0002]

2. Description of the Related Art FIG. 9 shows, for example, Nogaki et al., "Study on HDTV signal coding using motion and static adaptive denoising" (Image coding symposium PCSJ88, 8-6, pp125 to 12).
6 is a block diagram showing a conventional spatiotemporal image filter shown in FIG. In the figure, 120 is a horizontal bandpass filter (hereinafter referred to as BPF), 11 and 12 are frame memories cascade-connected to the output of the horizontal BPF 120, 63 is an adder circuit for adding the outputs of the horizontal BPF 120 and the frame memory 12, 50 , 51 are coefficient circuits connected to the adder circuit 62 and the frame memory 11, respectively, 70 is a subtraction circuit for subtracting the outputs of the coefficient circuits 51 to 50, and 71 is a subtraction circuit for subtracting the output of the subtraction circuit 70 from the input signal. .

Next, the operation will be described. The filter in FIG. 9 removes noisy frequency components from the HDTV luminance signal. Horizontal BPF120 is about 18M
A high frequency component centering on Hz is extracted. This frequency component is affected by the aperture correction of the camera,
Especially, it is an area where many noise components exist. The portion from the frame memory 11 to the subtraction circuit 70 constitutes a time axis high pass filter (hereinafter referred to as HPF) 300. The time axis HPF 300 extracts, as noise, only the signal component having a large temporal change from the signal component near 18 MHz. The output of the time base HPF 300 is removed from the original signal by the subtraction circuit 71.

In the time base HPF 300, the frame memories 11 and 12 delay the input signal for one frame period and output the delayed signal. Frame memory 11 at time t
If the signal input to X is set to X (t), the outputs of the frame memories 11 and 12 are X (t-F) and X (t-, respectively).
2F) (F is a frame period). The adder circuit 63 is X
(T) and X (t-2F) are added to obtain X (t) + X (t-2F). The coefficient circuit 50 multiplies this by 1/4 to obtain 1/4 · {X (t) + X (t-2F)}. On the other hand, the coefficient circuit 51 multiplies X (t-F) by 1/2 to obtain 1/2 · X (t-F). The subtraction circuit 70 includes the coefficient circuit 51 to the coefficient circuit 50.
Is subtracted, and finally, as the output of the time axis HPF 300, 1/2 · X (t−F) −1 / 4 · {X (t) + X (t−
2F)} is obtained. This is a HPF in the time direction in which the gain is 1 for a signal that changes with time of 30 Hz and 0 for a stationary signal that does not change with time.

Incidentally, in the above-mentioned literature, the time base HPF
Regarding the above, there is only a description that a FIR (Finite Impulse Response) filter is used. Therefore, the simplest and most general time-axis FIR filter has been described here.

[0006]

Since the conventional spatiotemporal image filter is constructed as described above, there is a problem that the edge portion of a moving object becomes a double image due to signal processing between frames. there were.

The present invention has been made to solve the above problems, and provides a spatio-temporal image filter that does not cause unnatural interference such as double images, and also provides a target rate. The objective is to obtain the code amount reduction effect according to the above and the subjective image quality of the restored image expected at the target rate.

[0008]

A spatio-temporal image filter according to a first aspect of the present invention is a spatial filter having a variable characteristic that restricts or enhances a spatial frequency band of an image signal, and the spatial filter has a spatial compression ratio depending on a compression rate of an image. By controlling the characteristics, spatial frequency limitation or enhancement of the image signal is performed.

The spatiotemporal image filter according to the second aspect of the present invention includes means for locally obtaining a motion vector of an image by utilizing the correlation in the time direction and variable characteristic for limiting or enhancing the spatial frequency characteristic of the image signal. A spatial filter is provided, and the characteristics of the spatial filter are controlled by the compression rate of the image and the size of the motion vector, thereby performing the spatial or temporal frequency limitation or enhancement of the image signal. It is a thing.

A spatiotemporal image filter according to a third aspect of the present invention uses means for locally obtaining horizontal and vertical components of a motion vector from an image signal by utilizing correlation in the time direction, and a frequency band in the horizontal direction of the image signal. It has a horizontal filter with variable characteristics that limits or emphasizes, and a vertical filter with variable characteristics that limits or emphasizes the vertical frequency band of the image signal.The characteristics of the horizontal filter are controlled by the image compression rate and the horizontal component of the motion vector. The characteristics of the horizontal filter are controlled by the image compression rate and the vertical component of the motion vector.

The spatiotemporal image filter according to the fourth aspect of the present invention has means for locally obtaining a motion vector of an image by utilizing the correlation in the time direction, and a characteristic for limiting or enhancing the frequency band in the horizontal and vertical directions of the image signal. A variable two-dimensional filter is provided, and the characteristics of the two-dimensional filter are controlled by the compression rate of the image and the magnitude of the motion vector, so that the spatial and temporal frequency limitation or enhancement of the image signal is performed.

[0012]

The spatiotemporal image filter according to the present invention equivalently performs filtering in the time direction by changing its frequency characteristic in accordance with the image compression rate and the local motion amount of the image.

[0013]

【Example】

Example 1. FIG. 1 is a block circuit diagram showing a spatiotemporal image filter according to an embodiment of the first invention. In the figure, 3
0 is a coefficient generation circuit which inputs the compression rate of the image, 100 to
Reference numeral 102 is a horizontal low-pass filter (hereinafter referred to as LPF) connected in parallel to the input, 40 to 43 are connected to the input and the horizontal LPFs 100 to 102, respectively, and the coefficient value is controlled by the coefficient generating circuit 30 to be variable. A coefficient circuit, 60 is an adder circuit for adding the outputs of the variable coefficient circuits 40 to 43.

Encoding compression rate (hereinafter referred to as rate)
Is, for example, an HDT sampled at 74.25 MHz
If V is quantized with 8 bits, a code amount of about 960 Mbps is obtained. For example, the target coding rate is 40 Mbps.
If set, the rate will be 1/24. In a coding device having a variable coding speed, this rate has a variable value.

LPFs 100 to 102, coefficient generation circuit 3
0, the variable coefficient circuits 40 to 43, and the adder circuit 60 constitute an adaptive horizontal filter 200 having variable characteristics. The input image signal is limited to different frequency bands by the LPFs 100 to 102. The passing area of this LPF is, for example, fs
Is set as the sampling frequency of the signal, such as LPF100: fs / 4 LPF101: fs / 8 LPF102: fs / 12. FIG. 8 shows these pass bands.
For HDTV signals, typically fs = 74.25 MHz.

Variable coefficient circuits 40 to 43 and adder circuit 60
Adds the input signal and the outputs of the LPFs 100 to 102 at a mixing ratio determined by the compression ratio of the image and outputs the result. X0 is the input signal and X is the output of the LPFs 100-102.
1 to X3 and the coefficient values of the variable coefficient circuits 40 to 43 are k0 to k3, respectively, the output Y of the adder circuit 60 is given by the following equation. Y = k0 * X0 + k1 * X1 + k2 * X2 + k3 * X3 Since X0 to X3 are signals having different frequency bands from each other, the frequency band of the output Y can be changed by controlling the values of k0 to k3. The coefficient generation circuit 30 uses the compression ratio (rat
According to e), the coefficient values k0 to k3 are controlled as follows, for example.

When 1/15 <rate, k0 = 1 k1 = 0 k2 = 0 k3 = 0

When 1 / 25≤rate≤1 / 15, k0 = 3/4 k1 = 1/4 k2 = 0 k3 = 0

When 1 / 40≤rate≤1 / 25, k0 = 1/2 k1 = 1/2 k2 = 0 k3 = 0

When 1 / 60≤rate≤1 / 40, k0 = 1/2 k1 = 0 k2 = 1/2 k3 = 0

When rate ≦ 1/60, k0 = 1/2 k1 = 0 k2 = 0 k3 = 1/2

That is, when 1/15 <rate, the output follows the following equation. Y = X0 Here, X0 is an input signal. Below, 1/25 ≦ ra
When te ≦ 1/15, Y = 3/4 · X0 + 1/4 · X1 where X0 is the input signal and X1 is the signal band-limited to fs / 4 or less by the LPF 100. It becomes a signal in which high frequency components of / 4 or more are attenuated. Similarly, when 1/40 ≦ rate ≦ 1/25, Y = 1/2 · X0 + 1/2 · X1 1/60 ≦ rate ≦ 1/40, Y = 1/2 · X0 + 1/2 · When X2 rate ≤ 1/60, Y = 1 / 2.X0 + 1 / 2.X3. Therefore, in the adaptive horizontal filter 200, the high frequency component of the image signal has a low encoding target rate, that is, compression. As the rate increases, it decays gradually.

Example 2. FIG. 2 is a block circuit diagram showing a spatiotemporal image filter according to an embodiment of the second invention. In the figure, 10 is a frame memory for inputting an image signal,
Reference numeral 20 is a motion vector detection circuit connected to the input and the frame memory 10. Reference numeral 30 is a coefficient generation circuit having the compression rate of encoding and the output of the motion vector detection circuit 20 as input.
Reference numerals 100 to 102 denote horizontal low-pass filters (hereinafter, referred to as LPFs) connected in parallel with each other, 40 to 40
43 are input and horizontal LPFs 100 to 102, respectively.
, A variable coefficient circuit whose coefficient value is controlled by the coefficient generating circuit 30, and 60 are variable coefficient circuits 40-4.
Adder circuits for adding the outputs of 3 and 40 to 43,
A variable coefficient circuit which is connected to the input and horizontal LPFs 100 to 102 and whose coefficient value is controlled by the coefficient generation circuit 30.
An adder circuit 60 adds the outputs of the variable coefficient circuits 40 to 43.

Next, the operation will be described. The temporal frequency of the image signal is proportional to the spatial frequency of the object and the speed of movement, where the temporal frequency is f (cycle / second), the speed of movement is v (pixel / frame), and the spatial frequency is μ (cycle / pixel). , The following relationship holds. f = k · μ · v where k (frames / second) is the frame frequency.
In the case of an HDTV signal, k = 29.97, and the relationship between μ and f with v as a parameter is a group of straight lines passing through the origin, as shown in FIG.

In FIG. 2, the frame memory 10 delays the input image signal for one frame period and supplies the motion vector detection circuit 20 with the signal of the current frame and the signal of the previous frame. The motion vector detection circuit 20 obtains a local image movement amount by a so-called block matching method.
That is, first, the current frame is divided into blocks each having a size of, for example, 16 pixels × 16 lines, then an image portion that best matches each block is obtained from the previous frame, and the displacement amount (vx, vy) Is the movement amount of the block. Here, vx is a horizontal component of the movement amount, and vy is a vertical component. Of these, vx is given to the coefficient generation circuit 30.

LPFs 100 to 102, coefficient generation circuit 3
0, the variable coefficient circuits 40 to 43, and the adder circuit 60 constitute an adaptive horizontal filter 200 having variable characteristics. The input image signal is limited to different frequency bands by the LPFs 100 to 102. The passing area of this LPF is, for example, fs
Is set as the sampling frequency of the signal, such as LPF100: fs / 4 LPF101: fs / 8 LPF102: fs / 12. FIG. 8 shows these pass bands.
For HDTV signals, typically fs = 74.25 MHz.

Variable coefficient circuits 40 to 43 and adder circuit 60
Is the input signal and the outputs of the LPFs 100 to 102, which are determined by the compression rate rate of encoding and the movement amount vx (kn = 0,
Add and output according to the mixing ratio of 1, 2, 3). Input signal is X
0, the outputs of LPF 100-102 are X1-X3, respectively.
Then, the output Y of the adder circuit 60 is given by the following equation. Y = k0 * X0 + k1 * X1 + k2 * X2 + k3 * X3 Since X0 to X3 are signals having different frequency bands from each other, the frequency band of the output Y can be changed by controlling the values of k0 to k3. The coefficient generation circuit 30 has a compression ratio rat.
For example, the coefficient values k0 to k3 are controlled as follows according to the value of e and the horizontal component vx of the movement amount.

When rate> 1/15: When 0 ≦ vx (pixel / frame, [hereinafter abbreviated as p / f]), k0 = 1 k1 = 0 k2 = 0 k3 = 0

When 1/25 <rate <1/15 When 0 ≦ vx ≦ 12 (p / f), k0 = 1 k1 = 0 k2 = 0 k3 = 0 12 ≦ vx ≦ 24 (p / f) When k0 = (24-vx) / 12 k1 = (vx-12) / 12 k2 = 0 k3 = 0 24 ≦ vx (p / f), k0 = 0 k1 = 1 k2 = 0 k3 = 0

When 1 / 40≤rate≤1 / 25 When 0≤vx≤8 (p / f): k0 = (8-vx) / 8 k1 = vx / 8 k2 = 0 k3 = 0 8≤vx When ≦ 16 (p / f), k0 = 0 k1 = (16-vx) / 8 k2 = (vx-8) / 8 k3 = 0 k4 = 0 16 ≦ vx ≦ 24 (p / f), k0 = 0 k1 = 0 k2 = (24-vx) / 8 k3 = (vx-16) / 8 When 24 ≦ vx (p / f), k0 = 0 k1 = 0 k2 = 0 k3 = 1

When 1 / 80≤rate≤1 / 40 When 0≤vx≤4 (p / f) k0 = (4-vx) / 4 k1 = vx / 4 k2 = 0 k3 = 0 4≤vx When ≦ 12 (p / f), k0 = 0 k1 = (12−vx) / 8 k2 = (vx−4) / 8 k3 = 0 When 12 ≦ vx ≦ 24 (p / f), k0 = 0 k1 = 0 k2 = (24-vx) / 12 k3 = (vx-12) / 12 When 24 ≦ vx (p / f), k0 = 0 k1 = 0 k2 = 0 k3 = 1

When 1 / 80≤rate When 0≤vx≤4 (p / f): k0 = (4-vx) / 4 k1 = vx / 4 k2 = 0 k3 = 0 4≤vx≤8 (p / F), k0 = 0, k1 = (8-vx) / 4, k2 = (vx-4) / 4, k3 = 0, when 12 ≦ vx ≦ 24 (p / f), k0 = 0, k1 = 0, k2 = 1 k3 = 0 24 ≦ vx (p / f), k0 = 0 k1 = 0 k2 = 0 k3 = 1

In this way, the movement amount vx is determined by rate.
Since the range is variable, the horizontal high frequency component gradually attenuates as the rate decreases. On the other hand, for example, 1
The case of / 80 ≦ rate ≦ 1/40 will be described. When 0 ≦ vx ≦ 4, the output is X0 according to the following equation.
To X1 continuously. Y = (1-vx / 4) * X0 + (vx / 4) * X1 where X0 is the input signal and X1 is the signal whose band is limited to fs / 4 or less by the LPF 100. The above high frequency components are gradually attenuated as vx increases. Similarly, 4 ≦
When vx ≦ 12, Y = {1- (vx-4) / 8} X1 + {(vx-4) /
8} · X2 12 ≦ vx ≦ 24 (p / f), Y = {1- (vx-12) / 12} · X2 + {(vx-1
2) / 12} · X3 When 24 ≦ vx (p / f), Y = k3 · X3. Therefore, in the adaptive horizontal filter 200, as the rate of the high frequency components of the image signal becomes lower, As the amount of movement increases, it gradually decreases.

In FIG. 7, vx = 4, 12, 24 (p
/ F), the horizontal frequency is fs / 4, f
Points indicating s / 8 and fs / 12 are indicated by. Frequency components higher than this are attenuated by the above description. Therefore, the present embodiment is a filter that equivalently removes signal components represented by diagonal lines in the time-horizontal frequency domain.

Example 3. In the above embodiment, the high frequency component of the signal is limited by the horizontal component of the movement amount, but the vertical compression component of the signal and the vertical component of the movement amount are further used to limit the vertical high frequency component of the signal. You may do it.

FIG. 3 is a block circuit diagram showing a spatiotemporal image filter according to an embodiment of the third invention. In the figure,
Reference numeral 10 is a frame memory for inputting an image signal, and 20 is a motion vector detection circuit connected to the input and frame memory 10. The configuration of the adaptive horizontal filter 200 is the same as that of the first embodiment. Reference numeral 31 denotes a coefficient generation circuit which receives the compression rate of encoding and the output of the motion vector detection circuit 20, and 11
0 and 111 are vertical LPFs connected in parallel to the output of the adaptive horizontal filter 200, and 44 to 46 are connected to the adaptive horizontal filter 200 and the vertical LPFs 110 and 111, respectively, and a variable coefficient circuit controlled by the coefficient generating circuit 31. , 61 are adder circuits for adding the outputs of the variable coefficient circuits 44 to 46.

Next, the operation will be described. The operations of the frame memory 10 and the motion vector detection circuit 20 are the same as in the first embodiment. The horizontal component vx of the movement amount obtained by the motion vector detection circuit 20 is supplied to the coefficient generation circuit 30 and the vertical component vx.
y is provided to the coefficient generation circuit 31. The band limitation of the horizontal frequency is performed by the adaptive horizontal filter 200 as in the first embodiment.

The coefficient generating circuit 31, the vertical filter 110,
111, the variable coefficient circuits 44 to 46, and the adder circuit 61 constitute an adaptive vertical filter 201 with variable characteristics. here,
The pass bands of the vertical LPFs 110 and 111 are set as follows. LPF110: fs / 4 LPF111: fs / 8 where fs is the sampling frequency in the vertical direction.
In the case of HDTV, considering interlace, fs = 11
25/2 (c / ph: cycle / picture h
Eight). Figure 7 shows this passband as a horizontal LP
Shown together with F. The outputs of the vertical LPFs 110 and 111 are mixed and added together with their input signals through the variable coefficient circuits 44 to 46 and the adder circuit 61.

The coefficient generating circuit 31 determines the coefficient values k4 to k6 of the variable coefficient circuits 44 to 46 as follows based on the encoding compression amount and the vertical component vy of the moving amount. Here, the movement amount counts the scanning line interval in the field as one line.

When rate> 1/15, k4 = 1 k5 = 0 k6 = 0

When 1/25 rate ≦ 1/15 When 0 ≦ vy ≦ 8 (line / frame, [hereinafter abbreviated as l / f]), k4 = 1 k5 = 0 k6 = 0 8 ≦ vy ≦ 12 When (l / f), k4 = (12-vy) / 4 k5 = (vy-8) / 4 k6 = 0 When 12 ≦ vy (l / f), k4 = 0 k5 = 1 k6 = 0

When 1 / 40≤rate≤1 / 25: When 0≤vy≤8 (l / f), k4 = (8-vy) / 8 k5 = vy / 8 k6 = 0 8≤vy≤12 ( 1 / f), k4 = 0 k5 = (12−vy) / 4 k6 = (vy−4) / 4 12 ≦ vy (1 / f), k4 = 0 k5 = 0 k6 = 1

When 1 / 80≤rate≤1 / 40: When 0≤vy≤4 (l / f): k4 = (4-vy) / 4 k5 = vy / 4 k6 = 0 4≤vy≤12 ( 1 / f), k4 = 0 k5 = (12−vy) / 8 k6 = (vy−4) / 8 12 ≦ vy (1 / f), k4 = 0 k5 = 0 k6 = 1

In the case of rate ≦ 1/80 When 0 ≦ vy ≦ 4 (l / f), k4 = (4-vy) / 4 k5 = vy / 4 k6 = 0 4 ≦ vy ≦ 12 (l / f) When, k4 = 0 k5 = 1 k6 = 0 12 ≦ vy (l / f), k4 = 0 k5 = 0 k6 = 1

In this way, as the rate becomes lower and the amount of movement increases, the high vertical frequency component is gradually attenuated. Therefore, the adaptive vertical filter 201 can equivalently perform band limitation in the vertical-time frequency domain.

Example 4. In the spatial filter according to the third embodiment, a horizontal filter and a vertical filter are cascade-connected, and the respective characteristics are controlled by the horizontal and vertical components of the motion vector or the compression ratio of the encoding. It consists of a two-dimensional filter that is not
The characteristics may be controlled by parameters of the magnitude of the motion vector and the compression rate of encoding.

FIG. 4 is a block circuit diagram showing a spatiotemporal image filter according to an embodiment of the fourth invention. In the figure,
Reference numeral 10 is a frame memory for inputting an image signal, 20 is a motion vector detection circuit connected to the input and frame memory 10, and 80 is a read-only memory (hereinafter referred to as ROM) having the output of the motion vector detection circuit 20 as input.
2 is a coefficient generation circuit that receives the compression rate of the encoding and the output of the ROM 80, and 130 is a two-dimensional LP connected to the input.
F, 47 and 48 are the input and the two-dimensional LPF 13, respectively.
A variable coefficient circuit connected to 0 and controlled by the coefficient generation circuit 32, and 62 is an adder circuit for adding the outputs of the variable coefficient circuits 47 and 48.

FIG. 5 is a block circuit diagram showing a two-dimensional LPF. In the figure, 90 is a line memory connected to the input, 140 and 141 are delay circuits cascade-connected to the line memory 90, 91 is a line memory connected to the delay circuit 141, and 52 to 56 are input and line memories, respectively. 90, delay circuits 140 and 141, line memory 9
1 are connected to the coefficient circuits, and 64 are these coefficient circuits 52 to
It is an adder circuit that adds the outputs of 56.

Next, the operation will be described. The operations of the frame memory 10 and the motion vector detection circuit 20 are the same as in the first embodiment. Horizontal component vx and vertical component v of the motion vector
The y is converted by the ROM 80 into the vector size V defined by the following equation. V = [vx ² + vy ² ] ^1/2

The portion from the two-dimensional LPF 130 to the adder circuit 62 constitutes an adaptive two-dimensional filter 202 having variable characteristics. The two-dimensional LPF 130 is a filter that attenuates spatial high frequency components of the image signal. The detailed operation will be described later. Input image signal to X0, two-dimensional LPF130
Is X1 and the coefficient values of the variable coefficient circuits 47 and 48 are k6 and k7, respectively, the output Y of the adder circuit 60 is given by the following equation. Y = k6 ・ X0 + k7 ・ X1

The coefficient generating circuit 32 sets the coefficients k6 and k7 according to the compression rate and the value of V, for example, as follows. In case of rate> 1/15 k7 = 1 k8 = 0

When 1 / 40≤rate≤1 / 15: When 0≤V≤6 (p / f), k7 = (6-V) / 4 k8 = V / 6 6≤V (p / f) When k7 = 0 k8 = 1

When 1 / 80≤rate≤1 / 40 When 0≤V≤12 (p / f), k7 = (12-V) / 12 k8 = V / 12 12≤V (p / f) When k7 = 0 k8 = 1

In the case of rate ≦ 1/80 When 0 ≦ V ≦ 24 (p / f) k7 = (24−V) / 24 k8 = V / 24 When 24 ≦ V (p / f) k7 = 0 k8 = 1 By this control, the spatial high frequency components of the output signal are continuously attenuated as the coding compression rate decreases and the motion amount increases.

Now, the operation of the two-dimensional LPF 130 will be described. In FIG. 5, the line memory 90, the delay circuits 140 and 141, and the line memory 91 constitute a memory circuit that delays a signal for a certain period. The delay amount of the delay circuits 140 and 141 is one sample period. The delay amounts of the line memories 90 and 91 are set so as to be one line period in combination with the delay amounts of the delay circuits 140 and 141 connected to the cascade, respectively. Therefore, the input image signal is x0, the line memory 9
0, the delay circuits 140 and 141, and the output of the line memory 91 are sequentially set to x1 to x4, as shown in FIG.
And x4 are adjacent points above and below x2, and x1 and x3 are adjacent points to the left and right of x2.

The coefficient circuit 54 halves the value of x2, and the other coefficient circuits 52, 53, 55 and 56 are respectively x.
The values of 0, x1, x3 and x4 are multiplied by 1/8. Adder circuit 6
4 adds all of these results and obtains X1 = x2 / 2 + (x0 + x1 + x3 + x4) / 8 as the output of the two-dimensional LPF 130. By this calculation, spatial high frequency components are attenuated.

In each of the above embodiments, the motion vector detection circuit 20 is adapted to obtain the motion vector from the amount of displacement of the signal between the current frame and the previous frame. The displacement may be subtracted to correct the motion vector. This allows
It is possible to prevent the high frequency of the signal from being attenuated in response to the movement of the eyes such as the pan of the camera.

In each of the above embodiments, the coefficient k0
The control of ~ k8 is performed for each block of motion detection, and the same band limitation is applied to the signals in the same block, but the motion vector for each pixel is obtained by interpolating from the motion amount of the adjacent block, The coefficient control may be performed for each pixel.

In each of the above embodiments, the LPF is used as the spatial filter, but the spatial filter is not limited to this, and
For example, using a so-called median filter,
You may take the structure which controls the characteristic.

[0060]

As described above, according to the present invention, the image code has the function of limiting the spatial frequency of the image according to the magnitude of the amount of movement of the object according to the compression rate of the image coding. If it is used in the preceding stage of the encoding, there is an effect that a double image resulting from inter-frame processing is not generated, and a time frequency that is suitable for human visual characteristics can be limited, and an image that can be easily subjected to compression encoding processing can be obtained. Since it is generated in advance, image quality deterioration due to mosquito noise, block distortion, etc. due to encoding / decoding processing is less likely to occur, it has the effect of effectively reducing the code amount, and the subjective image quality required for the target encoding speed is easy. Can be obtained.

[Brief description of drawings]

FIG. 1 is a block circuit diagram showing a spatiotemporal image filter according to an embodiment of the first invention.

FIG. 2 is a block circuit diagram showing a spatiotemporal image filter according to an embodiment of the second invention.

FIG. 3 is a block circuit diagram showing a spatiotemporal image filter according to an embodiment of the third invention.

FIG. 4 is a block circuit diagram showing a spatiotemporal image filter according to an embodiment of the fourth invention.

FIG. 5 is a block circuit diagram showing a two-dimensional LPF according to a fourth embodiment.

FIG. 6 is an explanatory diagram showing the operation of the two-dimensional LPF according to the fourth embodiment.

FIG. 7 is an explanatory diagram showing frequency characteristics of the space-time filters according to the first and second embodiments.

FIG. 8 is an explanatory diagram showing frequency characteristics of the horizontal filter and the vertical filter according to the first to third embodiments.

FIG. 9 is a block circuit diagram showing a spatiotemporal image filter according to a conventional example.

[Explanation of symbols]

 10-12 frame memory 20 motion vector detection circuit 30-32 coefficient generation circuit 40-48 variable coefficient circuit 50-56 coefficient circuit 60-64 addition circuit 70-71 subtraction circuit 80 ROM 90-91 line memory 100-102 horizontal LPF 110 -111 Vertical LPF 120 Horizontal BPF 130 Two-dimensional LPF 140-141 Delay circuit

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] August 26, 1993

[Procedure Amendment 1]

[Document name to be corrected] Drawing

[Correction target item name] Figure 9

[Correction method] Change

[Correction content]

[Figure 9]

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Takahiro Nakai 1 Baba Institute, Nagaokakyo City, Kyoto Prefecture Mitsubishi Electric Corporation Electronic Product Development Laboratory

Claims (4)

[Claims] 1. A spatial filter having a variable characteristic that limits or enhances a spatial frequency band of an image signal, wherein the spatial frequency characteristic of the image signal is limited by controlling the characteristic of the spatial filter according to an image compression rate. Alternatively, a spatiotemporal image filter characterized by performing emphasis. 2. A method for compressing an image, comprising means for locally obtaining a motion vector of an image by utilizing a correlation in the time direction, and a variable spatial filter for limiting or enhancing a spatial frequency characteristic of an image signal. A spatiotemporal image filter characterized by performing spatiotemporal frequency limitation or enhancement by controlling the characteristics of a spatial filter by the rate and the magnitude of a motion vector. . 3. A means for locally obtaining horizontal and vertical components of a motion vector from an image signal by utilizing correlation in the time direction, and a horizontal filter having a variable characteristic for limiting or enhancing the frequency band in the horizontal direction of the image signal. And a vertical filter with variable characteristics that limits or emphasizes the vertical frequency band of the image signal, controls the characteristics of the horizontal filter with the horizontal component of the image compression rate and motion vector, and A spatio-temporal image filter characterized by controlling the characteristics of a horizontal filter with a vertical component. 4. A means for locally obtaining a motion vector of an image using a correlation in the time direction, and a two-dimensional filter having a variable characteristic for limiting or enhancing the frequency band in the horizontal and vertical directions of the image signal, A spatio-temporal image filter characterized by performing spatio-temporal frequency limitation or enhancement of an image signal by controlling the characteristics of a two-dimensional filter by the image compression rate and the magnitude of a motion vector.