JPS60264188A - Detector of movement of picture - Google Patents

Abstract

PURPOSE:To attain simplicity of the titled detector and to improve the detecting accuracy by detecting the difference, slope of the brightness of two adjacent picture elements of a line and the difference of the brightness of two vertical picture elements of two adjacent lines of the same reproducing field or frame to detect the movement based on the detected signal. CONSTITUTION:A digital video signal read from a frame memory 11 and a digital video signal subject to delay for one picture element transmission period by a latch 12 are fed to a subtractor 14 respectively, where two horizontally adjacent picture elements are subtracted, then the difference and gradient of the brightness are obtained respectively. The difference of the brightness is fed to a comparator 16, where the difference is compared with the 1st threshold value from an input terminal 17. The digital video signal from the frame memory 11 and the digital video signal delayed by one line's share by a line memory 13 are fed respectively to a subtractor 15, where two vertically adjacent picture elements of the same reproducing field are subtracted and the difference of the brightness of them is obtained. The difference is fed to a comparator 20, where the difference is compared with the level of the 2nd threshold value from the input terminal 21.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to an image motion detection device, and in particular, it stores video signals transmitted by the dot interlace method ζ in a memory, reads them out, and detects the motion of an image to be reproduced and displayed. The present invention relates to a motion detection device.

BACKGROUND OF THE INVENTION Conventionally, the dot interlace method has been known as one of the methods for transmitting video signals in a narrow band. In this dot interlace method, subsampling is performed to compress the band of the video signal to 1/2 and transmit it. At this time, subsampling is performed by shifting the sampling position for each line and each field. To explain this dot interlacing method, in the first field, in the first field, the odd-numbered pixels in the odd-numbered lines and the even-numbered pixels in the even-numbered lines are sampled among the solid lines in the odd-numbered fields of the video signal. and transmit it. In the next second field, among the lines indicated by broken lines in the even field of the video signal, the odd-numbered pixels on the odd-numbered lines and the even-numbered pixels on the even-numbered lines are sampled and transmitted, and in the third and fourth fields, the sample positions are sampled. The pixels at the positions ■ and ■ are transmitted. That is, 1
The transmission frequency band is compressed to 1/2 by subsampling, in which the pixels of the original image for one frame are thinned out to 1/2 and then transferred over 4 fields.

On the other hand, there is a frame memory on the receiving side, and the above
The pixels of one frame transmitted in the fourth field are written to addresses corresponding to the original sampling positions in this frame memory. Pixels for one frame stored in this frame memory are read out and converted into video signals through a known signal processing circuit, and then reproduced and displayed on a display device. In this way, the video signal is transmitted after being band-compressed to 1/2, and the original image is reproduced without deterioration in image quality.

However, with this dot interlacing method, the image quality may deteriorate in the case of moving images. That is, since the pixels read from the frame memory are displayed in the same arrangement as in FIG. 1, in one field, the pixels of the first field and the third field are arranged alternately in the horizontal direction, In the field, the pixels of the second field and the fourth field are arranged alternately in the horizontal direction. Therefore, if the original image is, for example, a car traveling from the right to the left of the screen, the reproduced image will be as shown in Figure 9. As shown in the figure, the image (2) of the previous field and the image (2) of the current field are superimposed to form a double image, and the image quality deteriorates.

In order to prevent this image quality deterioration, it is necessary to detect the moving area of the image,
For moving images, it is necessary to improve the image quality by performing corrections such as displaying only the image of the current field.

Therefore, many methods for detecting image motion have been proposed in the past, and the simplest method is a method that uses inter-frame differences. That is, as shown in FIG. 10, the luminance difference between corresponding pixels of the previous frame and the ring frame is calculated, and if the difference is greater than a certain value, it is determined that there is movement.

Further, as a second conventional motion detection method, as shown in FIG. 11, one block consists of m×n pixels of the previous frame.
There was a method of shifting images around the corresponding blocks in the current frame, searching for the point where the difference between the images was minimal, and determining that displacement as movement. According to this motion detection method, the precision of motion detection can be made high.

Problems to be Solved by the Invention However, although the conventional image motion detection method shown in FIG. 10 has a simple processing circuit, motion detection is localized and highly accurate motion detection is difficult. There was a problem. Furthermore, although the conventional image motion detection method shown in Figure 11 can detect motion with high precision, it requires a large number of calculations, and requires a complex calculation circuit to detect motion in real time. The problem was that it was necessary.

SUMMARY OF THE INVENTION An object of the present invention is to provide an image motion detection device that solves the above-mentioned problems by utilizing the characteristics of the dot interlace method.

Means and Effects for Solving the Problems The present invention provides a means for solving the problems by using the same line of digital video signals taken out from a frame memory that stores digital video signals that are time-series composite signals of pixels subsampled by the dot interlacing method. a first detecting means for detecting a difference in brightness and a slope of brightness between two adjacent pixels of the frame memory; a second detection means for detecting a difference in luminance between two pixels in the vertical direction of two lines; a shift register for storing a signal regarding the slope of the luminance extracted from the first detection means; The first and second signals regarding the difference in luminance extracted from the people of the first and second detection means are separately compared with preset first and second threshold values, and the first and second signals are The signal levels of both the first and second signals are
control means for causing the shift register to perform a storage operation only when the value is greater than a threshold value, and resetting the shift register at other times; By constructing a decoder that generates and outputs a motion detection signal, it is possible to perform real-time image motion detection with a simple circuit configuration. This will be explained in conjunction with FIG. 7.

Embodiment FIG. 1 is a circuit diagram of a first embodiment of the apparatus of the present invention, and FIGS. 2 to 4 are diagrams for explaining the operating principle of the apparatus of the present invention, respectively. First, to explain the operating principle of the device of the present invention, in the dot interlace method, as shown in FIG. are arranged in

In addition, in FIG. 2, n', n+1. Each line of n+2 indicates a line of an odd field, and each line of n'+'nl+i indicates a do-in for each even field.

■Respectively, the above-mentioned 1. The pixels of the third field are shown.

Now, the image of the field before the previous field is shown as ■ in Fig. 3, its video signal (luminance signal) is shown in Fig. 4 (A>), and the image of the current field is shown as ■ in Fig. 3, and its video signal Assuming that (luminance signal) is as shown in Fig. 4 (B), in the case of the above moving image moving from ■→■, the luminance difference between pixels becomes large in the moving part. As mentioned above, since these pixels are read out consecutively in an alternating arrangement of pixels from the previous field and the current field, the output of the frame memory contains fluctuations in luminance as shown in FIG. 4(C). A signal (digital data) indicating the brightness is extracted.Therefore, by detecting this luminance fluctuation, it is possible to detect the motion of the image.

The present invention has been developed by paying attention to the above-mentioned principle, and each embodiment of the device of the present invention will be described next. 1st
In the first embodiment of the device of the present invention shown in the figure, input terminal 1
0 receives a digital video signal in which pixels obtained by shifting the sampling position for each line and each field using the dot interlacing method as explained in conjunction with FIG. It is written as shown in FIG. As described above, the input digital video signal is a pixel data series that transmits the information of the original picture for one frame over four fields.

A digital video signal (pixel data series) read out from the frame memory 11 in such a manner that the pixels of the previous field and the pixels of the current field are arranged alternately in the horizontal direction, as in the conventional case, is transmitted to the latch 12, Line memory 1 (described later)
3. are supplied to subtracters 1.4 and 15, respectively. Subtractor 1
4 is supplied with the digital video signal read out from the frame memory 11 and the digital video signal delayed by one pixel transmission period by the latch 12, and outputs both signals, that is, the signals of two horizontally adjacent pixels. Perform the subtraction,
The difference in brightness and the slope of brightness between these two adjacent pixels are obtained. The above luminance difference (absolute value) is supplied to a comparator 16, where it is compared with a first threshold value from an input terminal 17, and only when the difference is smaller than the first threshold value, 1lJII
Output the I signal. This control signal is applied as a reset pulse to the shift register 19 through the AND circuit 18 to reset it.

On the other hand, the above luminance slope obtained by the subtracter 14 is
If the difference in brightness between two horizontally adjacent pixels is positive, it is a logic II I 11 signal, and if the difference in brightness between the two pixels is negative, it is a logic 1+OI+ signal, and this signal is sent to the shift register. 19. Therefore, the shift register 19 stores the brightness slope only when the difference is greater than the first darkness value.

Here, the video signal of the field before Imisaki is shown in Fig. 5 (A>
, and the video signal of the current field is shown in the same figure (B
), the output digital video signal of the frame memory 11 shows information content in which the luminance oscillates as shown in FIG. 2(C). Therefore, the slope of the luminance output from the subtracter 14 becomes as shown in FIG. 5(D), and the output signal of the comparator 16 becomes as shown in FIG. 5('F). In FIG. 5(E), the low level period (A area and C area) is the control signal output period, which is the period during which the shift register 19 is reset, and the high level period (B area iii) is the period during which the shift register 19 is reset. This is a period during which the slope of luminance is stored in the register 19. Therefore, the output signal of the shift register 19 becomes as shown in FIG. 5(F).

In one case where there is movement in the image, the brightness gradient i shown in FIG. 19 and further supplied to the decoder 22, the decoder 22 detects a pattern of 1° o, i" or 1 "'o, i, o" from this and determines that it is a moving area. As a result, a high-level motion detection signal is output from the decoder 22 to the output terminal 23 as shown in FIG. 5(G). Using this motion detection signal, it is possible to perform corrections such as displaying only the image of the current field, thereby improving the image quality.

Note that in area A, the difference in brightness is smaller than the first threshold, so the shift register 19 is reset and no movement is detected. Further, in region C, the difference in brightness is larger than the first threshold value, but since the change in slope continues in one direction, the decoder 22 does not output a motion detection signal.

By the way, even if it is a still image, if we consider the case where the brightness changes in a short period, in this case - the brightness fluctuations for each pixel are large, and the above circuit section alone will falsely detect it as movement. is possible.

Therefore, in order to prevent such erroneous detection, in this embodiment, the line memory 13. Subtractor 15. A comparator 20 and the like are provided. In the dot interlacing method, as can be seen from Figure 8, the two pixels in the vertical direction of two adjacent lines in the same playback field are the pixel from the field before the previous field and the pixel from the current field. In contrast, for still images, the correlation between pixels is extremely strong and the brightness difference is small.

Therefore, in the present invention, motion is detected based on this difference, and this will be explained next.

The subtracter 15 is supplied with the digital video signal from the frame memory 11 and the digital video signal delayed by one line by the line memory 13, and subtracts both signals, that is, two vertically adjacent pixels of the same reproduction field. Perform subtraction to obtain the difference in luminance. This difference (
absolute value) is fed to a comparator 20, where the input terminal 21
The level is compared with a second threshold value. The comparator 20 outputs a low-level control signal only when the difference is smaller than the second threshold, and outputs a low-level control signal to the shift register 1 through the AND circuit 18.
9 as a reset pulse to reset it.

Now, suppose that a digital video signal related to a still image is input, such that the video signal of the previous field is shown in FIG. 6(A), and the video signal of the current field is shown in FIG. 6(C). , the digital video signal information of the previous line from the line memory 13 is as shown in FIG. show.

Therefore, the slope of the luminance taken out from the subtractor 14 becomes as shown in FIG. 6(E), the output signal of the comparator 16 becomes as shown in FIG. As shown in FIG. 3(G), the control signal is always at a low level.

Therefore, the output signal of the comparator 16 changes at time t1 as shown in FIG.
) Even if the level becomes high as shown in
Since the output signal of the AND circuit 18 is always at a low level as shown in FIG. Therefore, the output signal of the shift register 19 is
), it is always 1101+, so the decoder 22
The output signal of the comparator 16 is always at a low level, and even if the output signal of the comparator 16 becomes a high level at time 1+, no motion detection signal is erroneously generated.

Next, a second embodiment of the device of the present invention, which provides the same function without providing a line memory, will be described with reference to FIG. In the figure, the same components as in FIG. 1 are denoted by the same reference numerals, and their explanations will be omitted. The digital video signal inputted to the input terminal 10 is supplied to field memories 25 and 26, respectively, in which odd field pixels are stored in one, and even field pixels are stored in the other. field memory 25
and 26 correspond to the frame memory 11, and the storage area of the frame memory is divided into two to form a field memory 25.
．． 26 may be supplied.

The odd and even field digital video signals taken out simultaneously from the field memories 25 and 26 are branched into two and supplied to selectors 27 and 28, respectively. Therefore, pixels of two adjacent lines (an odd field line and an even field line) of the same reproduction frame are simultaneously supplied to each of the selectors 27 and 28.

Each of the two signals supplied to each of the selectors 27 and 28 is
These are image signals having a time difference of one field (1/60 seconds), and the selector 27 selects and outputs the digital video signal of the current playback field, and the latch 12. Subtractor 1
4 and 15, respectively, while the selector 28 selects and outputs the digital video signal of the next reproduction field and supplies it to the subtracter 15. As a result, the subtracter 14 receives a signal delayed by one pixel transmission period by the latch, and
Subtraction is performed with the signal from the selector 27 to generate a signal related to the difference in brightness between two horizontally adjacent pixels and a signal related to the slope of brightness, respectively, as in the first embodiment. On the other hand, the subtracter 15 is supplied with the digital video signals from the selectors 27 and 28, and the subtracter 15 is supplied with the digital video signals from the selectors 27 and 28, and selects two pixels on the vertical direction of two adjacent lines of the same reproduction frame (the same reproduction field in the first embodiment). A signal indicating the difference in brightness between pixels ■ and ■ (or ■ and ■) in FIG. 8 is generated.

The difference in luminance between pixels ■ and ■ (or ■ and ■), which have a time difference of one field from each other, is extremely small in the case of a still image, but large in the case of a moving image. Therefore, in this first example as well, by performing the same signal processing as in the first example on the output signal of the subtracter 15, it is possible to prevent a still image from being erroneously detected as a moving image. I can do it.

Effects of the Invention As described above, according to the present invention, from a digital video signal read out from a field memory that stores a digital video signal that is a time-series composite signal of pixels subsampled by the dot inter-race method, detecting a difference in brightness and a slope between two adjacent pixels on the same line, and detecting a difference in brightness between two vertical pixels on two adjacent lines in the same playback field or the same playback frame; Since motion detection is performed based on these detection signals, it is possible to perform image motion detection in real time without the need for complex arithmetic circuits, and motion detection can be performed locally. Since it is not a motion sensor, it has the advantage of being able to perform highly accurate motion detection.

[Brief explanation of drawings]

FIG. 1 is a circuit diagram showing a first embodiment of the device of the present invention;
FIGS. 2, 3, and 4 are diagrams for explaining the operating principle of the device of the present invention, and FIGS. 5 and 6 are signal waveform diagrams for explaining the operation of the block system shown in FIG. Fig. 7 is a circuit diagram showing a second embodiment of the device of the present invention, Fig. 8 is a diagram explaining transmission pixels of the dot interlace method, and Fig. 9 is a double image that may be reproduced and displayed using the dot interlace method. FIGS. 10 and 11 are diagrams illustrating each side of the conventional image motion detection method, respectively. 10...Digital video signal input 1 child, 11...Latch, 13...Line memory, 14.15...Subtractor, 16.20...Comparator,
17.21... Threshold input terminal, 19... Shift register, 22... Decoder, 23... Motion detection signal output terminal, 25.26... Field memory, 27.2
8...Selector.

Claims (1)

[Claims] (1) A frame memory that is supplied with a digital video signal that is a time-series composite signal of pixels subsampled by the dot interlacing method and stores it, and an output digital video signal of the frame memory is supplied,
A first detection means for detecting the difference in brightness and the slope of brightness between two adjacent pixels on the same line, and a digital video signal stored in the frame memory are supplied, and the same reproduction field or the same reproduction frame is supplied. a second detection means for detecting a difference in luminance between two pixels in the vertical direction of two adjacent lines; and a shift register for storing a signal regarding the slope of the luminance extracted from the first detection means. and separately compares the first and second signals regarding the luminance difference extracted by the first and second detection means with preset first and second thresholds, and control means for causing the shift register to perform a storage operation only when the levels of the first and second signals are both higher than the first and second thresholds, and resetting the shift register at other times; , a decoder which is supplied with the output signal of the shift register and generates and outputs an image motion detection signal based on a change pattern of its value. (2) The first detection means is supplied with a first delay circuit that delays the output digital video signal of the frame memory by one pixel transmission period, and the digital video signal output from the frame memory and the first delay circuit. The second detecting means includes a second delay circuit for delaying the output digital video signal of the frame memory by one line transmission period; 2. The image motion detection apparatus according to claim 1, further comprising a second subtracter to which the digital video signal of the delay circuit is supplied. (2) The written digital video signals of odd fields, yield fields, and even fields are simultaneously read out from the frame memory, and the first frame memory selects and outputs the digital video signal of the current field from among the read digital video signals. a second selector for selectively outputting the digital video signal of the other field among the read digital video signals;
selection means, and the first detection means has the first selection means.
a delay circuit for delaying the output digital video signal of the selection means by one pixel transmission period, and a first subtracter to which the digital video signal output from the first selection means and the delay circuit is supplied; Image motion detection according to claim 1, characterized in that the second detection means comprises a second subtractor to which the digital video signals of the first and second selection means are supplied. Device.