JPS6027287A - Motion detecting circuit - Google Patents

Abstract

PURPOSE:To realize a scanning line interpolation processing where the picture quality is not degraded even in dynamic pictures, by utilizing a frame difference signal between two adjacent fields to extract motion information of an object. CONSTITUTION:When field memories 21, 23, and 24 and line memories 22 and 25 are cascaded, positional relations of time/space between their input and output signals appear as signals on scanning lines 14-19. A signal of an interpolating scanning line 20 is calculated in accordance with a formula by adding circuits 26, 28, and 32, coefficient circuits 27 and 29, and multiplying circuits 30 and 31 (0<=k<=1). In this formula, a motion coefficient (k) is approximated to ''1'' in the part of a still picture and is approximated to ''0'' in the part of a dynamic picture. A motion detecting circuit consists of subtracting circuits 33, 35, and 37, absolute value circuits 34, 36, and 38, an adding circuit 39, and a coefficient converting circuit 40, and the motion coefficient (k) is obtained as the output of the coefficient converting circuit 40 which sets (k) to ''1'' if a signal resulting from addition of absolute values of three frame difference signals (14-17), (15-19) and (16-18) is small and approximates (k) to ''0'' according as this signal is larger.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a motion detection circuit, and more particularly, to a motion detection circuit for an image of a television signal, and particularly to a motion detection circuit for detecting motion of an image of an interlace-scanned television signal. This invention relates to a circuit that extracts motion information.

[Background of the invention]

In interlace-scanned television images, disturbances such as flicker occur in the horizontal stripes of characters and the like. To improve this, interpolate the scan lines.

There is a method in which scanning lines of two fields are displayed simultaneously, interlaced scanning is stopped, and a progressive scanning television image is created.

In this case, the signal of the interpolated scan line is interpolated from the signal of the previous field. In the case of a stationary subject, this allows you to display a high-quality image without any shuffling, but if the subject moves, the field image from 1/60 second ago and the current field image will be displayed overlapping each other. The edges of the moving parts look like the teeth of a comb, resulting in unsightly deterioration of image quality.

In order to avoid this interference with image quality, the motion information of the subject is extracted from the TV signal, and in the video part, instead of interpolating from the previous field, the signal of the interpolated scan line is used with the average value of the adjacent scan lines of the current field image. A method of making it is being tried.

FIG. 1 shows television images (u-2+) that are sent sequentially.

(' ”) + ' field).The number of scanning lines in one image is 525/2 due to interlaced scanning, and in the figure, some of the scanning lines are shown as solid lines (1, 2 , 3). From this incrementally scanned television image, the number of scan lines is doubled to create a progressively scanned television image in which each image consists of 525 scan lines. It is shown by a broken line.If the signal of the interpolated scanning line is a still image,
Interpolation is performed from the scanning line 3 at the same position in the vertical direction (V direction) of the previous field, and in the case of a moving image, interpolation is performed using the average value of the upper and lower scanning lines 1.2 of the same field. FIG. 2 shows this interpolation in terms of the time axis t and the vertical axis {circle around (2)}, where 0 marks indicate interlaced scanning lines and O marks indicate interpolated scanning lines. The symbol 1゜2.3.4 is the 4-mark shown in Figure 1.
Represents a scan line.

To determine whether or not the subject is moving, compare the scanning lines at the same position two fields (one frame) ago, as shown by the double arrows in Figure 2, and if the difference is small, it is determined that the image is still. If it is large, it is determined that it is a moving image. Accordingly, the signal of the interpolated scanning line 4 is switched to the average value of the scanning line 3 of the previous field or the scanning lines 1 and 2 of the same field.

Consider, for example, a case where an image as shown in FIG. 3, that is, a black image 5 consisting of three scanning lines, is moving upward in the above-described signal processing method.

The fourth prisoner is expressed by time t and vertical axis V, and the figure shows that it is moving upward at a speed of 93 lines per field period. In the case of Party B, areas A and C? ' can be regarded as a moving image because there is a difference from the scanning line one frame before,
As shown in the figure, the interpolated scanning lines are interpolated using the average fat of the upper and lower scanning lines of the same field. - force, in region B, the values of the l field and the (5-2) field scan line are equal, so the still image and the given awn% ('1) field scan line can be interpolated. It is done. Therefore, in region B, l
The black scan lines of the <i-x> field are interpolated onto the light background of the field. As a result, as shown in FIG. 5, the sequential image of the i-th field has area A
cD The black image 7 is blank, and below it (area B) black images 8 and 9 appear every other line, resulting in an eyesore and image quality deterioration.

[Purpose of the invention]

An object of the present invention is to solve the above-mentioned problems and provide a motion information detection circuit for obtaining a scanning line interpolation signal that does not cause image quality deterioration even in moving images.

[Summary of the invention]

To achieve the above object, the present invention provides a circuit for extracting object movement information included in an interlaced television signal, which extracts object movement information using successive two field frame difference signals. It is characterized by

[Embodiments of the invention]

The principle of one embodiment of the present invention will be described below with reference to FIG. FIG. 6 shows the arrangement of scanning lines in four successive fields, with the time axis and the vertical axis (■) representing a portion corresponding to region B in FIG. 4. The interpolation mode of interpolation scan #i14 is
Frame difference signals 11 and 12 of scanning lines 1 and 2, and frame difference signal 13 of scanning line 3 at the same position in the previous field read.
Taking into account the magnitude of these three frame difference signals 11, 1
2.

13 are all small, it is determined that it is a still image,
The signal of the interpolation scanning line 4 is interpolated from the scanning line 3 of the previous field read, and the other signals are determined to be moving images and interpolated with the average value of the upper and lower scanning lines 1.2 of the same field. In the case of interpolation scanning line 4 in the same figure, frame difference signal 11.12F:l is small, but signal 13 is thick, so it is determined to be a moving image, and interpolation is performed using the average value of scanning lines 1 and 2. As a result, image quality deterioration such as 8.9 shown in FIG. 5 does not occur.

In the embodiment of FIG. 6, the image signals of the i and (t) fields are used to obtain the signal of the interpolated scanning line of the l field. On the other hand, the signals of the (t-t+(i-2), (ta) fields) are used to detect whether the subject is still or moving in order to switch the interpolation processing mode. There may be a large time lag between the interpolation process and the correct static/dynamic adaptive process.

<i-3> ~ Using the il field static/motion determination results, interpolation processing of the (i-1)th field is performed. For the interpolation processing, the signal of the (t-2), u-1+ field is used for the interpolation processing. is used, and the signals of the (' 3) I field before and after it are used for motion determination, and the time difference between the motion determination result and the interpolation process using it is at least y correct motion adaptive processing. can be realized. That is, in FIG. 7, the static/moving determination of the signal of the interpolated scanning line 20 of the (i-1> field) is performed using the difference signal of scanning lines 14.17, 15.19, and 16.1.
Using the difference signal of 8 (if judged by 7, the maximum time interval is 2
Due to the field period, the time difference between interpolation calculation and static/motion determination is small.

FIG. 8 shows the motion coefficient k (0 k 1) shows one row of the circuit configuration calculated by changing the equation. In the same figure, if field memories 21, 23, 24 and line memories 22, 25 are continuously connected as shown in the figure, the time/space positional relationship of their input/output signals will be
19 signal. The signal (20) of the interpolation scanning line 20 is outputted as 1.L by the adder circuits 26, 28.32, the coefficient circuits 27.29, and the multiplication circuits 30.31 as shown in the following equation.

However, the motion coefficient approaches 1 in the still image part,
It is assumed that the value approaches 0 in the video portion. The motion detection circuit is
Subtraction circuit 33.35.37. Absolute value conversion circuit 34°36.
38 and addition circuit 39. It consists of a coefficient modification circuit 40 and three frame difference signals C(14)-(17)),
C(15) −(19)) , [(16) −(1
8) Coefficient modification circuit 40 which sets k to 1 if the signal obtained by adding the absolute values of) is small, and approaches k to 0 as it increases.
The motion coefficient k is obtained as the output.

The motion coefficient can have several values between l:Q and 1, but in order to simplify the circuit configuration by omitting the multiplication circuits 30 and 31, k may be limited to two values, O and 1. . in this case,
The multiplier circuits 30 and 31 can be removed and the adder circuit 32 can be replaced with a switch.

Further, by ORing the static/motion determination results of the absolute value signals of the frame difference signals (output signals of 34 and 36 degrees 38), the adder circuit 39. The coefficient modification circuit 40 can be replaced.

〔Effect of the invention〕

As described above, since the motion information of an interlaced image is extracted from the frame difference signal of two successive fields, the present invention has a problem in that the motion information of an interlaced image is extracted from the frame difference signal of two successive fields. Unsightly image quality deterioration can be avoided, and scanning line interpolation processing that does not cause image quality deterioration can be realized. As a result, it is possible to realize a high-quality television receiver that can display general tempo images with extremely high quality.

[Brief explanation of the drawing]

Figure 1 is a three-dimensional model of a television image, Figure 2 is an explanatory diagram of conventional motion adaptive scanning line interpolation processing, Figure 3 is an example of an Fi moving image, and Figures 4 and 85 are conventional scanning line interpolation. Examples of images with degraded image quality obtained by the process, FIGS. 6 and 7 are operation explanatory diagrams of the motion determination process of the present invention. FIG. 8 shows a block diagram of an embodiment of the present invention. 21.23.24...Field memory, 22°25
... Line memory, 26.28, 32, 33°35.
37.39... Addition (subtraction) circuit, 27.29... Coefficient circuit, 30.31... Multiplication circuit, 4o... Motion coefficient modification circuit. Blade 5 (2) Fig. 9 (L-3) ll-2) H-I) (L)yA8 Fig.

Claims (1)

[Claims] In a circuit that extracts IH information about the movement of a subject contained in an interlace-scanned television signal, a memory with a capacity of at least three fields, a subtraction circuit that calculates a frame difference signal, and a frame difference signal of two successive fields are used. 1. A motion detection circuit comprising: a motion calculation circuit that calculates a motion coefficient using the motion calculation circuit.