JPH0754978B2 - Motion compensation frame number conversion method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial field of application) The present invention relates to the conversion of the number of frames to be performed together with the conversion of the number of scanning lines in the conversion of the television system, based on the original image signal of two consecutive frames. BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motion-correction type frame number conversion method for forming an interpolated image signal of a new frame corrected in accordance with a motion, and particularly to a frame number conversion force of a portion that deviates from the image frame in only one of two consecutive frames. This is to prevent the image signal from becoming unpleasant due to the lack of the required image signal.

(Prior Art) In this type of conventional motion-correction type frame number conversion method, as described in, for example, Japanese Patent Application No. 59-244638, filed by the applicant of the present application, continuous two-frame original image signals are used. Form an interpolated image signal of a new frame by linear interpolation, and translate the entire image represented by the interpolated image signal according to the direction and distance of translation detected for the entire image represented by the original image signal. Converts the number of frames by forming a motion-compensated image signal, comparing the inter-frame difference of the original image signal with the inter-frame difference of the motion-compensated image signal, and determining the smaller inter-frame difference as the interpolated frame image signal. It was designed to form an image signal. That is, in the conventional motion compensation type frame number conversion method of this kind, motion compensation is uniformly performed on the entire image based on the motion vector of the image detected for the entire image represented by the original image signal, or at least relatively wide. Based on the motion of the image detected for the entire image portion of the area, the entire image portion is subjected to uniform motion correction.

(Problems to be solved by the invention) Therefore, when there is a large movement in the image portion located in the peripheral portion of the image frame, the image signal of the interpolation frame newly provided for converting the number of frames. Of two consecutive frames before and after the interpolation for forming the image is determined depending on the direction and size of the motion of the corresponding image portion, for example, the corresponding image portion is on the left side of the image frame. When moving in the direction, a part of the moving image portion in the previous frame preceding the interpolating frame often goes out of the image frame as shown in sequence in FIGS. 2 (a) to (d), In such a case, the moving image signal in any one of the two consecutive frames to be interpolated will be omitted. Therefore, at the peripheral portion of the image frame, only the inter-frame interpolated image signal to be applied between two consecutive frames is not normally formed due to the partial lack of the moving image portion. Of course, there is a possibility that the motion detection of the image, which is usually performed based on the difference between at least two consecutive frames, may be hindered. In particular, when the motion detection of such an image is hindered, the subsequent signal processing may be significantly disturbed, resulting in unacceptable noticeable image quality deterioration at the peripheral portion of the display image.

That is, for example, as shown in FIG. 2A, when a columnar moving object moves to the right on the screen by the magnitude of the motion vector v, it is input as two consecutive frames to the frame number conversion circuit. The image (3) of the current frame and the image (2) of the previous frame are horizontally enlarged images (3e) and (3e) as shown in FIGS. 2 (d) and 2 (b) due to the accumulation effect of the television camera. 2e). Therefore, the image of the interpolated frame is subjected to the second interpolation by position interpolation with respect to the images (2e) and (3e) of the two consecutive frames.
As shown in (c), in the motion-compensated frame number conversion of the conventional method, only the size of the motion vector v ₁ is added to the image (2e) of the previous frame, and the image of the current frame (3e
Since the interpolated image signal is formed as 1/2 of the sum of the two types of motion-corrected image signals whose positions are respectively corrected by the size of the motion vector v ₂ , the image of the previous frame ( When the portion indicated by the dotted line in FIG. 2 (b) of 2e) is out of the image frame and is omitted, the above-mentioned image distortion or disorder occurs in the interpolated image signal. was there.

An object of the present invention is to solve the above-mentioned conventional problems, and even when a part of a moving image portion is out of the image frame in the peripheral portion of the image frame, the image portion of the interpolation frame image An object of the present invention is to provide a motion-compensated frame number conversion method that does not significantly deteriorate the image quality due to image distortion or disorder caused by the deletion.

(Means for Solving Problems) That is, according to the motion compensation type frame number conversion method of the present invention, when the conversion of the number of frames is performed together with the conversion of the number of scanning lines in the conversion of the television system, the continuous original image signal of two frames is used. From the original image signal or the motion-corrected image signal obtained by correcting the original image signal with a motion vector in accordance with the movement of the image detected for the original image signal from the new frame formed by linear interpolation. In the motion-compensated frame number conversion method for forming a frame number-converted output image signal by weighting addition of the interpolated image signal and the motion-compensated image signal according to the magnitude of the difference between frames of the motion-compensated image signal, For the original image signal of the part which is out of the image frame in only one of the two frames, The frame number conversion output image signal of the corresponding portion is formed based only on the motion-corrected image signal obtained by correcting the original image signal of the corresponding portion in the other of the two consecutive frames with a motion vector. Is.

To describe the example shown in FIGS. 2A to 2D, the motion-corrected image that does not extend outside the image frame in the motion image portion, that is, the image of the current frame shown in FIG. 2D, for example. By forming the interpolated image signal by using only the position-corrected image in (3e), image distortion or disturbance is prevented from occurring in the peripheral portion of the image frame.

(Operation) Therefore, in the motion correction frame number conversion of the method of the present invention, there is no possibility that unacceptable image quality deterioration will occur in the peripheral portion of the image due to partial omission of the required image signal.

(Example) Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 1 shows an example of the configuration of a circuit device for performing motion compensation type frame number conversion by the method of the present invention. In the motion-compensation frame number conversion device for converting an N frame system image signal into an M frame system image signal with the configuration shown in the figure, an N frame system input image signal is supplied to the frame memory 1 to perform writing and reading in frame order. The image signal (2) of the previous frame read from the frame memory 1 with a delay of one frame period and the image signal (3) of the current frame to be simultaneously written in the frame memory 1 are written in the buffer memories 2 and 3, respectively. , To the linear interpolation circuit 4. In the linear interpolation circuit 4, inter-frame linear interpolation is performed on the image signal (2) of the previous frame and the image signal (3) of the current frame to extract the interpolated image signal (a), and the terminal a of the changeover switch 10 is used. Lead to.

On the other hand, in the puffer memories 2 and 3, the vector detection circuit 9 uses, for example, the above-described mutual relationship of the image signals of at least two consecutive frames such as the image signal (2) of the previous frame and the image signal (3) of the current frame,
Before detecting the vectors v ₁ and v ₂ as described above with reference to FIG. 2 and supplying the motion vectors v ₁ and v ₂ to the buffer memories 2 and 3 respectively and writing them in the buffer memories 2 and 3. Frame image signal (2)
And the image signal (3) of the current frame are read out at the timings of the position corrections corresponding to the directions and the magnitudes of the motion vectors v ₁ and v ₂ , respectively, to read the image of the previous frame that has been motion-corrected. The signal (b) and the motion-corrected image signal (d) of the current frame are respectively formed and led to the terminals b and d of the changeover switch 7, and the sum signal obtained by leading to the adder 5 is added to the coefficient unit 6 , And multiply by a coefficient of 1/2, and the resulting image signals of the previous frame and the current frame that have been subjected to motion compensation respectively, that is, the weighted average image signals (c) by the respective motion vectors, are switched. 7 to the terminal c, and switching control is controlled by a control signal applied from the control circuit 8 as described later, and the motion compensation selected from one of the terminals b, c, d Leading to the terminal e of the switch 10 switches the image signal (e).

In the change-over switch 10, the inter-frame difference between the image signals (2) and (3) of the previous frame and the current frame, or the image signals (b) and (d) of the previous frame and the current frame, which have been subjected to the motion correction, respectively, are set. Switching is controlled according to the magnitude, and either the interpolated image signal (a) from the terminal a or the motion-corrected image signal (e) from the terminal e is taken out as an M frame system image signal. In the conventional motion compensation type frame number conversion method of this kind, the changeover switch 7 is not provided, and only the motion compensation image signal (c) obtained by weighted averaging the previous frame and the current frame from the adder 5 and the coefficient unit 6 is used. Is directly led to the terminal e of the changeover switch 10 as the motion-corrected image signal (e).

Then, in the control circuit 8, the vector detection circuit 9
The changeover switch 7 is controlled according to the following conditions by the motion vector v detected by.

Note that the motion compensation image signal (b) or (d) according to the present invention is controlled by the control circuit 8 and the motion compensation image signal (b) or (d) according to the present invention is used in addition to the motion compensation image signal (c) of the continuous two-frame weighted average as in the conventional method. There is a possibility of being selected as (e): the third image signal of one of two consecutive frames in which the motion-corrected interpolated image signal is to be formed is cut off by the protrusion from the image frame. Only the vertical peripheral portions (1) and (2) having the width Δv and the horizontal peripheral portions (3) and (4) having the width Δy are shown by shading in the figure.

Here, V _x : horizontal direction vector V _{x max} : maximum value of horizontal direction vector V _y : vertical direction vector V _{y max} : maximum value of vertical direction vector V = (V _x , V _y ): motion vector, and scanning If the direction of is the positive direction of the vector, then v = v ₁ −v ₂ for the vectors v ₁ and v ₂ described above with reference to FIG. Then, the control circuit 8 controls the switching of the following conditions (1) to (5).

Condition (1) In the vertical peripheral portion (1), when v _x : The motion-corrected image signal (d) of the current frame is taken out from the terminal d of the changeover switch 7.

When v _x <0: The motion compensation image signal (b) of the previous frame is taken out from the terminal b of the changeover switch 7.

Condition (2) When v _{x in} the vertical peripheral edge portion (2): The motion-corrected image signal (b) of the previous frame is taken out from the terminal b of the changeover switch 7.

When v _x <0: The motion-corrected image signal (d) of the current frame is taken out from the terminal d of the changeover switch 7.

Condition (3) When v _{y in} the vertical peripheral portion (3): The motion-corrected image signal (a) of the current frame is taken out from the terminal d of the changeover switch 7.

When v _y <0: The motion-corrected image signal (b) of the previous frame is taken out from the terminal b of the changeover switch 7.

Condition (4) When v _{y at} the lateral peripheral edge (4): The motion-corrected image signal (b) of the previous frame is taken out from the terminal b of the changeover switch 7.

When v _y <0: The motion-corrected image signal (d) of the current frame is taken out from the terminal d of the changeover switch 7.

Condition (5) In the image areas other than the peripheral portions (1) to (4), the motion-corrected image signal (c) of two frames weighted average is taken out from the terminal c of the changeover switch 7.

The width Δ of the vertical and horizontal peripheral portions (1) to (4) shown in FIG.
It is preferable that x and Δy are set to Δ _x = 2V _{x max} and Δ _y = 2 _{y max} as the maximum widths in which distortion or disturbance of the motion correction image signal may occur due to the lack of the peripheral edge image. is there.

Further, the switching switch 10 is assumed to be switched and connected to the terminal e even in the peripheral portion of the image when the motion vector is detected.

Further, in the peripheral portion of the image frame, as described above, the motion-corrected interpolated image signal (e) is obtained only by the motion-corrected image signal (b) or (d) of the previous frame or the current frame.
Since the signal-to-noise ratio of the image is slightly deteriorated as compared with the central portion of the image frame using the motion-compensated image signal (c) of two frame weighted average, the image quality is substantially the same because it is the peripheral portion of the image. It does not deteriorate.

(Effect of the Invention) As is apparent from the above description, in the conventional motion compensation type frame number conversion method of this kind, inter-frame interpolation in which the same motion compensation is performed is performed over the entire image frame. As a result, the image is distorted or disturbed in the peripheral portion of the image frame of the interpolating frame due to the lack of the image generated in one peripheral portion of the continuous two frames to be interpolated between the frames. In the type frame number conversion method,
Since the interpolated image signal is skillfully formed by using only the frame that does not cause image deficiency for motion compensation in the peripheral portion of the image frame, the number of frames is converted over the entire surface of the image frame without substantial image quality deterioration. The remarkable effect is that an image can be obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration example of a motion-correction type frame number conversion apparatus according to the method of the present invention, and FIGS. 2 (a) to (d) are lines sequentially showing an example of an inter-frame interpolated image formation mode. FIG. 3 and FIG. 3 are line diagrams showing shaded areas in the image portion of the range to which the motion correction of the interpolated image according to the method of the present invention is applied. 1 ... Frame memory, 2, 3 ... Buffer memory 4 ... Linear interpolation circuit, 5 ... Adder 6 ... Coefficient unit, 7, 10 ... Changeover switch 8 ... Control circuit, 9 ... Vector detection circuit

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Yuichi Ninomiya 1-10-11 Kinuta, Setagaya-ku, Tokyo Inside the broadcasting technology research institute of Japan Broadcasting Corporation (72) Taiji Nishizawa 1-10 Kinuta, Setagaya-ku, Tokyo No. 11 Inside the Japan Broadcasting Corporation Broadcasting Technology Laboratory

Claims (1)

[Claims] 1. An interpolated image signal of a new frame formed by linear interpolation from an original image signal of two consecutive frames and the original when the number of frames is converted together with the conversion of the number of scanning lines in the conversion of the television system. The interpolated image according to the magnitude of the difference between the frames of the original image signal or the motion compensated image signal and the motion compensated image signal obtained by correcting the original image signal with a motion vector according to the motion of the image detected for the image signal. In a motion-compensated frame number conversion method for forming a frame number converted output image signal by adding a weight of a signal and the motion-compensated image signal, an original image signal of a portion outside the image frame in only one of the two continuous frames. As for the above, instead of adding the weight, the original image signal of the corresponding portion in the other of the two consecutive frames is moved. Motion compensation type frame number conversion method is characterized in that so as to form a frame number conversion output image signal of the corresponding portion based only on the motion compensated image signal corrected by the vector.