JPH0795590A - Method and device for processing video signal and image pickup device - Google Patents

Abstract

PURPOSE:To compensate a motion by detecting a motion vector for parallel movement or magnification or reduction as to a detected object in a video signal depending on the video signal and a zoom magnification signal so as to surely detect a motion vector even when the object is zoomed. CONSTITUTION:The processing unit is provided with a motion vector detection means 22 detecting a motion vector for parallel translation or magnification/ reduction as to a detected object in a video signal S1 based on the video signal S1, a video signal S5 before a predetermined time and a zoom magnification signal S21. Then the motion vector detection means 22 detects a motion vector and a motion compensation means 24 compensates the motion depending on a video signal S5 before a predetermined time, a motion vector and a zoom magnification signal S21. Thus, a current frame picture is reproduced by using a preceding frame picture and the motion is compensated with high accuracy. Furthermore, when the motion is detected, a zoom magnification obtained depending on the video signal is added for the detection, then the compression rate is improved without losing the picture quality.

Description

Detailed Description of the Invention

[0001]

[Table of Contents] The present invention will be described in the following order. Industrial Application Conventional Technology (FIGS. 3 to 5) Problem to be Solved by the Invention (FIG. 6) Means for Solving the Problem (FIG. 1) Action (FIG. 1) Example (FIGS. 1 and 2) ) The invention's effect

[0002]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video signal processing device, a video signal processing method and an image pickup device, and can be applied to a device for transmitting an image through an image pickup lens such as a television camera.

[0003]

2. Description of the Related Art Conventionally, in a so-called video signal transmission system for transmitting a video signal composed of a moving image to a remote place such as a video conference system and a video telephone system, in order to efficiently use a transmission line, The video signal is encoded by utilizing the correlation of (1) to improve the transmission efficiency of significant information.

For example, in the intra-frame coding process, as shown in FIG. 3, when transmitting each image PC1, PC2, PC3, ... Which constitutes a moving image at time t = t1, t2, t3. The image data to be processed is one-dimensionally encoded and transmitted within the same scanning line. In addition, the inter-frame coding processing uses the autocorrelation of the video signal with respect to the time axis to sequentially display the images PC1 and PC2, PC2
, And the image data PC12, PC23 ... Which is the difference in pixel data between PC3.

Further, the motion compensation predictive coding detects the motion vector between the images PC1 and PC2, PC2 and PC3 ... Which are next to each other in sequence, and compensates this motion vector to perform prediction between frames. As a result, the video signal transmission system high-efficiency-encodes the images PC1, PC2, PC3 ... To digital data having a significantly smaller amount of data as compared with the case of transmitting all the image data, and sends it to the transmission path. Is made possible.

The video signal processing apparatus 1 in such a video signal transmission system is configured as shown in FIG. 4, and the input video signal S transmitted from an image pickup section (not shown).
1 is supplied to the delay circuit 2 and the motion detection circuit 3. That is, the input video signal S1 of the frame transmitted first
Is supplied to the encoding unit 5 via the delay circuit 2 and the subtractor 4.

The encoding unit 5 outputs the input video signal to, for example, D
A CT (Discrete Cosine Transform) conversion circuit or the like performs orthogonal transformation, and then a quantization circuit quantizes and encodes. The encoded signal S2 output from the encoding unit 5 is
It is sent to the recording medium or the transmission path and is supplied to the decoding unit 6.

The decoding unit 6 decodes the input coded signal S2 by, for example, dequantizing it with an inverse quantizing circuit and then performing inverse orthogonal transformation with an inverse DCT converting circuit. The decoded signal S3 decoded by the decoding unit 6 is stored in the frame memory 8 via the adder 7. Here, the frame memory 8 is
It works as a delay circuit for each frame.

Input video signal S1 of n frames (n> 1)
Is supplied to a delay circuit 2 for adjusting the timings of the input video signal S1 and the motion compensation signal S4 and a motion detection circuit 3 for obtaining a motion vector.

An input video signal S1 of n frames and a previous frame signal S5 of (n-1) frames stored in the frame memory 8 are input to the motion detection circuit 3. Accordingly, the motion detection circuit 3 detects the motion by a technique such as block matching, and the motion vector obtained as a result is sent to the motion compensation circuit 9. The motion compensation circuit 9
The motion vector sent from the motion detection circuit 3 and the previous frame signal S5 are input for motion compensation, and the resulting motion compensation signal S4 is input to the subtractor 4 and the delay circuit 10.

The input video signal S1 input to the delay circuit 2 is also input to the subtractor 4 together with the motion compensation signal S4.
The difference signal S6 obtained as a result is input to the encoding unit 5. The encoding unit 5 encodes the input differential signal S6 and outputs it as an encoded signal S2 to the decoding unit 6 and the recording medium or the transmission path.

The decoding unit 6 decodes the input coded signal S2 and sends it to the adder 7. The adder 7 is a delay circuit 1
Motion compensated signal S4 and decoded signal S3 delayed by 0
To restore the input video signal S1 before being subtracted by the subtractor 4 (that is, n frames),
This is sent to and stored in the frame memory 8.

When the video signal is subjected to high-efficiency coding and transmitted in this way and the television camera is panned, as shown in FIG. 5, the motion detection circuit 3 (n-1) frames (see FIG. 5 (A)) and the displacement (ba) of n frames (FIG. 5 (B)) are detected as motion vectors, so that the entire image of (n-1) frames is translated in parallel to the image of n frames. Can be reproduced. Therefore, move the previous frame image by the amount of motion vector,
Since the current frame image can be reproduced, motion compensation can be performed by following the movement of the television camera, and the transmission efficiency is remarkably improved compared to the case of directly transmitting one frame of image data. obtain.

[0014]

By the way, in the video signal processing apparatus 1 as described above, when the video signal is encoded, it is assumed that the movement of the object as the detection target is the parallel movement of the rigid body, and the movement amount is Although it is processed as a motion vector, the subject may be formed not only by panning but also by a combination of motions such as rotation, reduction and enlargement.

When the object is actually panned with zooming as shown in FIG. 6, (n-1) frame (FIG. 6 (A)) and n frame (FIG. 6 (B)) are blocked. Even if the pattern is touched, the corresponding block cannot be found, and as a result, the current frame image must be directly transmitted, which causes a problem. At this time, since the block difference value does not approximate to 0, there is a problem that the compression efficiency is reduced as a whole.

In order to solve such a problem, it is conceivable to obtain the zoom magnification by image processing which performs various calculations on the video signal. According to this method, all the magnifications that the input device can take are multiplied. In addition, since (n-1) frames are prepared and block matching is performed with n frames, there is a problem that the amount of calculation becomes enormous and a large capacity memory is required.

The present invention has been made in consideration of the above points, and proposes a video signal processing apparatus, a video signal processing method, and an imaging apparatus capable of surely detecting a motion vector and compensating for a motion even when a subject is zoomed. Is what you are trying to do.

[0018]

In order to solve such a problem, in the present invention, in the video signal processing device 20 to which the zoom magnification signal S21 is inputted together with the video signal S1,
A motion vector detecting unit 22 for detecting a translation and / or enlargement / reduction motion vector of the detection target in the video signal S1 according to the video signal S1, the video signal S5 before a predetermined time, and the zoom magnification signal S21. To be provided.

Further, according to the present invention, in the video signal processing method for processing the input video signal S1 and zoom magnification signal S21, the motion vector detecting means 22 causes the video signal S1, the video signal S5 before a predetermined time, and the zoom. In accordance with the magnification signal S21, the translation and / or enlargement / reduction motion vector of the detection target in the video signal S1 is detected.

In the present invention, the image pickup lens means 21 having a zoom function for reducing and enlarging the object, and the video signal S1 according to the image pickup result of the image pickup lens means 21,
According to the video signal S5 before a predetermined time and the zoom magnification signal S21 obtained from the imaging lens means 21, the video signal S1
A motion vector detecting unit 22 for detecting a motion vector of translation and / or enlargement / reduction of an object to be detected is provided.

[0021]

The video signal S is detected by the motion vector detecting means 22.
1. Video signal S5 and zoom magnification signal S2 before a predetermined time
According to 1, the motion vector of translation and / or scaling is detected for the detection target in the video signal S1. Further, the motion compensating means 24 carries out motion compensation according to the video signal S5, the motion vector and the zoom magnification signal S21 of a predetermined time. As a result, the current frame image can be reproduced using the previous frame image, and highly accurate motion compensation can be performed.

Image pickup lens means 2 having a zoom function
1 and the image signal S1 according to the image pickup result of the image pickup lens unit 21, the image signal S5 before a predetermined time, and the zoom magnification signal S21 obtained from the image pickup lens unit 21, regarding the detection target in the image signal S1. Motion vector detection means 2 for detecting a translation and / or scaling motion vector
By including 2 and, it is possible to downsize the device, save power, and perform highly accurate motion compensation.

[0023]

Embodiments of the present invention will be described in detail with reference to the drawings.

In FIG. 1, in which parts corresponding to those in FIG. 4 are designated by the same reference numerals, numeral 20 indicates a video signal processing device as a whole, and in addition to the video signal processing device 1 of FIG. The image pickup unit 21 is provided as a unit.

The image pickup section 21 sends an image of a subject picked up by an image pickup lens having a zoom function to the delay circuit 2 and the motion detection circuit 22 as an input video signal S1. Further, in the image pickup unit 21, when a subject is zoomed by the image pickup lens, the lens drive signal S20 of the image pickup lens is input to the zoom magnification detector 23 to detect the zoom magnification, and the zoom magnification signal S21 obtained as a result is a motion. It is input to the detection circuit 22 and the motion compensation circuit 24.

The motion detecting circuit 22 is adapted to obtain a motion vector according to the input zoom magnification signal S21, the input video signal S1 and the previous frame signal S5. The motion compensation circuit 24 is adapted to perform motion compensation by inputting the motion vector, the zoom magnification signal S21 and the previous frame signal S5.

In the above structure, when the object is enlarged by zooming the image pickup lens, the lens drive signal S20 is input to the zoom magnification detector 23 to detect the magnification. The zoom magnification signal S21 obtained as a result is the motion detection circuit 2
2 and the motion compensation circuit 24. In addition, the imaging unit 21
The input video signal S1 transmitted from
And the motion compensation signal S4 are supplied to the delay circuit 2 for adjusting the timing and the motion detecting circuit 22 for obtaining the motion vector.

The motion detection circuit 22 includes a frame memory 8
From the previous frame signal S5 and zoom magnification signal S
21 and the input video signal S1 are input. Here, the motion detection circuit 22 enlarges the input previous frame signal S5 by the magnification of the zoom magnification signal S21 by interpolation calculation, detects the motion by the block matching with the input video signal S1, and obtains the motion vector.

For example, as shown in FIG. 2, the zoom magnification is m
If it is doubled, (n-1) frames (FIG. 2 (A)) are enlarged by m times by interpolation calculation (FIG. 2 (C)) and n frames (FIG. 2 (B)) are used to detect the motion. Find the vector. This motion vector is sent to the motion compensation circuit 24.

Further, the motion compensation circuit 24 enlarges the inputted previous frame signal S5 by the magnification of the zoom magnification signal S21. Thereafter, the motion vector is input from the motion detection circuit 22 to perform motion compensation, and the resulting motion compensation signal S4 is sent to the subtractor 4 and the delay circuit 10.

The input video signal S1 input to the delay circuit 2 is also input to the subtractor 4 together with the motion compensation signal S4.
The difference signal S6 obtained as a result is input to the encoding unit 5. The encoding unit 5 encodes the input differential signal S6, sends it as an encoded signal S2 to the recording medium or the transmission path, and supplies it to the decoding unit 6.

The decoding unit 6 decodes the input coded signal S2 and sends it to the adder 7. The adder 7 includes a delay circuit 10
By adding the motion-compensated signal S4 and the decoded signal S3 delayed by this, the input video signal S1 before being subtracted by the subtractor 4 is restored, and this is sent to the frame memory 8 and stored.

According to the above configuration, the zoom magnification signal S21 obtained according to the lens drive signal S20 sent from the image pickup unit 21 is taken into consideration when the motion is detected and the motion is compensated. In addition to parallel movement, highly accurate motion compensation can be performed for enlargement and reduction, and thus the compression rate can be improved without impairing the image quality of the video signal.

Further, according to the above-mentioned configuration, the zoom magnification is obtained according to the lens drive signal S20 of the image pickup section 21, so that it is necessary to obtain the zoom magnification by image processing for performing various calculations on the video signal. Therefore, the entire apparatus can be downsized and power can be saved.

In the above embodiment, the case where the object is enlarged by the image pickup lens of the television camera is described, but the present invention is not limited to this, and the same effect can be obtained even when the object is reduced. In this case, the motion detection circuit thins out the input previous frame signal to the magnification of the zoom magnification signal, and obtains the motion vector by the block matching with the input video signal.

Further, in the above-described embodiment, the frame memory 8 which plays the role of the delay circuit of the frame unit is used, but the present invention is not limited to this, and the memory which plays the role of the delay circuit of the field unit is used. Is also good. At this time, when the zoom magnification signal is input to the motion detection circuit and the motion compensation circuit, the previous field signal input to each circuit is enlarged or reduced according to the magnification of the zoom magnification signal to obtain an (n-1) field. Similar motion compensation can be performed between n fields.

Further, in the above-mentioned embodiment, the image pickup through the image pickup lens of the television camera is described, but the present invention is not limited to this, and the television is not limited as long as the zoom magnification is input together with the input image information. It may be something other than Jiyoung Camera.

Furthermore, in the above-described embodiment, the case where the motion compensation is performed using the detected motion vector has been described, but the present invention is not limited to this, and only the motion vector is detected and a predetermined process is performed on the video signal. It can also be applied to such a video signal processing device.

[0039]

As described above, according to the present invention, when the motion is detected, the zoom magnification obtained according to the video signal is taken into consideration. Can detect the motion vector, and thus improve the compression ratio without degrading the image quality of the video signal.

[Brief description of drawings]

FIG. 1 is a block diagram showing a video signal processing device according to the present invention.

FIG. 2 is a schematic diagram used for explaining motion detection during zooming.

FIG. 3 is a schematic diagram for explaining an intra-frame / inter-frame encoding process.

FIG. 4 is a block diagram showing a configuration of a conventional video signal processing device.

FIG. 5 is a schematic diagram for explaining panning.

FIG. 6 is a schematic diagram for explaining panning with zooming.

[Explanation of symbols]

1, 20 ... Video signal processing device, 3, 22 ... Motion detection circuit, 5 ... Encoding section, 6 ... Decoding section, 8 ... Frame memory, 9, 24 ... Motion compensation circuit, 21 ... Imaging Part, 2
3 ... Zoom magnification detector.

Claims (6)

[Claims] 1. A video signal processing device to which a zoom magnification signal is inputted together with a video signal, wherein a video signal in the video signal is received according to the video signal, the video signal before a predetermined time, and the zoom magnification signal. A video signal processing device comprising motion vector detection means for detecting a motion vector of translation and / or scaling of a detection target. 2. A motion compensating means for compensating a motion according to the video signal before the predetermined time, the motion vector and the zoom magnification signal, the video signal being the video signal using the motion compensation result. The video signal processing device according to claim 1, characterized in that 3. A video signal processing method for processing an input video signal and a zoom magnification signal, wherein the motion vector detecting means responds to the video signal, the video signal before a predetermined time, and the zoom magnification signal. ,
A video signal processing method, characterized in that a motion vector of translation and / or enlargement / reduction is detected for an object to be detected in the video signal. 4. A motion compensating means performs motion compensation according to the video signal before the predetermined time, the motion vector, and the zoom magnification signal, and encodes the video signal using the motion compensation result. 4. The method according to claim 3, wherein
The video signal processing method described in. 5. An image pickup lens unit having a zoom function for reducing and enlarging a subject, a video signal according to an image pickup result of the image pickup lens unit, and a zoom magnification obtained from the image signal and the image pickup lens unit before a predetermined time. An image pickup apparatus comprising: a motion vector detecting unit that detects a parallel movement and / or an enlargement / reduction motion vector of a detection target in the video signal according to a signal. 6. A video compensating means for compensating the motion according to the video signal before the predetermined time, the motion vector and the zoom magnification signal, wherein the video signal is converted into the video by using the motion compensation result. The imaging device according to claim 5, wherein the signal is encoded.