JPS62268282A - For image signal coding and decoding system and device - Google Patents

Abstract

PURPOSE:To reduce the amount information generated at the time of panning and zooming by detecting whether panning or zooming occurs, generating moving vectors corresponding to respective movings, and thereby shrinking an inter-frame predicted error signal. CONSTITUTION:A TV camera controling signal inputted to an input terminal 110 from a TV camera console or a TV camera control equipment, is supplied to an intial vector generating circuit 10 where a parameter representing a panning or a zooming in a TV vamera control signal is detected. A quantization circuit 13 adaptively switches a quantization characteristic Qa used when normal and quantization characteristics Qp and Qz exclusively used at the time of panning and zooming, in accordance with the TV camera controlling signal, in order to control the amount of generation of predicted-error information.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to panning and zooming in which the image moves on the entire TV screen.
The present invention relates to an image signal encoding/decoding method and an apparatus thereof, which are characterized in that they take into account processing.

(Prior Art) A predictive coding method is a typical coding method for image signals. In this predictive coding method, the total amount of transmitted information is reduced by transmitting the entire difference signal between the input image signal and the predicted signal, that is, the prediction error signal. The prediction signal is given by a prediction function (), and various predictive coding methods can be obtained depending on how this prediction function is selected. Describe the interframe predictive coding method using the conventional example and j~.

The value Q of the prediction function in the interframe predictive coding method,
It is determined based on the signal of the previous frame. For example, the seventh
As shown in the figure, the value of the prediction function for the signal X at the coordinates (xo, yo) is determined by the signal aK at the same coordinates of the previous frame. Since the signal a has a value very close to the signal X, the prediction error signal has an amplitude value close to O, and the encoding efficiency is high.

However, for a portion where there is a large movement of the image between frames, the signal a takes on all values far from the signal X, so the prediction error signal has a large amplitude value and the encoding efficiency is low.

(Problems to be Solved by the Invention) The above is an overview of image signal encoding systems that have been considered effective in the past. An image can be roughly divided into the subject and its background, and if only the subject is moving, it may be effective or require a little bit of vector information, but for example, when panning, like moving a TV camera filming the subject, the entire screen moves. This is especially true when the subject and the background move simultaneously in the same direction and at the same speed, including small movements of the subject.

At times like this, when panning and zooming, the Vi
This method has a problem in that much more force information is generated than information generated by the movement of the subject alone.

An object of the present invention is to provide an encoding/decoding method that takes into account panning and zooming, which are large force movements of the entire screen that occur when operating an RV camera, and reduces the amount of information generated during panning and zooming. The purpose is to provide equipment.

(Means for Solving the Problems) The present invention (-On the transmitting side, a vector F/I/W representing the movement of the screen is generated from the TV left camera control signal that controls the movement of the TV camera, The image signal is predictively encoded using the vector as an initial vector, the prediction error signal obtained by the predictive encoding is compression encoded, and the T
The V left camera control signal is compression encoded and transmitted [7. On the receiving side, the compression encoded prediction error signal and the TV left camera control signal are decompressed and decoded, and the decompressed and decoded T
An image signal code characterized in that the initial vector is fully restored from the V left camera control signal, and the expanded and decoded prediction error signal is fully predictively decoded using the restored initial vector and reproduced into an image signal. A decoding method is obtained.

Furthermore, according to the present invention, a control signal output from a TV camera control device that controls the movement of the TV camera is input,
an initial vector generation circuit that outputs a vector representing screen motion from the control signal; and a predictive coding circuit for an image signal using the initial vector, and a compression code for a prediction error signal obtained by the predictive coding and the control signal. An encoding device having an encoding circuit for encoding is obtained.

Furthermore, according to the present invention, a signal obtained by compression-encoding the prediction error signal and the control signal obtained from the encoding device on the transmitting side is input, and the compression-encoded prediction error signal and the control signal are input. a decoding circuit that performs decompression decoding; a vector recovery circuit that restores the initial vector from the decompression-decoded control signal; and predictive decoding of the decompression-decoded prediction error signal using the restored initial vector. A decoding device having a circuit for converting the image into an image signal and reproducing it into an image signal is obtained.

(Function) According to the present invention, all parameters during panning or zooming are detected from the control signal of the TV camera, and from these parameters, the vector amount by which the screen has moved, that is, the magnitude and direction of the vector, can be calculated as shown in FIG. The screen is divided into m×n blocks and each block is determined. For example, by referring to the entire vector table prepared in advance, (a) When panning, all blocks have the same size and direction (see Figure 5).
al), and (b) when zooming in on a pufferfish, a vector whose direction is radial with the center at -q2 in the screen (Figure 5 (b)
) −q is generated. This vector has the feature that it can be used directly as a motion vector as information about the position where the screen has moved. During panning or zooming, the initial vectors of all m x n blocks are multiplied by frames, and vectors of the same size and direction continue,
It often becomes a vector with regularity. At this time,
Vector 'f m representing regular screen movement
Instead of transmitting the signal X n times, the TV camera control signal may be transmitted. Furthermore, according to the present method, the amount of generated information can be easily controlled by detecting parameters during panning or zooming, so the encoding efficiency is improved compared to the conventional method.

(Example) An example of the present invention will be described with reference to FIGS. 1 and 2. However, in this embodiment, all the crowd signals are digital signals.

An embodiment applied to the all-frame predictive coding method of the present invention will be described. FIG. 1 is a diagram showing an embodiment of an encoding device. However, it is assumed that there is an A/D converter in front of the input terminal 100, and that there is a fiTV camera operation console or a TV camera control device in front of the input terminal 110.

In FIG. 1, a TV camera control signal input from a TV camera console or fdT camera control device to an input terminal 110 is supplied to an initial vector generation circuit 10.
A parameter representing panning or zooming is detected in the V camera control signal, and the entire vector is output for each block of m x n screen divisions of ta+ or (bl in FIG. 5, for example). The output vector can be treated as it is as a motion vector for the entire screen.The motion vector is supplied to the variable delay circuit 11, and the variable delay circuit 11 calculates the motion from the same position as block A of the current frame, as shown in FIG. The signal in block B of the previous frame stored in frame memo IJ1.4 at a position displaced by vector ■5 is output as the predicted signal of block A of the current frame.This predicted signal is added to P1 subtraction circuit]2. It is input to the circuit 15.

Subtraction circuit I A- is input from input terminal 100 f'1
．． A prediction error signal is generated by subtracting the prediction signal sound from the predicted image signal, and the prediction error signal is supplied to an encoding circuit 161 via a quantization circuit 13 having a function of limiting the total number of possible levels. -j', the prediction error signal output from the quantization circuit 13 is added to the prediction signal output from the variable delay circuit 11 in the addition circuit 15, and the resultant signal is stored in the frame memo 1114 as a locally decoded image signal. Here, in the quantization circuit 13, the TV camera control signal adaptively switches between the quantization characteristic Qa used in normal times shown in FIG. 3, the quantization characteristic Q exclusive for panning and zooming, and Q2. The amount of information generated by the prediction error is controlled. The generated information is a TV camera control signal and a prediction error signal outputted via the quantization circuit 13.

This generation information is compressed and encoded by the encoding circuit 16 and transmitted from the output terminal 120.

FIG. 2 is a diagram showing an embodiment of a decoding device.

In FIG. 2, when a compression-encoded TV camera control signal and a prediction error signal are input to an input terminal 300, the decoding circuit 20 decompresses and decodes them to separate them into a TV camera control signal and a prediction error signal. Among them, a motion vector is reproduced by a vector reproduction circuit 21 from the TV camera control signal.

This motion vector is supplied to the variable delay circuit 22, and the variable delay circuit 22 generates a prediction signal for decoding from the previous frame signal in the frame memory 24. The decompressed and decoded rL prediction error signal is added to the prediction signal in the addition circuit 23 to reproduce an image signal, and when it is output from the output terminal 310, it is simultaneously stored in the π frame memory 24.

Although shown in the embodiments of FIGS. 1 and 2, the initial vector generation circuit 10 and the vector regeneration circuit 21vcV
A memory is required to reference the vector from the iTV power camera control signal.

(Effects of the Invention) In conventional predictive coding methods, when an image in which the entire TV screen moves significantly, such as panning or zooming, is input, the amount of generated information increases; however, in this method, panning and zooming are The inter-frame prediction error signal can be reduced by fully detecting whether or not zooming has occurred, and by generating all motion vectors commensurate with each movement. Further, as described above, by controlling the amount of information generated using the TV camera control signal that detects panning or zooming, the increase in the amount of information generated during panning or zooming can be improved.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of an encoding device according to the present invention, FIG. 2 is a block diagram showing an embodiment of a decoding device according to the present invention, and FIG. A block diagram showing an embodiment of the method, FIG. 4 is a diagram for explaining the screen division in the present invention, and FIG. 5(a)
, (b) is a diagram for explaining the initial vector in the present invention, FIG. 6 is a diagram for explaining the predictive encoding method in the present invention, and FIG. 7 is a diagram for explaining the conventional predictive encoding method. This is a diagram for In the figure, 10: initial vector generation circuit, 11: variable delay circuit, 12: subtraction circuit, 13: quantization circuit, 14:
Frame memory, 15: addition circuit, 16: encoding circuit,
100: Input terminal, 11O two-power terminal, 120: Output terminal, 20: Decoding circuit, 21: Vector reproducing circuit, 22
; variable delay circuit, 23: addition circuit, 24: frame memory, 300: input terminal, 31O: output terminal, 50: quantization circuit normally used, 51: quantization circuit during panning, 5
2: Quantization circuit during zooming, 53: Cell 1/Cuter, 5
4: (child circuit, 150: TV camera control signal,
160: prediction error signal; 170: quantization output signal. ↑2↑0 6 figure t”to”

Claims (3)

[Claims] (1) On the transmitting side, a vector representing the movement of the screen is generated from a TV camera control signal that controls the movement of the TV camera, predictive coding is performed on the image signal using the vector as an initial vector, and the image signal obtained by the predictive coding is The predicted error signal is compressed and encoded, and the TV camera control signal is compression encoded and transmitted. On the receiving side, the compression encoded prediction error signal and the TV camera control signal are decompressed and decoded. An image signal characterized in that the initial vector is restored from the converted TV camera control signal, and the expanded and decoded prediction error signal is predictively decoded using the restored initial vector and reproduced into an image signal. encoding/decoding method. (2) An initial vector generation circuit that inputs a control signal output from a TV camera control device that controls the movement of a TV camera, and outputs a vector representing the movement of the screen from the control signal; An encoding device comprising an encoding circuit that predictively encodes a signal and compresses and encodes a prediction error signal obtained by the predictive encoding and the control signal. (3) A decoding circuit that inputs a compression-encoded signal of the prediction error signal obtained from the encoding device on the transmitting side and the control signal, and decompresses and decodes the compression-encoded prediction error signal and control signal. a vector reproducing circuit that restores the initial vector from the expanded and decoded control signal; and a vector reproducing circuit that predictively decodes the expanded and decoded prediction error signal using the restored initial vector and reproduces it into an image signal. A decoding device having a circuit.