JPH04186985A - Animated picture encoding device - Google Patents

Abstract

PURPOSE:To furthermore improve the picture quality by the generation code quantity being equal to a conventional one by deciding which of an inter-motion compensating frame prediction or an inter-motion compensating field prediction is executed, by a block unit, and executing the prediction, based on this decision. CONSTITUTION:Whether an inter-motion compensating field prediction is executed or an intermotion compensating frame prediction is executed is decided by a block unit divided into field units by using an interlace animated picture signal. In a motion vector detecting circuit 34, for instance, in the case the inter-motion compensating frame prediction is selected, a motion vector is detected against a reproducing picture element value of the previous frame stored in a frame memory 30. In a predicting circuit 32, by using a prediction decision result and the above detected motion vector, the inter-motion compensating field prediction or the inter-motion compensating frame prediction is executed. In such a way, in accordance with the quality of an input signal, the inter- motion compensating field prediction, and the inter-motion compensating frame prediction can adaptively be switched.

Description

[Detailed Description of the Invention] Industrial Application Field The present invention is used in video conference systems, video telephones, surveillance systems, image search systems that search for images, image signal distribution systems that distribute images of television programs, etc. The present invention relates to a moving image encoding device.

2. Description of the Related Art As a conventional video encoding device, after performing an operation to selectively switch between motion compensated interframe prediction, field doma prediction, and intrafield prediction, and calculating a prediction error value,
A block that is a set of multiple pixels (e.g. 8 lines x 8
There is a device that orthogonally transforms the prediction error value for each pixel and encodes the coefficients. An example of using such a video encoding device for image communication is broadcast television, which was standardized by CMTT, a joint committee of CCIR (Consultative Committee for International Radio Telecommunications) and CCITT (Consultative Committee for International Telegraph and Telephone). High-efficiency encoding methods are known.

FIG. 3 shows an example in which a conventional high-efficiency encoding device for moving image signals is used for image communication.

In FIG. 3, 12 is an A/D converter, which is connected to a subtracter 14 and a motion vector detection circuit 34 through a line 13. Further, the subtracter 14 is connected to the prediction circuit 32 through the line 35 or the orthogonal transform circuit 16 through the line 15. The orthogonal transform circuit 16
7 to the coefficient quantization circuit 18. The coefficient quantization circuit 18 is connected to a quantized coefficient encoding circuit 20 through a line 19 and to a quantization width control circuit 40 through a line 41. Quantization coefficient encoding circuit 20
is connected to the quantized coefficient decoding circuit 22 and the line encoding circuit 36 through the line 21.

Quantized coefficient decoding circuit 22 is connected through line 23 to coefficient dequantization circuit 24 .

The coefficient inverse quantization circuit 24 is connected to the orthogonal inverse transform circuit 26 through a line 25 and to the quantization width control circuit 4 through a line 41.
0, and the orthogonal transform circuit 26 is further connected to an adder 28 through a line 27. Adder 28 is connected to prediction circuit 32 through line 35 and to frame memory 30 through line 29. Frame memory 30 is connected through line 31 to prediction circuit 32 and motion vector detection circuit 34 . The motion vector detection circuit 34 is connected to the prediction circuit 32 and the line encoding circuit 36 through a line 33. Line encoding circuit 3
6 is connected through line 37 to transmit buffer 38. The transmission buffer 38 is connected to a quantization width control circuit 40 through a line 39 and to a line through a line 42. Further, the quantization width control circuit 40 is connected to the coefficient quantization circuit 18, the coefficient dequantization circuit 24, and the line encoding circuit 36 through a line 41.

Next, the operation of the above conventional example will be explained.

In FIG. 3, an analog interlaced moving image signal is input to an A/D converter 12 through a line 11, and after being converted into a digital interlaced moving image signal, a line 13
The signal is input to the subtracter 14 and the motion vector detection circuit 34 through the subtracter 14 and the motion vector detection circuit 34.

In the motion vector detection circuit 34, the above-mentioned input digital interlace moving image signal (current field input signal)
Then, using the reproduced pixel values of the previous field and the reproduced pixel values of the previous frame (the field before the previous) read from the frame memory 30, a plurality of candidate motion vectors for the previous frame + previous A motion vector is detected using an evaluation function such as using a zero vector for the reproduced pixel value of the field as the detected motion vector, for example, a vector that minimizes the motion compensation prediction error. Adaptive selection of motion compensated interframe prediction and interfield prediction is performed depending on whether the zero vector for the previous field is selected or the zero vector for the previous field is selected. The detected motion vector is input to a prediction circuit 32 and a line encoding circuit 36 through a line 33.

In the prediction circuit 32, motion-compensated inter-frame prediction or inter-field prediction is performed using the motion vector, the current field, the previous frame, and the reproduced pixel values of the previous field read from the frame memory 30, and the calculated predicted value is: Through line 35, subtracter 14 and adder 2
8 is input. The subtracter 14 uses the input digital incremental video signal and the predicted value to calculate a prediction error value, and the orthogonal transform circuit 16 calculates the prediction error for each block, which is a set of a plurality of pixels. Perform orthogonal transformation.

The orthogonal transform coefficients output from the orthogonal transform circuit 16 are quantized in the coefficient quantization circuit 18, or the quantization width at this time is determined in the quantization width control circuit 40. In the quantization width control circuit 40, the quantization width is set according to the residual information in the transmission buffer 38.When the amount of residual information is large, the quantization width is set to be large in order to suppress the amount of generated information, and when the amount of residual information is small, the quantization width is The wound is made smaller in order to transmit detailed information.

The quantization width is input through line 41 to the line encoding circuit 36 for encoding and transmission to the decoder. The coefficients quantized in the coefficient quantization circuit 18 are encoded in a quantized coefficient encoding circuit 20, and the encoded coefficients are passed through a line 21 to a line encoding circuit 3.
6 is input.

The coefficients encoded by the quantization coefficient encoding circuit 20 are
The quantized coefficient decoding circuit 22 performs local decoding, and then the coefficient dequantization circuit 24 performs dequantization. The inverse quantization width at this time is either the same value as the quantization width used in the coefficient quantization circuit 18, or is input from the quantization width control circuit 40 through the line 41. The predicted error value is reproduced by inversely transforming the inversely quantized coefficients in the orthogonal inverse transform circuit 26, and the predicted value is added to the reproduced predicted error value in the adder 28 to obtain the pixel value. is reproduced and the result is written into the frame memory 30 as previous field data for the next input field.

Note that the motion vector information (including selection information of motion compensated interframe prediction or interfield prediction), orthogonal transform coefficient information, and quantization width input to the line encoding circuit 36 are as follows:
It is line encoded here, rate smoothed in a transmit buffer, and output to the line through line 42.

In this way, even in the above-mentioned conventional video encoding device,
Motion-compensated inter-frame prediction and inter-field prediction make it possible to efficiently transmit moving images with less deterioration in image quality.

Note that this conventional example describes an example of motion-compensated inter-frame and inter-field prediction, and also describes a device that can adaptively select three types of prediction by adding intra-field prediction to this. The same effects can be obtained.

Problems to be Solved by the Invention However, in the above-mentioned conventional video encoding apparatus, motion compensation is not performed for inter-field prediction. However, there is a limit to reducing the prediction error.
There was a problem in that it was difficult to obtain high image quality.

The present invention solves these conventional problems, and aims to provide an excellent video encoding device that can improve image quality with the same amount of generated code compared to conventional devices. That is.

Means for Solving the Problems In order to achieve the above object, the present invention makes it possible to determine whether to perform motion-compensated inter-frame prediction or motion-compensated inter-field prediction on a unit or field basis when detecting a motion vector. It is equipped with a determination circuit that performs determination on a block-by-block basis divided into several regions, and a motion vector detection circuit that performs motion vector detection according to the above determination, so that motion compensation is also performed for inter-field prediction.

Further, in the present invention, the above-described determination is made on a field-by-field basis based on the operation status detected by the camera or the signal from the camera.

Effects The present invention has the following effects due to the above structure. In other words, it is determined whether to perform motion-compensated inter-frame prediction or motion-compensated inter-field prediction in units of units or blocks in which the field is divided into several regions, and prediction is performed based on this determination. The prediction error can be further reduced depending on the nature of the image.
As a result, there is an effect that the image quality can be further improved with the amount of generated codes, such as the conventional amount.

Furthermore, the present invention determines whether to perform motion-compensated inter-frame prediction or motion-compensated inter-field prediction on a field-by-field basis based on operating conditions such as zooming or panning of the detected camera or signals from the camera. Therefore, any of the above predictions can be performed efficiently, the prediction accuracy is improved, and the image quality can be further improved with the amount of generated codes, such as the conventional amount.

Embodiment FIG. 1 shows the configuration of an embodiment of the present invention. In FIG. 1, the same numbers as in FIG. 3 indicate the same components as in FIG. 3. The configuration and operation of the parts in FIG. 1 that are different from the conventional configuration example shown in FIG. 3 will be explained below. Note that this embodiment is an example in which the present invention is applied to image communication.

In FIG. 1, 102 is a motion compensation inter-field prediction/motion compensation compensation inter-frame prediction determination circuit, and a line 101
It is connected to the A/D converter 12 through a line 103, and to a motion vector detection circuit 34, a line encoding circuit 36, and a prediction line 32 through a line 103.

Next, the operation of the above embodiment will be explained.

Using the interlaced video signal input from line 101 in FIG. 1, motion compensation interfield prediction is performed on a field-by-field basis or on a block-by-block basis obtained by dividing a field into several regions as shown in FIG. Determine whether to perform interframe prediction. As an example of the criterion at this time, motion compensated interframe prediction is used for a stationary region or a portion that can be considered as a parallel movement of a rigid body in the horizontal direction.
The translation of the rigid body and the parts that are not shown may be made into motion compensated inter-field prediction. The results determined by the motion compensated inter-field prediction/motion compensated inter-frame prediction determination circuit 102 are sent to the motion vector detection circuit 3 through a line 103.
4, input to the line encoding circuit 36 and prediction circuit 32. In the motion vector detection circuit 34, if motion compensated inter-field prediction is selected according to the judgment result of the motion-compensated inter-field prediction or the motion-compensated inter-field prediction/inter-frame prediction judgment circuit 102, it is stored in the frame memory 30. A motion vector is detected between the reproduced pixel value of the previous field, and if motion compensated interframe prediction is selected, the detected pixel value is calculated between the reproduced pixel value of the previous frame (the field before the previous field) stored in the frame memory 30. Detects motion vectors. The prediction circuit 32 performs motion-compensated inter-field prediction or motion-compensated inter-frame prediction using this motion-compensated inter-field prediction, the motion-compensated inter-frame prediction determination result, and the detected motion vector.

Further, the line encoding circuit 36 performs line encoding on the motion compensated inter-field prediction, the motion compensated inter-frame prediction determination result, the motion vector information, the orthogonal transform coefficient information, and the quantization width, and sends them to the transmission buffer 38.

As described above, according to the above embodiment, since it is possible to adaptively switch between motion compensated inter-field prediction and motion-compensated inter-frame prediction according to the properties of the input signal, the generated code amount is equivalent to that of the conventional device. This has the advantage that it is possible to further improve image quality.

Note that, according to the above embodiment, motion compensated inter-field prediction;
The motion compensated interframe prediction determination is performed using an input signal, or it is possible to use not only the input signal but also other necessary signals. Further, the above determination can be made on a field-by-field basis based on the operation status detected by the camera, such as zooming or panning, or an output signal from the camera. Even in this embodiment, the same effects as those described above can be obtained.

Effects of the Invention As is clear from the above embodiments, the present invention determines whether to perform motion-compensated inter-frame prediction or motion-compensated inter-field prediction on a unit or block basis in which a field is divided into several regions. Since the prediction is made based on this judgment, the prediction error can be further reduced according to the characteristics of the input image, and as a result, the image quality can be further improved with the amount of generated code, such as the conventional amount. There are advantages.

Furthermore, since the present invention determines the above judgment on a field-by-field basis based on the operation status detected by the camera or the signal from the camera, any of the above predictions can be made efficiently. There is an advantage that image quality can be further improved by changing the amount of generated codes.

[Brief explanation of drawings]

FIG. 1 is a schematic block diagram of a video encoding device according to an embodiment of the present invention, FIG. 2 is an explanatory diagram for explaining the operation of the present invention, and FIG. 3 is a schematic diagram of a conventional video encoding device. It is a block diagram. 12... A/D converter, 14... Subtractor, 16...
・Orthogonal transformation circuit, 18...Coefficient quantization circuit, 20...
・Quantization coefficient encoding circuit, 22... Quantization coefficient decoding circuit, 24... Coefficient dequantization circuit, 26... Orthogonal inverse transform circuit, 28... Adder, 30... Frame Memory, 32... Prediction circuit, 34... Motion vector detection circuit, 36... Line encoding circuit, 38... Transmission buffer, 40... Quantization width control circuit, 101... Line, 102... Motion compensation inter-field prediction/inter-frame prediction determination circuit, 103... Line. Agent: Patent attorney Akira Okaji and 2 others! @2 Figure 1 Odd field scanning line

Claims (2)

[Claims] (1) A determination circuit that determines whether to perform motion-compensated inter-frame prediction or motion-compensated inter-field prediction when detecting a motion vector, in units of units or in units of blocks obtained by dividing a field into several regions; a motion vector detection circuit that detects a motion vector based on the judgment of the judgment circuit; 1. A moving picture encoding device that encodes a prediction error signal that is a difference between a predicted value compensated for by a motion vector detected as a target and an input value, or a signal obtained by processing the predicted error signal. (2) A claim characterized in that the determination of whether to perform between the current field or between the current field and the previous frame is determined on a field-by-field basis based on the operation status detected by the camera or the output signal of the camera. The moving image encoding device according to item 1.