JPS642276B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a predictive encoding device for encoding television signals.

A predictive encoding device is a type of encoder that compresses the band of a television signal and converts it into a digital signal. In such a predictive coding device, if a request arises to add new transmission path frame bits or error correction bits or to multiplex audio or data signals due to transmission path conditions, the encoding characteristics for the television signal may deteriorate. It is necessary to meet this requirement without any restrictions.

Conventionally, as a method to meet such requirements, there has been a method of setting the number of samples in the blanking interval to a fixed value, reducing the amount of information generated in the television signal, and obtaining time slots (surplus information bits) necessary for new information. However, in such a method, a part of the blanking interval is set to a fixed value, and therefore it is not possible to satisfy the request from broadcast television personnel to faithfully transmit a waveform. Another method is to obtain time slots for multiplexing new information by somewhat lowering the sampling frequency of the predictive encoder. However, this method has the disadvantage that the prediction function needs to be changed due to a change in the sampling frequency, resulting in a major hardware design change, and that the high-frequency characteristics of the encoding characteristics deteriorate.

The purpose of the present invention is to maintain constant transmission speed and image quality through encoding, and to easily add new information bits such as error correction codes and new frame synchronization signals without changing the hardware of the television signal encoding system. An object of the present invention is to provide a predictive coding device that uses

According to the present invention, prediction has a specific quantizer with a small number of quantization levels that operates only in a predetermined section from among the portions where the signal does not change in a pulse-like manner in the horizontal or vertical blanking section of the television signal. It is composed of an encoder, a buffer memory that stores the output of the predictive encoder, and a multiplexing circuit that multiplexes the output of the buffer memory and new information, and the number of quantization levels of the specific quantizer is assigned to the predictive encoder. A predictive coding device for television signals is characterized in that new information can be multiplexed without deteriorating the coding characteristics of the predictive encoder for the image signal by setting the number of quantization levels to be smaller than the number of quantization levels of the quantizer. can get.

With such a configuration of the present invention, bits for a new transmission path frame and bits for error correction can be added without changing the parameters for the image signal portion of the predictive encoder that have already been set. Therefore, by setting the operating period of a specific quantizer with a small number of quantization levels in accordance with the number of bits to be added, new information bits can be added without causing any deterioration in image quality. This is because the prediction of the predictive encoder is accurate for the signal in the blanking interval, which does not change in a pulse-like manner, and the quantization characteristic of the narrow dynamic range can be applied because the prediction error distribution range is small. For this reason, a blanking interval in which the signal does not change in a pulse-like manner is selected as the interval in which the specific quantizer operates. Therefore, the selected sections may not be continuous depending on the case, and it is only necessary that the sum of the sections becomes the required section length. Note that when the input television signal is a color television composite signal, a high-order prediction function is used.

As described above, according to the present invention, addition of error correction bits and changes in the transmission line frame due to unique circumstances on the transmission line side can be easily performed by changing the transmission line interface part.

Next, the present invention will be explained using the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of a predictive coding apparatus for television signals according to the present invention.

In FIG. 1, an input television signal from an input terminal 1 is sampled and quantized by an A/D converter 3 using a sampling clock phase-synchronized with the input television signal generated by a pulse generator 2, and then converted into a predictive code. A difference circuit 4 calculates the difference between the output of the prediction circuit 5 and the output of the prediction circuit 5. The output of the difference circuit 4 is sent to a 5-bit nonlinear quantization circuit 6,
The output is quantized by a 3-bit nonlinear quantization circuit, that is, a specific quantizer 7, and the output of either the quantization circuits 6 and 7 is selected by the switch 8 in response to a switching signal from the pulse generator 2, and the prediction circuit 5 and is input to the buffer memory 9. The multiplexing circuit 10 multiplexes the output of the buffer memory 9 with the information at the input terminal 11, and sends the multiplexed information from the output terminal 12 to the transmission line. As the prediction function of the prediction circuit 5, a high-order prediction function is selected that can accurately predict color components of a color television signal. The 3-bit quantization circuit 7 has the same fine quantization characteristics as the 5-bit quantization circuit 6.

FIG. 2 is a diagram for explaining the present invention. FIG. 2a shows a television signal waveform, where 1 indicates a blanking section and 2 indicates a picture signal section.
The output of the 3-bit quantization circuit 7 is selected during a certain period of the blanking interval, for example, period A.
In the section, except for the rising and falling sections of the synchronization signal, the prediction by the prediction circuit 5 is very accurate, so that the waveform can be faithfully transmitted using the characteristics of the 3-bit quantization circuit 7. A pulse for writing and reading data to the buffer memory 9 is generated by the pulse generator 2 based on the horizontal synchronization signal of the input television signal.
The write/read operation is started at the pulse timing shown in FIG. 2b, and this becomes the starting point of one digital frame of data on the transmission line (see FIG. 2c).

The multiplexing circuit 10 multiplexes the audio data S and the transmission line frame bits F from the input terminal 11 onto the output of the buffer memory 9, and sends the multiplexed data to the transmission line from the output terminal 12. FIG. 2d shows the structure of a digital frame sent to the transmission line. From FIG. 2D, the capacity of the buffer memory 9 is required to be at least the sum of the number of frame bits and audio data multiplexed by the multiplexing circuit 10. For example, set the sampling frequency _S to _S
= 680 _H ( _H is the horizontal scanning frequency), the digital data generated within one horizontal scanning section is 5 bits (DPCM) x 680 samples = 3400 bits. If a specific quantization circuit, that is, a 3-bit quantization circuit 7, is selected for only 18 samples within the blanking interval, 18×(5-3)=36 extra bits are created.
It is now possible to allocate 12 bits to frame synchronization and 24 bits to audio data, so it is only necessary to provide a buffer memory with a capacity of 36 bits or more.

As described above, according to the present invention, additional data can be inserted within the range of the number of samples in the section a in Figure 2, and the characteristics of the quantizer for the section are not changed, so new data can be multiplexed without affecting image quality. . Further, since the input and output of the buffer memory is calculated in the horizontal scanning period, the capacity of the buffer memory is required to be at least as large as the information bits added within one scanning period.

In this embodiment, the number of quantization circuits is two, and the description is made using 3-bit and 5-bit quantization circuits, but it goes without saying that the present invention is not limited to this. Further, although the horizontal blanking interval is used as the period during which the specific quantizer operates, the present invention is not limited to this, and a vertical blanking interval may also be used. In this case, the capacity of the buffer memory becomes large. In this embodiment, an A/D converter is provided in the input section, and the predictive encoder is constructed entirely of digital circuits; however, the present invention is not limited to this; a nonlinear A/D converter is provided in the quantization circuit, and the predictive circuit It can also be configured in a bibrite type with a D/A converter provided at the output. In addition, in this embodiment, the quantization circuit is expressed as a specific quantizer.
However, a limiter function may be added to the quantization circuit of a conventional predictive encoder, and the lower two bits and the sign bit may be output as the output of a specific quantizer. Furthermore, in this embodiment, the case where the number of samples within one horizontal scanning period of the television signal is constant is explained, but when the number of sampling points within one horizontal scanning period fluctuates, the buffer memory as described below is used. It is necessary to control the operating period of a specific quantizer based on the buffer occupancy rate and to send information on the operating period of the specific quantizer to the receiving section. The number of samples operated by a specific quantizer is placed on the frame sent out on the transmission path, and the position of the sample point between transmission and reception is determined in accordance with the number of samples. Then, the buffer memory occupancy rate of the buffer memory is monitored, and if the buffer memory occupancy rate at the end of one horizontal scanning period is greater than or equal to a fixed value, the number of samples operated by the specific quantizer in the next one horizontal scanning period is increased by, for example, 2 samples. , if the opposite is true, the number of samples is reduced by, for example, 2 samples. In addition, the capacity of the buffer memory is set to twice the additional information bits generated in one horizontal scanning section, and the operation is controlled so that the buffer memory occupancy rate when the device is powered on is 1/2.

As is clear from the above explanation, according to the present invention, the transmission speed and the image quality due to encoding are kept constant,
Moreover, it has the advantage that new information bits such as error correction codes and new frame synchronization signals can be easily added without changing the hardware of the television signal encoding system.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of a predictive coding device for television signals according to the present invention, and FIG.
The figure is a diagram for explaining the operation of the present invention. Explanation of symbols: 1 is the input terminal, 2 is the pulse generator,
3 is an A/D converter, 4 is a difference circuit, 5 is a prediction circuit,
6 and 7 are quantization circuits, 8 is a switch, 9 is a buffer memory, 10 is a multiplexing circuit, 11 is an input terminal, 12
represent the output terminals, respectively.

Claims (1)

[Claims] 1. In a predetermined section of the portion that does not change in a pulse-like manner in the horizontal or vertical blanking section of the television signal, quantization is performed with a number of quantization levels smaller than the normal predetermined number of quantization levels, and the above-mentioned a predictive encoder having a quantization means for performing quantization at the predetermined number of quantization levels outside the predetermined interval; a buffer memory for storing the output of the predictive encoder; and a buffer memory for storing the output of the buffer memory and other A predictive coding device for television signals comprising a multiplexing circuit that multiplexes information.