JPH0799662A - Dynamic image signal transmission method - Google Patents

Abstract

PURPOSE:To adopt the image compression system using the encoding in the direction of the time base as an image compression system without damaging the real time property. CONSTITUTION:An image signal is inputted frame by frame and is compressed and encoded (steps S11 and S12). It is checked whether the signal of the compressed frame should be used as the reference signal of the succeeding frame or not (step S13) If it should be used as the reference signal of the succeeding frame, the compressed signal is transmitted by a TCP protocol (step 14); but otherwise, it is used by a UDP protocol (step S15).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving picture signal transmitting method for compressing an information amount and transmitting a moving picture signal by utilizing correlation in the time axis direction such as interframe correlation or interfield correlation.

[0002]

2. Description of the Related Art Conventionally, as a method of compressing the amount of information and transmitting a moving image signal, for example, "" Multimedia communication over TCP / IP and its performance "(IPSJ Research Report vol.93) , No.58, pp.59-66,93-os-60-9,1993.7.8) ”
There is known a method as described in.

FIG. 2 is a diagram for explaining the moving image signal transmission method described in this document.

In the figure, the moving image signal input from the input terminal 201 is JPEG compression section 202 for each frame.
And is sent to the UDP packetizing unit 203. In this UDP packet unit 203, the input data is segmented and a UDP packet is generated. This U
The DP packet is output from the output terminal 204 to the transmission line.

FIG. 3 is a diagram for explaining the operation of the UDP packet unit 203.

In the figure, the compressed data continuously output from the JPEG compression unit 202 is first segmented so as to match the size of the UDP packet. The figure shows a case where the data of frame 2 is divided into three segments. The segmented data is output with a header for the UDP protocol added.

[0007] In a system such as a video conference that strongly requires real-time processing, even if an error occurs in received data, sending a new screen is given priority over sending an old screen correctly by resending. Therefore, in such a system, the UDP protocol is adopted as the protocol.

The UDP protocol is a protocol that does not retransmit data even if a packet is discarded or an error occurs during transmission. In a system adopting this protocol, when a frame is not completed on the receiving side due to a packet loss or an error, the frame is not displayed, the frame is dropped, and the next frame is displayed.

[0009]

However, in the conventional moving picture signal transmission method using the UDP protocol, there is a high probability that the data of the preceding frame will not be correctly received, and therefore MPEG is a representative image compression method. Such an image compression method using inter-frame coding cannot be used, and only an image compression method using intra-frame coding such as JPEG can be used. As a result, in general, encoding in the time axis direction, which is said to generate a small amount of information, cannot be adopted, so that a sufficient compression rate cannot be obtained.

The problem is that for every frame, T
This can be solved by adopting a method of confirming delivery according to a protocol typified by the CP protocol and retransmitting when a packet discard or error is detected on the receiving side.

However, this increases the probability that data will be correctly received, but increases the amount of transmitted information.
There is a new problem that the real-time property is lost.

Therefore, according to the present invention, a moving image signal transmission in which the image compression method using the encoding in the time axis direction such as the interframe encoding can be adopted as the image compression method without impairing the real-time property. The purpose is to provide a method.

[0013]

SUMMARY OF THE INVENTION The present invention provides a transmission signal,
Separation into a first transmission signal that serves as a reference signal of a subsequent signal and a second transmission signal that does not serve as a reference signal, and different error control methods are used for the first transmission signal and the second transmission signal. Is.

[0014]

According to the present invention, since the signal serving as the reference signal of the subsequent signal is received with high probability without error, a compression method using encoding in the time axis direction can be used as the image compression method. In addition, since the amount of transmitted information does not increase with respect to a signal that does not serve as a reference signal, it is possible to perform transmission with an emphasis on real time.

[0015]

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

FIG. 1 is a flow chart showing the first embodiment of the present invention. In the figure, an image signal is input frame by frame and then compression encoded (step S1).
1, S12).

FIG. 4 is a diagram for explaining an example of image compression coding. In the illustrated example, intra-frame coding is performed every three frames, and inter-frame coding with the previous frame is performed in other frames. Here, the intra-frame coding is a method of coding using only the information of the frame, and the inter-frame coding is a method of coding using the information of the inter-frame difference.

In general, intraframe coding produces more information than interframe coding. Further, the intra-frame coded image can be decoded if only the information of the frame can be received, whereas the inter-frame coded image cannot be decoded even if the information of the frame can be received if the information of the previous frame cannot be received.

Next, it is checked whether or not the signal of the compressed frame is used as the reference signal of the subsequent frame (step S13). When it becomes a reference signal of the subsequent frame, the compressed signal is transmitted using the TCP protocol (step S14), and when it does not become the reference signal, it is transmitted using the UDP protocol (step S).
15).

FIG. 5 is a diagram for explaining packetization. The signals of frame A and frame B are transmitted with a TCP header. The signal of frame C is transmitted with a UDP header. Although the figure shows the case where each frame is transmitted in one packet, it may be segmented into a plurality of packets and transmitted.

In the example of FIG. 4, the frame A becomes the reference frame of the frame B, and the frame B becomes the reference frame of the frame C. Frame C does not become a reference frame of any frame. Therefore, frames A and B are TCP
The frame C is transmitted using the UDP protocol.

If frame A is not received, frame B
Cannot be decoded, and if frame B is not received, frame C cannot be decoded. However, even if the frame C cannot be received, it does not affect the subsequent frames. Therefore,
When the packet of the frame C is discarded or mistaken, the real-time property is emphasized, the data retransmission request is not issued, the frame is dropped, and the transmission of the next compressed signal D is started.

FIG. 6 is a diagram for explaining the retransmission protocol. For the packet transmitted by the TCP protocol, the receiving side performs error detection, and when the occurrence of the error is detected, the transmitting side is requested to retransmit. Error control is not performed for packets transmitted by the UDP protocol.

In FIG. 6, first, the signal of frame A is transmitted. Since the signal of frame A is transmitted using the TCP protocol, an error check is performed on the receiving side.
If there is no error, an Ack (acknowledgement) signal is returned to the transmission side, and the transmission side continues to transmit the signal of frame B.

When an error is detected in the signal of frame B,
A Nack (negative acknowledgment) signal is returned to the sender. When the transmission side receives the Nack signal, the frame B is transmitted again.
The signal of is transmitted.

When the Ack signal is returned in response to the frame B signal, the transmitting side starts transmitting the frame C signal. Since the signal of the frame C is transmitted by the UDP protocol, error control is not performed. Therefore, the transmission side transmits the signal of frame D without waiting for the Ack signal to be returned.

When an error occurs in the signal of frame C, the signal of frame C is discarded. This allows
On the receiving side, the image is displayed with the portion of frame C dropped.

In this embodiment, these operations are repeated to transmit continuous image signals.

According to this embodiment described in detail above, strong error protection is performed for the signal used as the reference signal of the subsequent frame, so that this signal is accurately received with high probability. Therefore, it is possible to suppress the error of the previous frame from propagating to the subsequent frame, and it is possible to use the image compression method using the interframe coding as the compression coding method.

Further, since the amount of transmitted information does not increase for a signal that does not serve as a reference signal, it becomes possible to perform transmission with an emphasis on real-time property.

Furthermore, since the error control method is switched by switching the protocol type,
If the reference signal of the subsequent frame has an error, the signal is retransmitted. This allows the probability that this signal will be correctly received to be very high.

FIG. 7 is a flow chart showing the second embodiment of the present invention. Note that in FIG. 7, the same steps as those in FIG.

In the first embodiment, the case has been described in which the error control method is switched by switching the protocol type based on whether or not it is used as the reference signal of the subsequent frame.

On the other hand, in the second embodiment, based on whether or not it is used as the reference signal of the subsequent frame,
The error control method is switched by using error correction codes having different error correction capabilities.

When used as a reference signal for a subsequent frame, the compressed signal has five BCH (3
1, 11) code is used (step S1)
6). On the other hand, when it is not used as the reference signal, it is encoded using one BCH (15, 11) code capable of error correction (step S17).

FIG. 8 is a diagram for explaining the error correction coding.

A signal of, for example, frame A used as a reference signal is divided every 11 bits, and is transmitted with a 20-bit check code. On the other hand, the signal of the frame C which is not used as a reference signal is divided every 11 bits and is transmitted with a 4-bit check code.

The receiving side uses the check code added at the transmitting side to perform error correction when an error occurs.

Even in such a configuration, strong error protection can be performed on the signal used as the reference signal of the subsequent frame. Therefore, as in the first embodiment, the image using interframe coding is used. The compression method can be used as the compression encoding method.

In the above description, the case where the image compression method using interframe coding is used as the image compression method using the coding in the time axis direction is described. An image compression method using axial encoding, for example, inter-field encoding may be used.

In the above description, the BCH (31,11) code and BCH (1
5, 11) The case where a combination of codes is used has been described, but in the present invention, another combination of error correction codes may be used as long as the combination has a difference in error correction capability.

[0042]

As described in detail above, according to the present invention, strong error protection is performed on the signal used as the reference signal of the subsequent signal, so that this signal is accurately received with high probability. Therefore, the error of the preceding signal is suppressed from propagating to the subsequent signal, and the image compression method using the encoding in the time axis direction can be used as the compression encoding method.

[Brief description of drawings]

FIG. 1 is a flowchart showing a first embodiment of the present invention.

FIG. 2 is a diagram for explaining a conventional moving image signal transmission method.

FIG. 3 is a diagram for explaining the operation of a UDP packetizing unit.

FIG. 4 is a diagram for explaining image code compression in the first embodiment.

FIG. 5 is a diagram for explaining packetization in the first embodiment.

FIG. 6 is a diagram for explaining a retransmission protocol in the first embodiment.

FIG. 7 is a flowchart showing a second embodiment of the present invention.

FIG. 8 is a diagram for explaining error correction coding according to the second embodiment.

[Explanation of symbols]

 A, B, C, D ... Frame.

Claims (1)

[Claims] 1. A moving image signal transmitting method for transmitting a moving image signal by compressing an amount of information by utilizing a correlation in a time axis direction, wherein a transmission signal is a reference signal of a subsequent signal. A moving image signal transmission method, characterized in that a transmission signal and a second transmission signal that does not serve as a reference signal of a subsequent signal are separated, and different error control methods are used for the first transmission signal and the second transmission signal.