JPH01284188A - Packet switching type picture coding transmission equipment - Google Patents

Abstract

PURPOSE:To share the decoding part of a reception side circuit and to simplify the title device by providing the other party separating buffer, a buffer control part to control the writing and reading of the other party separating buffer, and a frame memory selecting control part for the receiving side circuit. CONSTITUTION:For a transmission side circuit, a header signal to indicate the head of each image frame unit and information to indicate which is a transmission device for every packet are added. On the other hand, for the reception side circuit, the other party separating buffer 18 composed of plural buffers of respective image frame units is provided, and the writing and reading of the other party separating buffer 18 are controlled based on transmission equipment information from a packet decomposing part 16 and image frame header information detected before the information is inputted to the other party separating buffer 18. Further, one or plural frame memories 26, which are equipped as so many the frame memories equipped with the other party and store a signal in one frame before, can be switched by a frame memory selecting control part 30 by the output signal of a buffer control part 20. Thus, the decoding part can be shared.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an improvement of a packet-switched image coding and transmitting device, particularly an image coding and transmitting device used in video conferences, video telephones, and the like.

[Prior Art] Image encoding and transmission devices used for video conferences and video telephones have conventionally been intended for transmission paths such as circuit switched networks, and have not been suitable for multi-destination communications.

For this reason, systems that perform multi-point communication using packet-switched networks have begun to be considered.

FIG. 2 shows the configuration of a conventional packet-switched image coding and transmitting device for multi-point communication.

In the same figure, the preprocessing unit (2) has an image input signal (100
) is a circuit that blocks adjacent pixels on the image one by one and creates a dimensional vector signal (101) for each block, and this vector signal (101) is sent to the encoder (101).
1) is done manually.

This encoding unit (1) includes a motion compensation encoding unit (4) and a frame memory (5), and the frame memory (5) stores the encoded and decoded blocks of the previous frame of the current image signal. The image signal is stored.

Furthermore, the motion compensation encoding unit (4) also processes the current vector signal (1
01) and a block stored in the frame memory at the same position on the image as the block corresponding to the current vector signal (101), are created as reference blocks, and the block position information of the reference block and A reference block generation unit that calculates a vector signal, and calculates distortion (for example, Euclidean distortion or absolute value distortion) between the current vector signal (101) and the reference vector signal, and selects a block with the minimum distortion from among the reference blocks. and a distortion calculation section that selects the distortion calculation section. Further, the subtracter (3) calculates the difference between the current vector signal (101) and the vector signal (102b2) of the selected block selected by the motion compensation encoder (4), and calculates the interframe difference signal ( 105) is sent to the quantization encoder (6).

The quantization encoder (6) generates an interframe difference signal (105)
is converted into a quantized encoded signal (10B) and output, and the quantized encoder (7) converts the quantized encoded signal (106) into an encoded/decoded difference signal (107) and outputs it. do. The interframe adder (8) combines the encoded/decoded difference signal (107) and the vector signal (102) of the selected block.
b2) is added and output.

The variable length encoder (9) generates a quantized encoded signal (106)
is converted into a variable-length encoded signal (109), output, and written into a transmission buffer (10).

The sender code addition section (13) contains information (1) unique to the sender station.
10), the packet assembly unit (12) generates a transmission packet from the output of the transmission buffer (10) and information (110), and the transmission line interface unit (14) converts this transmission packet into an electrical Convert to level.

(15) is the transmission line interface section on the receiving side;
(16) is a packet disassembly unit that disassembles a transmission packet, (
31) is a selector that separates packetized signals for each source, and (35) is a decoding unit. This decoding unit (35) includes a variable length decoding unit (21) that decodes variable length codes, a reception buffer (22) that stores variable length decoded signals, and a quantization code output from the reception buffer. a quantization decoding unit (23) that quantizes and decodes the encoded signal;
A motion compensation decoding unit (2) decodes the vector signal of the selected block from the selected position information output from the reception buffer.
5) includes an interframe adder (24) that adds the quantized decoded signal and the vector signal of the selected block.

(26) is a frame memory that stores the encoded/decoded image signal of one frame before the current image signal;
27) is an output buffer that temporarily stores encoded/decoded image signals.

Next, the operation will be explained.

First, an image human input signal (100) is divided into blocks in a preprocessing section (2) and converted into a vector signal (101).

Then, a reference block is generated in the motion compensation encoding unit (4), and in the reference block, the vector signal (101
) is selected, and the selected block position information (102bl) and the selected vector signal (1
02b2) is output to the subtractor (3).

Then, in subtraction 8(3), the vector signal (101
) and the selected vector signal (102b2), and an inter-frame difference signal (105) is output to the quantization encoder (6). The quantization encoder (6) converts the interframe difference signal (105) into a quantization encoded signal (106), and converts the interframe difference signal (105) into a quantization encoded signal (106) and sends it to the quantization decoder (7) and the variable length encoder tR3 (9).
), and the quantization decoding section (7) encodes the quantization encoded signal (106) and converts it into a decoded difference signal (107).
The interframe adder (8) converts the encoded/decoded difference signal (107) into the vector signal (102b) of the selected block and outputs it to the interframe adder (8).
2) and output to the in-loop frame memory (5). The variable length encoder (9) takes in the quantized encoded signal (106) and converts it into a variable length encoder (10).
9) and output to the transmission buffer (10).

The source code adding section (13) supplies source code information (110) unique to the transmitting station to the packet assembling section (12), and the packet assembling section (12) supplies the source code information (110) unique to the transmitting station to the packet assembling section (12).
110) and the transmission buffer (10), and sends this transmission packet to the interface section (14).
In step 4), this transmission packet is converted to an electrical level suitable for the transmission path.

Next, in the transmission line interface unit (15) of the receiving side circuit, the transmission line interface unit (14) of the transmitting side circuit
), and the packet decomposition unit (16) decomposes the packetized signal.

The selector (31) separates this depacketized signal for each source based on the above-mentioned source code, and sends each variable length decoding section (21), (21), . . .
Send it to... The variable length decoding section (21) performs variable length decoding processing on the packet-decomposed signal and outputs it to the reception buffer (22), which accumulates these variable length decoded signals. Quantization decoding unit (
23) and a motion compensation decoding unit (25).

In the motion compensation decoding section (25), the reception buffer (22)
The interframe adder (24) reads out the signal of the selected block using the selected position information from the signals output from the frame adder (24).
The interframe adder (24) adds the quantized decoded signal and the selected block signal and outputs the result to the frame memory (26) and output buffer (27) as an image output signal.
). Based on the image signal from this output buffer (27), information sent from the transmitting side is displayed on the CRT.

[Problem to be solved by the invention]

Conventional Issues However, in the conventional transmission device described above, when receiving video from two or more points, it is necessary to decode the video separately in the receiving circuit, so the number of decoding sections equal to the number of transmitting parties is required. Ta. Therefore, there was a problem in that the cost of the device increased.

This invention was made to solve the problem,
An object of the present invention is to provide an image encoding/transmission device in which a decoding section of a receiving side circuit is shared so that it is not necessary to prepare as many decoding sections as there are partner stations.

[Means for Solving the Problems] In order to achieve the above object, the present invention takes in information after packet disassembly in a receiving circuit and separates it for each destination based on a header signal indicating the beginning of a video frame unit. a buffer control section that controls writing and reading of the destination separation buffer, and a frame memory selection control that switches frame memories based on the video frame header information and source information from the packet disassembly section. The decoding unit is characterized in that the decoding unit is shared when holding a multi-point conference or the like.

[Function] According to the present invention with the above configuration, in the transmitting side circuit, a header signal indicating the beginning of a video frame unit and information indicating which device is the source are added to each packet. The receiving side circuit has a destination separation buffer consisting of a large number of buffers in units of video frames, and the writing/reading control of the destination separation buffer is performed using the source information from the packet disassembly unit and the destination separation buffer. Control is based on video frame header information detected before input to the buffer. Further, the frame memories, which are provided in the same number as the destinations and store the signal of one frame before, are switched by the output signal of the destination separation buffer control. Furthermore, the receiving side circuit is provided with an output buffer for each destination that can store image output signals for a plurality of frames.

[Example] Hereinafter, preferred embodiments of the present invention will be described based on the drawings.

FIG. 1 shows the overall configuration of a packet-switched image encoding and transmitting apparatus according to the present invention.

Incidentally, the same members as those in FIG. 2 described above are given the same reference numerals, and their explanations will be omitted.

The characteristics of the present invention are that the receiving side circuit includes a destination separation buffer that takes in information after packet disassembly and separates it for each destination based on a header signal indicating the beginning of a video frame unit; and a frame memory selection control section that switches the frame memory based on the video frame header information and source information from the packet disassembly section.

In this embodiment, the decoding section (35), which was conventionally provided for each transmitting station, is shared, and a destination separation buffer (18), a buffer control section (28), and a frame memory selection control section (30) are newly added. has been added.

First, a header addition signal indicating the beginning of the video frame unit is inputted from the video frame header addition section (11) to the transmission buffer (10) of the transmission side circuit, and further, the sender code addition section (13) inputs a header addition signal indicating the beginning of the video frame unit. For each □, information indicating which device is the source is added.

In the receiving side circuit, the transmission line interface section (1
5) The incoming signal is decomposed into packets by the packet decomposition unit (16), and at this time, the source of each packet is identified and sorted by the source code added to the head of each packet. The destination separation buffer (18) consists of tk number of buffers in units of video frames, and serves to take in information from the packet decomposition unit and store packets for each source. Also, this destination separation buffer (18)
is equipped with a buffer control section (20), to which the destination discrimination signal is input from the destination discrimination section (17), and the signal from the video frame header detection section (19) is inputted. Based on the command information, the destination separation buffer (
18) write/read control is performed. The destination separation buffer (18) has several layers for changing video frames, and the video frame header detection unit (19) creates change timing. At this time, a flag is attached to the destination separation buffer (1
8).

In this way, when the writing is completed for each packet and the video frame is ready, the destination molecule it < buffer (18)
is sent to the decoding unit (35) and subjected to variable length decoding (21).
will be held. The loop itself in this decoding section has the same configuration as a normal one.

However, when framing and decoding are performed, the previous difference is taken and that difference is decoded. Furthermore, since the previous information must remain in the receiving circuit, a frame memory (26) in the loop is required for each destination.

At this time, for example, the information that has been read is from a specific source A.
, the buffer control unit (20) connects the buffer of the source A for reading, and at the same time, the buffer control unit (20) connects the buffer of the source A to the frame memory selection control (30).
A command is issued to the user to wait as information from source A is coming in. Then, when the information from source A is decoded and accumulated in multiple output buffers (27) prepared for each source, the image output signals from multiple destinations are combined on one screen. Output.

As explained above, according to the embodiments of the present invention, even when holding a multipoint conference, it is possible to use a single decoding unit in common, without having to prepare as many decoding units as there are partner stations. As a result, as the number of partner stations increases, the effect of sharing becomes more pronounced, and the receiving side circuit can be simplified.

[Effects of the Invention] As described above, the present invention provides a receiving side circuit with a separate destination buffer, a buffer control unit that controls writing/reading thereof, and a frame memory selection control unit. By sharing the conversion section, the device can be simplified.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the overall configuration of a packet-switching type image encoding and transmitting apparatus according to the present invention, and FIG. 2 is a diagram showing the overall configuration of a conventional packet-switching type image encoding and transmitting apparatus. In the figure, (1) is an encoding unit, (10) is a transmission buffer, (12) is a packet assembly unit, (16) is a packet disassembly unit, (18) is a destination separation buffer, and (20) is a buffer control unit. , (30) is a frame memory selection control section, (
35) is a decoding unit. In the drawings, the same reference numerals indicate the same or corresponding parts. Agent: Patent Attorney Masuo Oiwa / MQ Pell)

Claims (1)

[Claims] A preprocessing section that collects a plurality of pixels located close to each other on an image of an image input signal into blocks, and creates and outputs a vector signal for each block; and a preprocessing section that takes in signals from the preprocessing section and processes the an encoding unit that encodes a vector signal; a packet assembly unit that packetizes the encoded signal via a transmission buffer; and a transmission side circuit that outputs the packetized signal to a transmission path via an interface unit. , a packet disassembly section that inputs the signal sent from the transmission side circuit via the interface section and disassembles the packetized signal; and a packet disassembly section that divides the packetized signal into each source code and decodes the image signal. A packet that includes a decoding unit that encodes the image, a frame memory that stores a signal for one image of the previous frame, and an output buffer that stores image signals for a plurality of frames and outputs them to a display unit. In an interchangeable image coding transmission device,
The receiving side circuit includes a destination separation buffer that takes in information after packet disassembly and separates it for each destination based on a header signal indicating the beginning of a video frame unit, and a buffer that controls writing and reading of the destination separation buffer. A control unit and a frame memory selection control unit that switches frame memories based on the video frame header information and source information from the packet disassembly unit, and the decoding unit is shared when performing multipoint transmission. A packet-switched image encoding and transmitting device characterized by: