JP2000092496A - Compressed dynamic image data communicating method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a compressed dynamic image data communicating method, which can communicate a dynamic image in real time at low cost and is tolerant of data errors. SOLUTION: Respective frames 1 of the dynamic image to be transmitted are divided into many small areas s1 to s60, frame by frame, by the compressed dynamic image data communicating method for compressing and transmitting the dynamic image, sequential difference data (d) are generated from the small areas s1 to s60 in the small area group, and sequential time-series data of the difference data 8d obtained by the small areas s1 to s60, while the difference data (d) are replaced periodically in specified order with a key frame (k) for restoring image data of small areas by itself.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a compressed moving image data communication method for compressing and transmitting moving images, and more particularly to a compressed moving image data communication method which does not require retransmission of data and which can correct errors. .

[0002]

2. Description of the Related Art Surveillance cameras are provided at a large number of monitoring points provided in large-scale factories or plants, and these monitoring cameras are connected to a central monitoring room so that each monitoring point can be monitored in real time. Surveillance systems are known.

In this surveillance system, moving image data input from a surveillance camera at each surveillance point is, as shown in FIG. 6, a sequence 70 of images (frames) arranged on a time axis t.
Since an enormous amount of data is required for communication as it is, each image is compressed and communicated.

Conventionally, in this moving image compression, a starting point compressed image 80 called a key frame is first created as shown in FIG. 7 by using Motion JPEG for compressing data frame by frame, and this starting point compressed image (frame 0) is created. ) The difference between 80 and the next image (frame 1) on the time axis t is taken and compressed, and subsequent compression is performed on the change (difference) between adjacent images on the time axis t, such as MPEG. Is common.
In this case, since only the difference needs to be processed, a high compression ratio can be obtained.

[0005] In addition, the form of communication of this data over a network can be broadly classified into two forms, "unicast" and "multicast", depending on the number of communication start and end points.

[0006] Unicast is a "telephone type" communication with a start point: end point = 1: 1, and multicast is a "broadcast type" communication with a start point: end point = 1: many. With multicasting, a single data sent by a source computer can be received "simultaneously" by multiple destination computers (instead of sending the data multiple times, the network layer of the protocol automatically sends the data. Multiplex).
A problem in this network communication processing design is whether or not the data reaches the end point without error (whether or not the arrival is guaranteed). For this reason, in unicast, arrival guarantee can be easily provided by a handshake process between the start point and the end point (connection-oriented communication). On the other hand, in multicast, in principle, the number of computers on the end point side can be increased as much as possible, and it is difficult to guarantee data arrival for all end points. It is not given and has a best effort scheme (connectionless communication).

If an attempt is made to provide the same guarantee of arrival as in unicast in multicast communication, there is a problem that the real-time performance of communication is impaired due to error detection and data retransmission. There are problems that the protocol (communication protocol) with the end point becomes complicated and the processing load on the start point increases.
In addition, retransmission frequency prediction and network load prediction become difficult, and a retransmission storm may occur. To perform multicast communication of compressed moving image data in real time, some contrivance is required.

[0008]

SUMMARY OF THE INVENTION For example, Motion JPE
When transmitting compressed video data compressed by G by multicast, it is resistant to communication errors because the data is independent for each frame, but the data compression rate cannot be so high, so the data amount becomes large and the video data is transmitted in real time. Communication is difficult.

[0009] Further, when the compressed moving image data compressed by MPEG is transmitted by multicast, there is a problem that an algorithm is complicated and an expensive dedicated system is required for real-time encoding.

It is an object of the present invention to provide a compressed moving image data communication method capable of real-time moving image communication at low cost and resistant to data errors.

[0011]

According to a first aspect of the present invention, there is provided a compressed moving image data communication method for compressing and transmitting a moving image, wherein each image frame of the moving image to be transmitted is provided. Is divided into a large number of small regions for each frame, difference data is sequentially created from each small region of the small region group, and time-series data of the difference data obtained for each small region is arranged in a predetermined order. This is a method in which time series data is sequentially transmitted while periodically replacing the difference data with a key frame capable of restoring image data of a small area by itself.

According to a second aspect of the present invention, a small area is compressed and transmitted using some of the divided small area groups as a key frame, and the other small areas correspond to the image of the corresponding small area in the immediately preceding image frame. This is a method of calculating the difference between the data and the difference data.

According to a third aspect of the present invention, the receiving side stores data including a key frame for each small area, reproduces a key frame from the data, and previously stores difference data of another small area. This is a method of developing and reproducing the data together with the key frame data, and distributing the reproduced data on a screen to restore a moving image.

According to the configuration of claim 1, since the data amount of each frame is divided, no problem occurs when an error occurs.

According to the configuration of the second aspect, since the difference is obtained from the image of the corresponding small region in the immediately preceding image frame in each region, the image of the small region is updated in a short time.

According to the third aspect of the present invention, the transmitted key frame data alone can restore an image in a small area. Therefore, even if a data error occurs, the entire screen is not affected. The error is easily corrected by the key frame, and the image is restored.

[0017]

Next, a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

In this embodiment, an example will be described in which moving image data having a screen composed of 240 × 320 pixels is communicated using Ethernet.

FIG. 2 is a block diagram of a transmitting side system for implementing the compressed moving image data communication method according to the present invention.
FIG. 4 shows the configuration of the receiving system.

As shown in FIG. 2, the transmitting side system uses an area divider 10 for dividing an image (frame) into a plurality of small areas, and uses data of the divided small areas with a temporal change. To create compressed moving image data while creating the difference data and creating a key frame that can compress the image data as it is and reproduce the data by itself, creating the data as needed and periodically replacing it And a transmission buffer 30 for storing the compressed moving image data.

The small area compressor 20, as shown in FIG.
A key frame cycle setting register 21 for setting a cycle for including a key frame in the difference data; a frame number counter 22 for assigning a frame number and an address to image data input in each small area; and setting of key frame data A compression mode determiner 23 for determining a compression mode of a key frame or difference data from a cycle and a frame number; a compressor 24 for compressing image data of a small area according to the compression mode; And a header generator 25 for adding a header based on information from the frame number counter 22. The compressor 24 holds an image after being divided into small areas as image data, and compares the image data with the next input image data to obtain a difference.

As shown in FIG. 4, the receiving side system includes an area distributor 40 for distributing the transmitted compressed moving image data according to the address in the header, and expands the distributed compressed moving image data. And an area synthesizer 60 for synthesizing the image of each expanded small area into one image.

As shown in FIG. 5, the small area expander 50
A decompression waiting buffer 51 for storing the compressed moving image data of each divided small area, a decompression frame number register 52 for recognizing a frame number of the stored compressed moving image data, It comprises an expansion order controller 53 for arranging the frames in the order of frame numbers and an expansion unit 54 for sequentially expanding the compressed moving image data in the order of numbers.

Next, a compressed moving image data communication method using this system will be described with reference to FIG.

As shown in FIG. 1, a moving image input as a set of a large number of images (frames) 1 arranged on a time axis t by a surveillance camera, etc. The divider vertically divides into a small area group including a large number of small areas s _{1 to} s _{60 each} having a size of 4 × 320 pixels. And with a small area compressor,
A starting point compressed image called key frame k or difference data d is created according to the above-described compression mode. The compressed moving image data p is periodically generated by including the key frame data k in the difference data d by using the key frame period setting register and the frame number counter described above. Thereafter, a header is generated by the header generator by the frame number counter, and the header is added to the compressed moving image data sequence and transmitted. That is, communication control of each small area is performed as a small moving image that is internally independent of the compressed moving image data series.

As described above, since the data is transmitted as independent compressed moving image data p each having a key frame, even when a communication error occurs, the influence on the reproduction screen is reduced.

The communication is UDP (User Datagram Pr
The compressed moving image data transmitted according to the otocol protocol is automatically multiplexed and transmitted at the network layer.

UDP has no functional information other than the port number for data transfer to the application layer in the UDP segment format as its header, that is, does not confirm whether packets have been exchanged normally. Therefore, the algorithm is simple and communication can be performed at higher speed.

Then, in the receiving side system, the input compressed moving image data is allocated to a predetermined address by the area distributor and stored in the expansion waiting buffer 51, and then the expanded moving image data is stored in the expansion order controller 53. The original images 1 are restored by arranging the images of the small areas s _{1 to} s _{60 by} the expander 54 and synthesizing the small areas by the area synthesizer.

As described above, the screen is divided into a number of small areas s _{1 to} s ₆₀ , and the difference data series d is controlled for each of the small areas s _{1 to} s ₆₀ to transmit the key frame k periodically. By doing so, the receiving side can start receiving at any time from any of the small areas. Further, even if a communication error occurs, the error can be corrected by using the next transmitted key frame k, so that the communication can be repaired within a predetermined time without retransmission.

The conventional unicast retransmission requires computation, communication, and time costs. Therefore, the retransmission is stopped, and an error correction signal (key frame data) is periodically transmitted regardless of the presence or absence of an error. As a result, the key frame data can be independently developed and reproduced without depending on the context, so the small area on the end point side in the error state is
It is possible to recover from the error by waiting for the next error correction signal (key frame data). Furthermore, the error correction signal for each small area is not continuously transmitted as in the related art,
Since interleaved transmission (transmission at intervals) is performed, burst errors can be prevented.

Further, in the conventional monitoring system, a moving image at each monitoring point can be viewed only in the central monitoring room. However, according to the present invention, a telephone line or an ISDN installed in a factory or a plant is used. By connecting the data terminal carried by the worker to the network, a moving image at a desired monitoring point can be viewed from anywhere.

In this embodiment, since communication is performed by Ethernet, the data amount of the key frame is set to 4 × 32
Although data communication was performed by dividing a frame into 60 parts for 0 pixels, it goes without saying that the number of divisions is determined according to the transfer capacity of the communication medium.

[0034]

In summary, according to the present invention, the following effects are exhibited.

(1) Even if a communication error occurs in a certain small area, the error does not affect other small areas, so that the entire screen is not affected and a compressed moving image data communication method that is resistant to data errors is provided. realizable.

(2) Portion (Small Area) Disturbed by Communication Error
Is restored within a fixed period, and the receiving side can start receiving at any time with the key frame data to be restored as a starting point, at which time no special processing is required on the transmitting side.

(3) Since the key frame data is not continuous, it is possible to prevent the key frame data from being continuously lost when a burst error occurs.

[Brief description of the drawings]

FIG. 1 is a schematic diagram of a compressed moving image data communication method according to the present invention.

FIG. 2 is a configuration diagram of a transmission side system in FIG. 1;

FIG. 3 is a detailed view of the small area compressor of FIG. 2;

FIG. 4 is a configuration diagram of a receiving-side system in FIG. 1;

FIG. 5 is a detailed view of the small area expander of FIG. 4;

FIG. 6 is a conceptual diagram of a moving image.

FIG. 7 is a diagram for explaining a moving image compression method using a difference between images constituting a moving image.

[Explanation of symbols]

 Reference Signs List 10 region divider 20 small region compressor 30 transmission buffer 40 region distributor 50 small region expander 60 region combiner

Continuation of the front page (72) Inventor Yukihiro Kono 3-1-1, Toyosu, Koto-ku, Tokyo Ishikawajima-Harima Heavy Industries Co., Ltd. Toji Technical Center F term (reference) 5C059 KK15 LC03 MA05 PP04 RC22 RF05 SS06 SS08 SS14 UA02 UA05

Claims (3)

[Claims] In a compressed moving image data communication method for compressing and transmitting a moving image, each image frame of a moving image to be transmitted is divided into a number of small areas for each frame.
Differential data is sequentially created from each of the small regions of the small region group, and the difference data itself is periodically generated in a predetermined order with respect to the time-series data of the difference data obtained for each small region. A compressed moving image data communication method, wherein time-series data is sequentially transmitted while replacing the image data with a key frame that can restore the image data. 2. A small area is compressed by using some of the divided small area groups as a key frame and transmitted, and the other small areas are obtained by taking the difference from the image of the corresponding small area in the immediately preceding image frame. 2. The compressed moving image data communication method according to claim 1, wherein the data is difference data. 3. A receiving side stores data including a key frame for each small area, reproduces a key frame from the data, and develops difference data of another small area together with previously stored key frame data. 2. The compressed moving image data communication method according to claim 1, wherein the moving image is restored by reproducing the moving image and distributing the reproduced data on a screen.