JP2005269276A - Image server and image server system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable the client side to execute multi-screen display different from the monitor output connected to an image server. <P>SOLUTION: The image server has a function of image distribution output independent of a multi-screen monitor output image. The image server simultaneously receives image distribution requests from the same client. The multi-screen display peculiar to the client side can be executed. The difference between the distribution speeds is generated by providing the order of precedence of the client side. Therefore, it is possible to make the image distribution efficient. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a server / client system using a network, and particularly to an image distribution function, for example.

For example, in a conventional network image distribution in which videos from a plurality of monitoring cameras are transmitted to a client via a network, the client side needs to view images from a plurality of monitoring cameras in order to check the status of the monitoring target. At that time, the browser on the client side updates the image after all the image data has been sent. That is, the image cannot be updated unless all the image data is received. In addition, multi-screen display is used to confirm multiple monitoring targets at the same time. To display multi-screen on the client side, the multi-screen is created by combining the screens on the server side. A system is used in which one sheet is distributed to the client side and the client side displays the multi-screen distributed.
Related prior art documents could not be found.

  When a multi-screen is created on the server side and an image is distributed, the resolution of the image of each channel (for each surveillance camera) deteriorates by half or 1/3. On the other hand, if one image is distributed for each channel, the client does not request the next channel image unless the image is received, so that the image update speed becomes slow.

  Also, creating a multi-screen on the server side and distributing the image is the same as the multi-display screen that is output to the monitor of the recording device that records each image for each channel, and this multi-screen is sent to the client. Although it is possible for the client on the network to display a multi-screen by distributing, the image is the same as the output image from the recording device, so the image output in the output screen mode of the recording device and the network Will be the same screen output.

  Furthermore, in conventional multi-user management, both low-level user requests and high-level user requests are treated as equivalent on the server side and image distribution is performed. Therefore, users with low levels during multi-user access have a faster image update speed. There was a case.

  Multiple requests can be received at the same time for requests from clients, and screens for each channel are displayed side-by-side on the multi-screen on the network display screen, enabling network screen output different from the screen output mode of the main unit .

  Expressing the difference between efficient image update and image update speed for each user by performing exclusive control and memory management for requests for each channel and exclusive control and memory management for requests for each user Make it possible.

  Efficient by checking multiple request channels for sequential image distribution requests by allowing multiple image distribution requests from the client browser and exclusive control and memory management dedicated to each channel. Distributing an image can prevent a decrease in image update speed. In addition, since the image for each channel is distributed this time, output in which the screen mode of the main body and the screen mode of the network screen are different is possible. Further, by providing exclusive control and memory management separately for user management, a low-level user does not have a higher image update speed than a high-level user.

  That is, by using the present invention, it is possible to enable network screen mode output different from the screen mode of the main body recording device, and by performing exclusive control and memory management for each channel that allows multiple access from the client, It is possible to realize an improvement in image update speed and a difference in image update speed during multi-user.

  An embodiment of the present invention will be described below.

  FIG. 1 is a block diagram of the server side of this embodiment.

  Reference numeral 1 denotes a camera that captures video, and a decoder 2 converts an analog image signal from the camera 1 into a digital image signal. D1-I / F (IN) 3 converts the image signal from the decoder 2 into a signal suitable for the data bus 8. The SDRAM controller 4 is an SDRAM controller for controlling the SDRAM 6 for Monitor Out, writes the image signal from the D1-I / F (IN) 3 to the SDRAM 6, reads the image signal from the SDRAM 6 and outputs it to the data bus 8, and the SDRAM controller. Reference numeral 5 denotes an SDRAM controller for controlling the Rec Out SDRAM 7, which writes image data from the D 1 -I / F (IN) 3 to the SDRAM 7, reads an image signal from the SDRAM 7, and outputs it to the data bus 8.

  The Monitor Out SDRAM 6 includes a plurality of storage areas (banks), holds image signals in the plurality of storage areas, and the Rec Out SDRAM 7 includes a plurality of storage areas (banks). The image signal is held in the storage area.

  The data bus 8 is connected to the decoder 2 via the D1-I / F (IN) 3, the SDRAM 6 for Monitor Out via the SDRAM controller 4, and the SDRAM 7 for Rec Out via the SDRAM controller 5 to the D 1 -I / F ( OUT) 9 and D1-I / F (OUT) 10 are connected to each other. The D1-I / F (OUT) 9 receives the image signal read from the Monitor Out SDRAM 6 via the SDRAM controller 4 as D1-I / F1 is output to F (IN) 11, and D1-I / F (OUT) 10 outputs the image signal read from Rec Out SDRAM 7 via SDRAM controller 5 to D1-I / F (IN) 12. D1-I / F (IN1) 11 converts the image signal from D1-I / F (OUT) 9 into a signal suitable for the data bus 22, and D1-I / F (IN2) 12 is D1-I / F. (OUT) The image signal from 10 is converted into a signal suitable for the data bus 22, and the IDE-I / F 13 receives an instruction from the CPU 21 and controls the HDD 14.

  The HDD 14 records an image signal (JPEG file) or the like from the data bus 22 in accordance with the control of the IDE-I / F 13, reads the image signal or the like and outputs it to the data bus 22, and the JPEG codec 15 The image signal input from the F (IN1) 12 and temporarily stored in the SDRAM 17 via the data bus 22 is encoded according to the JPEG method, and the image signal obtained by encoding is output to the data bus 22.

  Further, the encoded image signal input from the data bus 22 is decoded according to the JPEG method, and the image signal obtained by decoding is output to the data bus 22. The SDRAM controller 16 is an SDRAM controller that controls the SDRAM 17, and is D1-I. Write image signals from / F (IN) 11 and D1-I / F (IN) 12 to SDRAM 17, read data from HDD, write to SDRAM 17, and encode and decode via JPEG codec via JPEG codec Write the converted image signal. Also, image signals and the like are read from the SDRAM 17 and output to the data bus 22 and the data bus 23. The SDRAM 17 includes a plurality of storage areas (banks), and the image signals, JPEG data, and HDD are stored in these storage areas. Each of the write and read data is stored.

  The expansion bus I / F 18 mediates exchange of data between the network controller 19 and the data bus 23, and the network controller 19 receives an external network access request signal. In response to an instruction from the CPU 21, the image signal or the like held in the SDRAM 17 is transmitted to the network.

  A program describing the processing procedure of the CPU 21 is stored in the flash memory 20, and the CPU 21 controls each of the above components according to the program in the flash memory 20. As a result, various processes such as recording and playback and network transmission of JPEG files are realized.

    The data bus 22 is connected to the HDD 14, the JPEG codec 15, and the SDRAM controller 16 via the SDRAM controller 16 via the D1-I / F (IN 1) 11, D 1 -I / F (IN 2) 12, and IDE-I / F 13, respectively. The data bus 23 is connected to the network controller 19, the flash memory 20, the CPU 21, and the SDRAM 17 via the SDRAM controller 16 via the expansion bus I / F 18.

  The D1-I / F (OUT) 24 is input from the D1-I / F (IN2) 11 and outputs the image signal once held in the SDRAM 17 to the external monitor via the data bus 22 and also held in the HDD 14. The image signal obtained by decoding the JPEG file with the JPEG codec 15 is output to an external monitor.

  This image distribution apparatus is roughly divided into a multiplexer (MPX) unit for processing images and a digital video recorder (DVR) unit for storing images. The MPX unit is connected to a plurality of surveillance cameras, captures and processes the images of each surveillance camera at a predetermined timing, and outputs the images to the DVR unit that is a storage unit. This image distribution device captures video from the outside such as a monitoring camera, creates a recording output (Rec Out) and a monitor output (Monitor Out) in the MPX unit, records the recording output in the DVR unit, The monitor output is output for monitoring. The recording part of the DVR part can record up to a maximum speed of 60 fps. The monitor output uses an on-screen display (OSD) in the DVR unit to superimpose recording time, REC display, etc. on the monitor out from MPX.

  Traditionally, multiple screens on a network were realized using the Monitor Out path. As shown in the block diagram of FIG. 1, the Monitor Out path refers to the input processing system of the ASIC of the image input to the D1-I / F IN 1 in the DVR ASIC unit. Say.

  However, if this path is used, the monitor output screen of the main unit and the network screen are the same image output, so the main unit monitor output screen and the network output screen cannot be made different outputs. Also, in order to make the multi-screen in Monitor Out one screen, for example, if there are four screens, the resolution of the image of each channel is halved. In the present invention, the Rec Out path is used to make the monitor output of the main unit independent of the screen mode of the network, and at the same time, it prevents the deterioration of the resolution of the image reduced by Monitor Out.

  Next, FIG. 2 shows a system configuration diagram of this embodiment. If a multi-screen is configured on the network using the Rec Out path, the image size and the number of images increase, which affects the image update speed on the network. In other words, when the input from the conventional Monitor OUT is received with the D1 input, the MPX screen mode (FULL screen, 4-split screen, or MULTI screen) is set at the time of input. Will be output as is (720x240). However, since REC OUT is used in this method, the input image is the image of each channel, and to create a multi-screen on the network, the channel images must be pasted together, and the 4-screen multi-screen To make the image, 4 images of 720x240 are required, resulting in an increase in image size and number.

  In order to prevent this problem, at present, when a request from a client requests an image once, the next request is not made until the image is acquired. Since only one image output memory is prepared and if the next request is received during distribution, the contents of the memory being distributed may be destroyed. The request is subjected to exclusive control and waited.

  In contrast, the present invention allows simultaneous requests for multiple images. The implementation method is exclusive control and memory management for each channel, and accepts requests other than image requests for the same channel. The client is controlled for each channel, and if the requested channel can be acquired, the client displays it on the browser, and requests the next image from the server after the display. While the image acquisition time is taking, another channel is operating, and even if it takes time to acquire an image of one channel, a request or processing of another channel is not waited for.

  FIG. 3 is a flowchart showing the operation on the server side of this embodiment. When a request from the client is received, the network task wakes up (S1), and after acquiring the requested channel of the client (S2), if there is no exclusive control of the designated channel (S3), an image request is accepted (S4), Channel exclusive control is started (S5). After that, it waits until the acquired image from the Rec Out path becomes the requested channel (S6). When it matches the requested channel (S7), JPEG compression is performed (S8), and it is expanded in the memory for the requested channel (S9). After the memory is expanded, JPEG is transmitted to the client (S10). When the transmission is completed (S11), the exclusive control for the request channel is released (S12), and the task is paused (S13).

  A request from the client can be accepted even during the waiting time existing in the meantime, and the processing of the flowchart can be similarly performed. If it is a request from another channel, it results in waiting for channel matching from the Rec Out path without performing exclusive control. A plurality of channel requests can wait for the matching of the respective request channels at the same time, so that a queue can be created for each channel, and there is no waste in acquiring the channel image of the Rec Out path. The image update speed can be increased by performing such efficient image acquisition and distribution.

FIG. 4 is a timing chart of the channel control method of this embodiment and the conventional method. ID1 is the one with a low user level. In addition, ID2 or ID3 is used here if the user level is high. Conventional control is used for ID1, and channel control is used for ID2 or ID3. An image CH in the figure indicates an input image channel from the Rec Out path. The sections a, b, and c in the figure indicate S1 to S4, S4 to S8, and S8 to S13, respectively, in the flowchart of FIG. The pulse display such as CH1, CH2, and CH3 in the figure indicates that the server has received an image request from the client at the rising edge, and that the server has finished distributing the image to the client at the falling edge. In the channel control, if the channel is different, a request from the client is accepted and each processing operation is performed. On the other hand, in the conventional method, if there is an image request for any channel, the processing operation is performed in sequence until the processing is completed. In this way, the conventional method is used for ID1 with a low user level, and the channel control method is used for ID2 or ID3 with a high user level, so that efficient image acquisition and distribution is performed, and the image update speed is increased. By changing the control method, it is possible to express the difference in image update speed for each user.

It is a block diagram which shows an example of a structure by the side of the server of one Example of this invention. 1 is a system configuration diagram configured on a server side and a client side according to an embodiment of the present invention. FIG. It is a flowchart figure which shows the operation | movement by the side of the server of one Example of this invention. FIG. 3 is a timing diagram of a channel control method and a conventional method according to an embodiment of the present invention.

Explanation of symbols

1: Camera, 3: D1-I / F (IN), 4: SDRAM controller, 5: SDRAM controller, 6: SDRAM (for Monitor Out), 7: SDRAM (for Monitor Out), 9: D1-I / F (OUT), 10: D1-I / F (OUT), 11: D1-I / F (IN), 12: D1-I / F (IN2), 13: IDE-I / F, 15: JPEG codec, 16: SDRAM controller, 17: SDRAM 17, 18: Expansion bus I / F, 19: Network controller, 20: Flash memory, 21: CPU, 21: D1-I / F (OUT)

Claims (4)

      In an image server in which images of multiple channels are stored, an image distribution request is simultaneously received from one client connected to the image server via a network, and an image is distributed to each channel that has received the image distribution request. An image server characterized by       When the channel of the image input to the image server matches the channel of the image requested by the client, the image is distributed to the client, and an image queue waiting for distribution is created for each channel while the channel does not match. The image server according to claim 1.       In an image server system comprising an image server storing images of a plurality of channels connected to a network and a client requesting images from the image server, the client simultaneously receives image delivery requests for a plurality of channels from the client, An image server system, wherein a multi-screen is created based on an image distributed by. An image server having a multi-user access to an HTTP server, wherein the image update speed varies depending on a priority of clients.