JP5336993B2 - Video recording apparatus and video recording method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To achieve shortening of a repair time and the reduction of rewriting times at the same time. <P>SOLUTION: A video recording device includes: a plurality of recording devices 10 provided with a video data circulation recording area, a content information recording area and an address management table in each camera device 3; and a work memory 80 for temporarily holding video data from each camera device 3 in the work area of each recording device, writes content information from the header of the video data to be a writing waiting state to the recording device 10 within the individual work area of the work memory 80, and controls writing of the video data and the content information to update the content information in the recording device 10 at the point of time when the update amount of the content information exceeds a reference value. Here, the synchronization time of the content information and the video data is fixed when mismatching between the video data and the content information occurs by periodically determining the mismatching between the content information in the video data recorded about a correspondence area of the recording device 10 to the camera device 3 and the content information within the content information recording area. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a video recording apparatus for recording video data on a network, and particularly to a case where a plurality of recording devices having different characteristics are used in combination.

  With the speeding up and cost reduction of network technology, it has become common to use an IP (Internet Protocol) network for video data transmission. Along with this, network-compatible video recording apparatuses have been provided, and for example, introduction into video surveillance systems has begun.

  Here, the recording devices used in the network-compatible video recording apparatus are mainly HDD (Hard Disk Drive) and SSD (Solid State Drive), but the characteristics differ depending on the type. First, in terms of the characteristics of the HDD, random access is accompanied by a head seek and can be performed only about 100 times per second. On the other hand, the number of rewritable times exceeds 1 million, and there is no practical problem. On the other hand, in the characteristics of SSD, the random read speed is several tens of thousands times per second, which is two orders of magnitude faster than HDD, but in the case of random write, it is several hundred to several thousand times per second. slow. On the other hand, the number of rewritable times is thousands to tens of thousands, which is about one to two digits less than that of HDD. Further, even if writing is performed with a small data size (several kilobytes) from an application, the access unit (about 64 KB) is rewritten at once in the SSD, so writing with a small size tends to shorten the life of the SSD.

  By the way, the network type digital video recording apparatus has a cyclic recording mode in which overwriting is performed cyclically in the case of a surveillance video, for example (see, for example, Patent Document 1). In this cyclic recording mode, in a steady state, the table of contents information and the video data writing state are always synchronized on the buffer memory of the input stage. That is, the video data from the camera is stored on the buffer memory so as to fit in one block, and then written to the video block that is the storage destination of the recording device. At the same time, the table of contents information on the buffer memory is updated.

  Here, how to write out the table of contents information on the memory to the corresponding area of the recording device becomes a problem. That is, there is a method of writing the table of contents information on the buffer memory to the recording device every time it is updated, and a method of writing it at once after collecting it to some extent.

  In the latter method of writing after a certain amount of data, if an unexpected power failure occurs, a mismatch occurs between the table of contents information and the video data, and the table of contents information is restored from the table of contents information in the header of the video data. Will occur. In the case of an HDD, since random read takes time, it is convenient that the time required for restoration is short when the amount of mismatch between the table of contents information and the video data is small, that is, the rewrite frequency of the table of contents information is high. On the other hand, in the case of SSD, if the rewrite frequency of the table of contents information is increased, the writing becomes a small size, which becomes a factor of shortening the lifetime. In SSD, random read is faster than HDD, so even if the amount of mismatching is large, the repair time is not so long.

  As described above, in the circular recording using the table of contents information and the video data, when using a recording device having different characteristics with respect to the rewriting frequency of the table of contents information such as HDD and SSD, how to reduce the number of times of rewriting the table of contents information, The issue is how to extend the lifespan, how to shorten the repair time, and how to write the table of contents information to the recording device.

Japanese Patent No. 4237549 JP 2007-052516 A

  As described above, in the conventional network-compatible video recording apparatus, in the circular recording using the table of contents information and the video data, a recording device having different characteristics with respect to the rewriting frequency of the table of contents information such as HDD and SSD is used together. The problem is how to reduce the number of times the table of contents information is rewritten and extend the life, how to shorten the time for repairing the table of contents information, and how to write the table of contents information to the recording device.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a video recording apparatus capable of simultaneously reducing the time required for restoring the table of contents information and the number of rewrites.

  To achieve the above object, the present invention provides a plurality of recording devices that circulate and record video data from arbitrary video data supply sources, and a plurality of recording devices that temporarily hold video data from the video data supply sources. A work memory to be sent to each device, and in each recording device, for each video data supply source, a circular recording area for circular recording of the video data, and the table of contents information from within the header of the video data The table of contents information recording area for managing the recording status of the video data and the address management table for managing the recording addresses of the video data and the table of contents information are formed, and the written table of contents information generated at the time of recording the video data Recording apparatus that repairs inconsistencies with the plurality of recording devices in the work memory A plurality of work areas for temporarily storing video data from the video data supply source are formed corresponding to the areas formed in this manner, and the table of contents information is written out from the header of the video data in each work area and the recording is performed. When the writing status of the circular recording area and the table of contents information recording area in the corresponding area in the device is set to the standby state, the rewrite status of each unit block in the corresponding area is managed, and the update amount of the table of contents information exceeds the reference value The work memory control means for controlling the writing of the video data and the table of contents information so as to update the table of contents information in the recording device and the corresponding areas of the video data supply sources of the plurality of recording devices are recorded. Inconsistency between video data and table of contents information by periodically judging inconsistency between table of contents information in video data and table of contents information in table of contents information recording area In the event, which is constituted so as to comprise a synchronization means for constant the synchronization time of the index information and the image data.

  Here, when the plurality of recording devices have different characteristics with respect to the rewriting frequency of the table of contents information, the work memory control means obtains an average access time for the recording device, and calculates the number of updates and the average access time. The time required for repair is obtained from the product, and when the total time required for repair of all the contents blocks exceeds the repair target time, the table of contents information having the longest time required for repair is written to the recording device, and the synchronization means When the inconsistency of the table of contents information occurs, the synchronization time of the table of contents information and the video data is made constant, and at the same time, the number of times of writing the table of contents information to the recording device is suppressed.

  ADVANTAGE OF THE INVENTION According to this invention, the video recording apparatus which can implement | achieve shortening of restoration time of table-of-contents information and reduction of the rewrite frequency simultaneously can be provided.

1 is a block configuration diagram showing an embodiment of a video recording apparatus according to the present invention. The conceptual diagram for demonstrating the recording area according to camera of embodiment shown in FIG. The block diagram which shows the structure of the internal component of embodiment shown in FIG. The conceptual diagram for demonstrating the storage format of the video data of embodiment shown in FIG. The conceptual diagram for demonstrating the address conversion table of embodiment shown in FIG. The conceptual diagram for demonstrating the table of contents information block and video block of embodiment shown in FIG. The conceptual diagram which shows the structure of the non-volatile RAM of the network type video recording apparatus of embodiment shown in FIG. The conceptual diagram which shows the structure of volatile RAM at the time of the steady operation (at the time of video reception / recording / delivery execution) of embodiment shown in FIG. The conceptual diagram for demonstrating the operation | movement at the time of steady operation of the network type video recording apparatus of embodiment shown in FIG. 2 is a flowchart for explaining processing at the time of starting the network type video recording apparatus of the embodiment shown in FIG. 1. The flowchart for demonstrating the processing content of the write thread of embodiment shown in FIG.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a block diagram showing an embodiment of a network compatible video recording apparatus according to the present invention. In FIG. 1, reference numeral 1 denotes a network-compatible video recording apparatus according to the present invention, which is connected to a network 2 and has IP camera apparatuses 3-1, 3-2,. Video signals are transmitted and received between 4-1,..., 4-n.

  The video recording apparatus 1 includes a plurality of recording devices 10-1, 10-2, 10-3 including recording devices such as a hard disk (hereinafter referred to as HDD) and a semiconductor memory (hereinafter referred to as SSD) in a dedicated case. , ..., 10-k. In the following description, when a recording device is represented, it is referred to as a recording device 10.

  The network 2 is a transmission path for transmitting the above-described video signal, and includes a network such as a LAN (Local Area Network), the Internet, or a public line, and connects a plurality of computers to transmit signals such as control signals and data.

  Each of the camera devices 3-1, 3-2,..., 3-m has an IP (Internet Protocol) communication function, and each video signal obtained by photographing the subject or the status of the monitoring area is converted into a JPEG (Joint Photographic Experts Group) format or MPEG4. (Moving Picture Experts Group 4) Compressed in the format, etc., put on IP packet and output to network 2. In the following description, when a camera device is represented, it is referred to as a camera device 3. Some camera apparatuses 3 are provided with sensors such as a microphone or a human sensor (not shown), and these sensor signals are also included in the video signal.

  The information terminal devices 4-1,..., 4-n are, for example, data processing devices such as personal computers, which are equipped with a web browser or the like by network connection and are also called viewing PCs. In the following description, when representing an information terminal device, it is referred to as an information terminal device 4.

  FIG. 2 is a conceptual diagram for explaining that the recording device 10 of the network-compatible video recording apparatus 1 includes a circular recording area 5 for storing video data for each camera apparatus 3. As shown in FIG. 2, in the recording device 10, circulating recording areas 5-1, 5-2,..., 5-m corresponding to the camera devices 3-1, 3-2,. Is secured. Each of the circular recording areas 5-1, 5-2,..., 5-m is an area for holding the video data from the camera devices 3-1, 3-2,.

  That is, in the video recording apparatus 1, when video data is acquired from any of the camera apparatuses 3 via the network 2, the video recording apparatus 1 stores the video data in the circular recording area 5 corresponding to the recording device 10. When a video request from the information terminal device (viewing PC) 4 is received via the network 2, the designated video data is read from the recording device 10 and transmitted to the requesting information terminal device 4.

  From the information terminal device 4, an arbitrary playback method (forward playback / reverse playback, normal speed, fast forward, frame advance, etc.) can be performed from an arbitrary time on an arbitrary camera circular recording area in the circular recording area 5. The operation setting of the network type video recording apparatus 1 is performed using a Web screen or dedicated software that operates on the information terminal apparatus 4 via the network 2.

  FIG. 3 is a functional block diagram showing the internal configuration of the network type video recording apparatus 1. In FIG. 3, reference numeral 6 denotes a computer board. The computer board 6 includes a network interface (I / F) 61, a recording device interface (I / F) 62, a CPU 63, and a volatile RAM (random access memory) 7. The nonvolatile RAM 8 is mounted and connected through the internal bus 60.

  The network interface 61 is a signal conversion unit for Ethernet (registered trademark), and converts signals between the network 2 and the internal bus 60.

  The recording device interface 62 is a signal conversion unit for a recording device bus such as ATA (Advanced Technology Attachment) or SCSI (Small Computer System Interface), and converts a signal between the internal bus 60 and the recording device bus 64. . Here, it is assumed that HDDs 10a, SSDs 10b,... Serving as recording devices are connected to the recording device bus 64 in a mixed manner. For the sake of simplicity, the number of recording devices connected to the apparatus is denoted as D.

  The CPU 63 executes an OS (operating system) and software for operating as the network type video recording apparatus 1. The nonvolatile RAM 8 stores an OS and software for operating as the network type video recording apparatus 1. The volatile RAM 7 is used to temporarily store OS and NDR (Network Digital Recorder (registered trademark)) software temporary information and video data to be transmitted and received between the network and the HDD.

  FIG. 4 shows a method of storing video data on the recording device 10-1 (the same applies to other recording devices). The recording device 10-1 is equally divided into areas having a size sufficiently larger than one unit of input video data. For ease of explanation, this area is hereinafter referred to as a “Block”. For example, if one unit of video data is several tens of kilobytes to several hundreds of kilobytes, the byte size of the block is appropriately several tens of times from several megabytes to several tens of megabytes.

  The recording device 10-1 includes an address conversion table 0 (device → camera area) 11, an address conversion table 1 (camera area → device) 12, a table of contents information block 13 (13-1 to 13-i), The video block 14 (14-1 to 14-j) exists. A necessary number of blocks are secured for each of the tables 11 and 12, and the rest are used as video blocks. In this example, the address conversion table 0 (device → camera area) 11 and the address conversion table 1 (camera area → device) 12 are one block, and the table of contents information block 13 is four blocks 13-1 to 13-i. The video block 14 shows j blocks (14-1 to 14-j).

  The circulation recording area (5-1 to 5-m in FIG. 2) of each camera device 3 is realized by allocating a necessary number of video blocks 14 as an area for each camera device 3 and using the circulation.

  FIGS. 5A and 5B are schematic diagrams showing the format structures of the address conversion tables 11 and 12 shown in FIG. 4, respectively. In the block of the address conversion table 0 (device → camera area) 11 shown in FIG. 5A, the address information 110 exists for the number of video data blocks (j). Information on “what number block of which camera device 3” is stored in the address information. This table 11 reveals what number block on the device 10-1 is assigned to which number block of which camera device 3. If no camera device 3 is assigned, information indicating “free” is stored. This table 11 is also used when searching for a free area when changing the area size for the camera.

  In the block of the address conversion table 1 (camera area → device) 12 shown in FIG. 5B, the address table (12-1 to 12-m) for each camera device 3 and where the table is in this block. There is camera area information 12-0 indicating whether the camera is arranged. The camera area information 12-0 includes information for the number of cameras of 12-0-1 to 12-0-m as information of each camera, and the number of blocks in the circular recording area corresponding to each camera device 3 And information on where in the block (12) the table information of the camera area is described.

  12-1 shows the table information of the circular recording area for the camera device 3-1, and there is address information of 12-1-1-1 to 12-1-X1 (X1 is the number of blocks of the camera 3-1). This address information stores information of “what number block in the area of the camera device 3-1 is what number block in which device” (the same applies to other camera devices 3-2 to 3-m). ). When storing the video data acquired from the camera device 3, the blocks are used in the order described in this table, and when reaching the end, the blocks are used again from the beginning, so that a circular recording area is provided for each camera. Is realized.

  In the above example, for ease of explanation, the address conversion tables 11 and 12 have been described as having one block. However, depending on the capacity of the recording device 10 and the number of bytes of the block, a plurality of blocks may be included. It may be used.

6A and 6B are conceptual diagrams for explaining the table of contents information block 13 and the video block 14 shown in FIG. 4, respectively. As shown in FIG. 6B, the video block 14-1 includes a header portion 140 and video frames 141-1 to 141-f. Video frames 141-1 to 141-f are unit data of video data, and indicate video data from one frame in JPEG and from I frame to the next I frame in the MPEG system.

  The header information 140 includes table of contents information 130, video data address 142, block number 143, and camera number 144. The video data address 142 is information indicating where in the block each video data is stored. The block number 143 indicates a block number indicating what number in the circular recording area this block is created. This block number is a sequential number from 0 and indicates the number of the block created from the start of use of the circular recording area. It should be noted that it does not return to 0 after the start of circular use of the area. The camera number 144 indicates the identification number (1 to m) of the camera device 3.

  On the other hand, as shown in FIG. 6A, the table of contents information 130 includes information on a status flag 131, intra-block head data time 132, intra-block head data number 133, and intra-block frame number 134. The status flag 131 indicates the table of contents information and the status of the video block. The flag states include “unused”, “video data normal”, and “video data corruption”. The in-block start data time 132 indicates the time of the image block start image data. The in-block head data number 133 indicates the data number at the time of the video block head video data. The data number is a serial number from 0 and indicates the number of video data acquired from the start of use of the circulating recording area. The intra-block frame number 134 indicates how many pieces of video data are stored in the video block.

  Here, even when the status flag 131 is in the “video data corruption” state, information when the video data is normal is maintained as the information 132 to 134. If the information is lost when the video data is normal, it is a dummy value that is inferred from the previous and subsequent normal table of contents information and has no contradiction.

  The table of contents information block 13 uses a plurality of blocks 13-1 to 13-i. In this table of contents information block 13, there are as many table of contents information 130 as the number of video blocks (j). The table of contents information 130-1 to 130-j is table of contents information of the video blocks 14-1 to 14-m, respectively. The table of contents information block 13 is used in combination with the address conversion table 1 (camera area → device) to search for video data at an arbitrary time from the camera area.

  The present invention is characterized in that the frequency of writing out the table of contents information is automatically controlled from the upper limit value of the repair time and the characteristics of the recording device. This characteristic processing method will be described with reference to FIGS.

  FIG. 7 shows the contents held in the nonvolatile RAM 7 of the network type digital video recording apparatus 1. The nonvolatile RAM 7 includes an execution file 71 and a setting file 72.

  The execution file 71 is a software group (OS, software for receiving / recording / distributing video data, software for changing settings via the Web, and video recording for each camera device 3 for operating as the network type digital video recording apparatus 1. This is an execution file of software for setting an area.

  The setting file 72 is a file that describes setting information for designating the operation of software for receiving, recording, and distributing video data. In this setting file 72, there are the target restoration time 721, the viewing user upper limit number 722, and the setting information 723 for each camera for the number of cameras (723-1 to 723-m). The target repair time 721 is a numerical value that indicates a time that can be used to repair inconsistency between the table of contents information 13 and the video data 14 when an unexpected power failure occurs during video recording. The viewing user upper limit number 722 is a numerical value indicating an upper limit value of the simultaneous access number from the information terminal device (viewing PC) 4. Hereinafter, this value will be referred to as U for ease of explanation. The per-camera setting information 723-m is setting information for the camera device 3-m, and includes camera connection information 724-m and camera image designation information 725-m. 723-1 includes 724-1 and 725-1. Similarly, 723-m includes 724-m and 725-m. The camera connection information 724 includes information necessary for connection to the camera device 3-1, such as the IP address and port number of the camera device 3-1. The camera image quality designation information 725 includes information for designating image quality such as the number of pixels in one frame, compression rate, frame rate, and the like.

  FIG. 8 shows the state of the volatile RAM 8 during the steady operation of the network type digital video recording apparatus 1 (during video reception / recording / distribution). In the volatile RAM 8 at the time of steady operation, the work memory 80 for each camera is the number of cameras (80-1 to 80-m), and the work memory 81 for each user is the upper limit number of viewing users 722 minutes (81-1 to 81-U). There are as many recording device work memories 82 as the number of recording devices in the apparatus 1 (82-1 to 82-D).

  The per-camera work memory 80 includes an address conversion table (camera → device) 801, a video storage buffer 802, and a table of contents information table 809. As information, the number of blocks 803, latest video time 804, latest video block number 805, It has an old video time 806, an oldest video block number 807, and a block number 808 being created.

  An address conversion table (camera → device) 801 is a table in which the camera circulation area table in the address conversion table 12 of the recording device 10 is read out to a memory. When the received video data and table of contents information are stored in the recording device, the storage location calculation = the “number of blocks in the camera area” value is the value of the “storage destination recording device and its block number”. Used to convert to

  The video storage buffer 802 is a buffer that is used to store video data received from the camera device 3 within a range that can be stored in a block and to write it in the video block 14 of the recording device. In order to store the video data received from the camera device 3 and write to the recording device 10 in parallel, it is composed of a double buffer or more.

The block number 803 indicates the number of blocks secured in the camera circular recording area. This value is obtained by reading the number of blocks in the camera circular recording area of the camera area information 12-0 of the recording device 10 into the memory. The latest video time 804 and the latest video block number 805 indicate the values of the in-block head data time 131 and the block number 143 of the latest video block in the circulating recording area for the camera. The oldest video time 806 and the oldest video block number 807 indicate the values of the in-block head data time 131 and the block number 143 of the oldest video block in the camera circular recording area. The block number 808 being created indicates what number block in the circular recording area for the camera the buffer that is receiving and stuffing video from the camera device 3 is.

  The table of contents information table 809 is table of contents information sorted in the order of addresses of the circulating recording area of the camera. When the network type video recording apparatus 1 is started, the table of contents information 130 of the table of contents information block 13 arranged in the order of addresses on the device is sorted using the address conversion table 801 in the order of addresses in the circulation recording area of the camera. It is a table of contents information table 809.

  The work memory 81 for each user includes a video block reading buffer 811 and has a login ID 812 and a last access time 813 as information. The video block reading buffer 811 is a buffer used for storing video data read from the recording device. The login ID 812 is information for identifying the information terminal device (viewing PC) 4 and is information that the information terminal device 4 issues when accessing when starting viewing. The last access time 813 indicates the last access time from the information terminal device 4. This is used when timeout processing is performed when there is no access for a certain period of time.

  In the work memory 82 for each recording device, the average access time 822, the table of contents information blocks 821 (821-1 to 821-i) and the number of updates 820 (820-1 to 820-1) are stored as information. 820-i). The average access time 822 indicates the average time required for read access from the device measured during the initialization process when the network type video recording apparatus 1 is activated. In order to calculate the average access time 822, a total access time 8221 and a total access count 8222 are provided. During the start-up process, each time a read access is made to the recording device, the access time at that time is added to the total access time 8221, the total access count 8222 is incremented by 1, and the total access time 8221 is calculated as the total access count 8222. The average access time 822 is obtained by division.

  In the table of contents information blocks 821-1 to 821-i, there is table of contents information 8210 (8210-1 to 8210-j) for the number of video blocks of the recording device. The table of contents information block 821 is read from the table of contents information block 13 of the recording device 10 when the network type video recording apparatus 1 is activated. At the time of steady operation, the index information block 821 on the RAM 8 is updated each time a video block is written to the recording device, and the index information 821 on the RAM 8 is stored on the recording device 10 when the update amount exceeds a reference to be described later. Write to the table of contents information 13. At the end, all the table of contents information 821 on the RAM 8 is written back to the table of contents information 13 on the recording device 10 as (821-1 to 821-m).

  The update count 820 indicates the number of differences in the table of contents information between the table of contents information 821 on the RAM 8 and the table of contents information 13 on the recording device 10. The initial value is 0. During steady operation, a video block is written to the recording device 10 and the update count 820 is incremented by 1 each time the table of contents information block 821 on the RAM 8 is updated. When the table of contents information 821 on the RAM 8 is written to the recording device 10, it is reset to zero.

  FIG. 10 is a flowchart showing the startup process of the network type video recording apparatus 1. Step 200 “Start” indicates a point in time when the boot process of the software for the network type video recording apparatus 1 is started after the boot process of the OS unit after the network type video recording apparatus 1 is turned on is completed.

  In Step 201 “Confirm Recording Device Number”, processing for confirming how many recording devices 10 for video recording are connected to the network type video recording apparatus 1 is performed. When there are no recording devices 10 for recording video, in this embodiment, an error state is set, log file output, external notification, and the like are performed, and the startup process is terminated. There may be an operation in which the video data is not relayed but only the video data is relayed, but it is omitted from the description of the present invention. If one or more recording devices for video recording exist, the process proceeds to step 202.

In step 202 “Confirmation of capacity of each recording device”, the capacity in each recording device 10 = the number of blocks is confirmed. The capacity check is performed by actually performing read access and searching for the end. In the process of designating the read address from the recording device 10 and performing the read access, when the address is out of capacity, an “out of range error” indicating that is returned. In the tail search, the read address is doubled until the “out-of-range error” returns, and after the “out-of-range error” is obtained, the read address at the boundary that becomes the “out-of-range error” is narrowed down by the binary search. The read access read size is the byte size of the block. Read count is 2 bottom and the pair about twice the numeric value of (byte size of the volume / block of the recording device) (Example: If the block size 1MB = 2 ^20, 1TB the size of the recording device = 2 ⁴⁰ Log ₂ (2 ²⁰ ) × 2 = 40 times ), and the processing time has no problem in practical use. At this time, the time required for the read access is added to the total access time 8221, the total access count 8222 is added by 1 and used to calculate the average access time 822 for each device. When the capacity confirmation of all the video recording devices is completed, the process proceeds to step 203.

  In step 203 “Securing work memory for each recording device”, the work memory 82 for each recording device corresponding to the number of video recording devices detected in step 201 is secured in the volatile RAM 8. If the memory is insufficient, processing such as log file output and external notification is performed to that effect, and the start-up process is terminated.

  In step 204 “read table of contents information”, the table of contents information block 13 of each video recording device 10 is read and stored in the table of contents information block 821 of the work memory 82 for each recording device secured in step 203.

  In step 205 “camera number confirmation”, the number of cameras is acquired from the camera area information 12-0 of the address conversion table 12 of each video recording device 10. If there are a plurality of video recording media, the process proceeds to step 206 when the address conversion table 12 matches in all video recording media. If they do not match, it is determined that there is a discrepancy between the video recording devices, and processing such as log file output and external notification is performed, and the start-up processing is terminated.

  In Step 206 “Securing work memory for each camera”, the work memory 80 for each camera for the number of cameras acquired in Step 205 is secured in the volatile RAM 8. If the memory is insufficient, processing such as log file output and external notification is performed to that effect, and the start-up process is terminated.

  In step 207 “read address conversion table”, the camera area portion (12-1 to 12-m) of the address conversion table 12 of each video recording device 10 is read and stored in the address conversion table 801 of the work memory for each camera. To do.

  In step 208 “sort by table of contents information camera”, a table of contents information table 809 in the order of addresses of the circulating recording area of the camera is created from the table of contents information 821 arranged in the order of addresses for each recording device 10 using the address conversion table 801. .

  In Step 209 “Search by Table of Contents Information for Each Camera”, it is confirmed whether the table of contents information is continuous. The search is performed in three stages: searching for the table of contents information of the oldest block, searching for table of contents information of the latest block, and confirming that it is continuous between the oldest and latest.

  In the oldest block search, the table of contents information 130 having the smallest in-block head data number 133 is searched, and in the latest block search, the table of contents information 130 having the largest in-block head data number 133 is searched. In the confirmation of continuity, the table of contents information table 809 is treated as a ring buffer, and from the oldest block to the latest block, the value of the first data number 133 in the block of the table of contents information 130 + the number of frames in block 134 is within the block of the next table of contents information. Check if it is continuous with the first data number. If all are continuous, the oldest block information is stored in the oldest video time 806 and the oldest video block number 807 of the camera work memory, and the latest block information is similarly stored in the latest image of the camera work memory. The value stored in the time 804 and the latest video block number 805 is entered, and the value obtained by adding 1 to the latest video block number 805 is entered in the block number 808 being created. If a discontinuous portion is detected, processing such as log file output and external notification is performed to that effect, and the start-up processing is temporarily terminated and the routine proceeds to step 210.

  In step 210 “repair”, processing for repairing inconsistency between the table of contents information block 13 and the video block 14 is performed. When the video block 14 of the inconsistent portion is read from the recording device 10 and the video block 14 is normal, the table of contents information 130 of the header portion of the video block 14 is adopted, and when the video block 14 is abnormal, Consistent dummy table of contents information 130 is created from the table of contents information of the normal video block 14. Details of these restoration processes are described in Patent Document 2. After step 210 ends, the process restarts from step 200.

  In step 211 “calculation of sampling number for each camera”, the sampling number of reading of the video block 14 is calculated. If the number of blocks of the camera device 3 is 1024 or less, the sampling number is 10, and if the number of blocks exceeds 1024, the sampling number = lg (number of blocks).

  In step 212 “camera eye information sampling check”, the table of contents information 130 read from the video data 14 is compared with the table of contents information 130 in the table of contents information table 809. Since the comparison check of all the table of contents information is not realistic from the required time, the oldest block and the latest block acquired at step 209 are compared with the block at the point where the table of contents information table 809 is equally divided by the sampling number calculated at step 211. To do. If all match, the process proceeds to step 213. If an inconsistency is found, processing such as log file output and external notification is performed to that effect, and the start-up process is temporarily terminated and the process proceeds to step 210. The time required for the read access at the time of this comparison is also used for calculating the access average time 822 for each device 10. The time required for read access is added to the total access time 8221, and 1 is added to the total access count 8222.

  In step 213 “read setting file”, the information in the setting file 72 is read into the work memory of the volatile RAM 8.

  In step 214 “steady operation”, the start-up process is terminated and the steady operation is started.

  FIG. 9 is a conceptual diagram for explaining the operation at the time of steady operation of the network type video recording apparatus 1. The processing (reception / recording / distribution) of the network type video recording apparatus 1 is performed in cooperation with threads (Thread) for performing the respective processes, that is, the reception thread 91, the writing thread 92, and the distribution thread 93. It is assumed that the video storage buffer 802 operates with a configuration of a double buffer or more, and there are three states of each buffer: “clogging”, “waiting for writing”, and “empty”.

  The reception thread 91 receives video data with the image quality specified by the image quality information 725 from the camera device 3 specified by the connection destination information 724 of the camera information 723 of the setting file 72 via the network 2 and stores the video. The buffer 802 is filled in the “packing” state buffer. If no more video data can be packed in the “packing” buffer, the buffer status is set to “waiting for writing”, and another attempt is made to secure a buffer 802 in the “empty” status. If it can be secured, 1 is added to the block number 808 being created, the buffer status is set to “clogging”, the value of the block number 808 being created is entered as the block number 143 of the header 140, and the camera number 144 The identification number (1 to m) of the camera device 3 in charge is entered, and the video stuffing is continued. If there is no “empty” buffer, it waits until an “empty” buffer appears.

  The reception thread 91 operates for the number (m) of the camera devices 3 that perform reception. The writing thread 92 performs processing for writing video data and table of contents information from the volatile memory 7 to the recording device 10. Details will be described later with reference to FIG. The distribution thread 93 receives a video request from the information terminal device (viewing PC) 4 via the network 2 and searches the table of contents information table 809 to determine which video block 14 contains the requested video data, and the video block number. The video block is read from the recording device 10 to the video block reading buffer 811 and the requested video data is distributed to the information terminal device (viewing PC) 4.

  FIG. 11 is a flowchart showing the processing contents of the writing thread 92. In FIG. 11, Step 300 “Start” indicates a point in time when the writing thread 92 is generated.

  In step 301 “search for“ waiting for writing ”video reception buffer”, the video storage buffer 802 of each IP camera work memory (80-1 to 80-m) is examined to search for a buffer whose status is “waiting for writing”. If it does not exist, the process proceeds to step 307, and if it exists, the process proceeds to step 302.

  In step 302 “address conversion”, the block number 143 and the camera number 144 in the header portion 140 of the buffer detected in step 301 are used as the address of the video block 14 to which the recording device 10 is written, using the address conversion table 801. Perform the conversion.

  In step 303, the buffer found in step 301 is written to the video block 14 at the address on the recording device 10 converted in step 302, and the state is set to "empty".

  In step 304, the table of contents information 130 of the video block recorded in the device in 303 is stored in the table of contents information table 809 sorted by camera area of the work memory 80 for each camera and the table of contents information block 821 of the work memory 82 for each recording device.

In the storage processing in the table of contents information table 809, the storage destination address in the table of contents information table 809 is calculated, the table of contents information 130 of the header section 140 is stored at the address, and the latest and oldest time of the work memory 80 for each camera. The block number (804 to 807) is also updated. The address also updates the block number (804 to 807).

  In the storing process in the table of contents information block 821, the table of contents information 130 of the header section 140 is copied to the table of contents block 8210 on the work memory 82 for each recording device based on the address obtained in step 302, and the table of contents information 8210 is stored. One is added to the update count 820 of the table of contents block 821 to which it belongs.

  In step 305, the repair time required for each table of contents block 821 is obtained from the product of the number of updates 820 and the average access time 822, and it is determined whether the total repair time for all table of contents blocks 821 exceeds the repair target time 721. If exceeded, the process proceeds to step 306, and if not exceeded, the process proceeds to step 307.

  In step 306, the table of contents information 821 having the longest time required for repairing among the time required for repairing each table of contents block 821 obtained in step 305 is written to the table of contents information block 13 of the recording device, and the number of times the table of contents information block 13 is updated Reset 820 to zero.

  In step 307, the process proceeds to step 301 after waiting for several milliseconds.

  As described briefly in FIG. 11, in step 305, in addition to the above determination, whether or not the number of updates 820 for each block exceeds a threshold value may be added to the determination. If this determination is added, when updates are concentrated only on a specific table of contents block 821, such as when the writing destination of a video block is biased, most of the table of contents information 8210 in the table of contents block 821 is updated. In other words, it is difficult to reach 305 in the total number, and it is possible to avoid the phenomenon that writing to the recording device is difficult to be performed.

  When this processing is added, the threshold value should be a variable given from the setting, and the value should be a number equal to or less than the number of table of contents information 8210 in the table of contents block 821. The number of table of contents information 8210 in the table of contents block 821 may be given as a percentage.

  Alternatively, if the total repair time of all the table of contents blocks 821 exceeds the repair target time 721 in step 305, the table of contents 821 is stored in the recording device in step 306. You may make it write out.

  By using this table-of-contents information write-out control, in the case of a recording device with fast read access like the SSD 10b, the write-out frequency is reduced, so that the effect of reducing the number of rewrites and extending the life of the SSD can be obtained. Also, in the case of a recording device with slow read access such as the HDD 10a, the writing frequency increases, so that the effect of reducing the number of inconsistencies and reducing the time required for repair processing can be obtained. Therefore, even when the HDD 10a and the SSD 10b are used in combination in the recording device 10, the effect of making the table-of-contents information restoration time a predetermined time or less can be obtained.

  In the description of the above embodiment, the writing thread 92 searches the video storage buffer 802 for a simple explanation, but the receiving thread 91 writes video blocks in the order in which video data has been packed. It is better to mount the video storage buffer 802 in the form of FIFO (First In First Out) from the reception thread 91 to the writing thread 92.

  In the description of this embodiment, it has been described that the writing thread 92 monitors the time required for repairing each table of contents information 821 and writes out the recording device 10, but this is not a limitation, and other threads such as a network, for example, Processing may be performed by a thread for monitoring the software state of the type video recording apparatus 1.

  In addition, the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. Various inventions can also be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine the component covering different embodiment suitably.

1 ... Network compatible video recording device,
2 ... Network,
3,3-1 to 3-m ... IP camera device,
4,4-1 to 4-n ... information terminal device (viewing PC),
5,5-1 to 5-m ... circulation recording area,
6 ... Computer board, 60 ... Internal bus, 61 ... Network interface (I / F), 62 ... Recording device interface (I / F), 63 ... CPU, 64 ... Bus for recording device,
7 ... Volatile RAM (random access memory), 71 ... Execution file, 72 ... Setting file, 721 ... Restoration target time, 722 ... Viewing user upper limit number, 723, 723-1 to 723-m ... Setting information for each camera, 724 ... Camera connection information, 725 ... Camera image quality designation information, 8 ... Non-volatile RAM, 80, 80-1 to 80-m ... Work memory for each camera, 801 ... Address conversion table (camera to device), 802 ... Video storage Buffer, 803 ... Number of blocks, 804 ... Latest video time, 805 ... Latest video block number, 806 ... Oldest video time, 807 ... Oldest video block number, 808 ... Block number being created, 809 ... Table of contents information table, 81 81-1 to 81-U: Work memory for each user, 811: Buffer for reading video block, 812: Login ID 813 ... Last access time, 82, 82-1 to 82-D ... Work memory for each recording device, 820, 820-1 to 820-i ... Number of updates, 821, 821-1 to 821-i ... Table of contents information block, 8210 , 8210-1 to 8210-j ... Contents information, 822 ... Average access time, 8221 ... Total access time, 8222 ... Total access count,
91 ... Receiving thread, 92 ... Writing thread, 93 ... Delivery thread,
10, 10-1 to 10-k ... recording device, 10a ... HDD, 10b ... SSD,
11 ... Address conversion table 0 (device → camera area), 110 ... address information,
12, 12-1 to 12-m ... address conversion table 1 (camera area → device), 12-0, 12-0-1 to 12-0-m ... camera area information, 12-1-1 to 12-1 -X1 ... address information,
13, 13-1 to 13-i ... Contents information block, 130, 130-1 to 130-j ... Contents information, 131 ... Status flag, 132 ... First data time in block, 133 ... First data number in block, 134 ... Number of frames in block,
14, 14-1 to 14-j: video block , 14-1 to 14-m: video block, 140: header portion, 141-1 to 141-f ... video frame, 142 ... video data address, 143 ... block Number, 144... Camera number.

Claims (4)

A plurality of recording devices for circulatingly recording video data from arbitrary video data supply sources,
A work memory that temporarily stores video data received from the video data supply source and sends it to each of the plurality of recording devices;
Individual plurality of recording devices, assigned to each video data source in units of predetermined blocks, and record area for circulating recording the video data, the header block of the image data to be recorded there are a table of contents information recording areas Contents writes the information for managing the recording state of the video data in units of the predetermined block, and address management table for managing the recording address of the image data is formed,
Video recording for repairing inconsistency between the video data circularly recorded in the recording area formed in any of the plurality of recording devices corresponding to the video data supply source and the written table of contents information A device,
When starting the video recording device, the table of contents information of the table of contents information recording area is copied to the work memory,
A buffer is formed in each of the work memories, and the received video data is packed into the buffer within a range that can be accommodated in the predetermined block. The contents information on the work memory is updated with the contents information written from the header, while writing to the contents information recording area is in a standby state, and the contents information in the waiting state is updated. work memory control value calculated from the number of times to control the writing of the video data and the index information so as to perform writing for the contents information in the recording device at the time exceeds the reference value prescribed for each of the plurality of recording devices Means,
Check the inconsistency between the table of contents information in the video data in the recording area and the table of contents information in the table of contents information recording area, and if there is a mismatch, the block of the video data of the inconsistent portion from the corresponding recording device reading, if the read video data is normal, repair contents information table of contents information recording area from the table of contents information of the header of the read video data, and a synchronization means that they can synchronize the contents information and the video data Zoki recording apparatus movies characterized by including. Before SL have different properties plurality of recording devices with respect to the rewriting frequency of the index information, TOC information recording area of each of said recording device is composed of a plurality of contents blocks in the recording device, the update count Is counted for each table of contents block ,
Said work memory control means obtains the average access time for each of the plurality of recording devices, the number of updates and obtains the average access time and the repair time required from the product of any repair of all contents block of the recording device total Zoki recording apparatus movies according to claim 1, wherein the write out a long table of contents information of the most repair time required at the time of exceeding the restoration target time to the recording device of the required time. One of the plurality of recording devices is a hard disk drive, the other is a solid state drive having a higher random read speed and a smaller number of rewritable times than the hard disk drive, and the average access time is at least measured read access 3. The video recording apparatus according to claim 2, wherein the frequency of writing to the table of contents information to the solid state drive is determined based on time and less than that of the hard disk drive. A plurality of recording devices that circulate and record video data from any video data supply source, and a work memory that temporarily stores video data received from the video data supply source and sends it to each of the recording devices; Each of the plurality of recording devices has a recording area for circular recording of the video data and a header of the block of the video data to be recorded, which is allocated to each video data supply source in units of a predetermined block. A table of contents information recording area for writing the table of contents information and managing the recording status of the video data in units of the predetermined block, and an address management table for managing the recording address of the video data are formed, and the video data supply source Correspondingly, the video that is cyclically recorded in the recording area formed in any of the plurality of recording devices And over data, a method for use in a video recording apparatus for repairing a mismatch between the exported contents information,
  When starting the video recording device, the table of contents information of the table of contents information recording area is copied to the work memory,
  A buffer is formed in each of the work memories, and the received video data is packed into the buffer within a range that can be accommodated in the predetermined block. The contents information on the work memory is updated with the contents information written from the header, while writing to the contents information recording area is in a standby state, and the contents information in the waiting state is updated. Controlling the writing of the video data and the table of contents information so as to write to the table of contents information in the recording device when a value calculated from the number of times exceeds a reference value determined for each of the plurality of recording devices;
  Check the inconsistency between the table of contents information in the video data in the recording area and the table of contents information in the table of contents information recording area, and if there is a mismatch, the block of the video data of the inconsistent portion from the corresponding recording device If the read video data is normal, the table of contents information in the table of contents information recording area is restored from the table of contents information of the read video data, and the table of contents information and the video data are synchronized.
A video recording method comprising:

Images