JP2012216090A - Data collection device, data collection program, and data collection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform efficient scheduling for data collection in a network.SOLUTION: A data collection device collects data from each node of a network once for each cycle having a predetermined number of clocks. Moreover, the data collection device creates a success list including node information of nodes corresponding to a collection result which could be normally collected and a failure list including node information of nodes corresponding to a collection result which could not be normally collected. Furthermore, the data collection device allocates the node information included in the failure list distributively to each of clocks obtained from the total number of nodes of the network and the number of nodes collectable by a clock unit and corresponding to the number of clocks which is less than the predetermined number of clocks, and creates a collection list for determining the data collection order of the next cycle. The node information included in the success list is also allocated to the collection list.

Description

  The present invention relates to a data collection device, a data collection program, and a data collection method.

  2. Description of the Related Art Conventionally, in a network system having a tree-like hierarchical structure, for example, data is collected from a plurality of base devices that are lower in the hierarchy with respect to the headquarters by a headquarter device at the top of the hierarchical structure. One aspect of such a network system is a video surveillance system.

  In the video monitoring system, for example, video streams received from a plurality of bases are accumulated as moving images, or still pictures are generated from the received video streams to monitor each base. Here, in the video monitoring system, when the video stream is stored as a moving image, it is preferable to use a still image in order to press the storage capacity.

  Also, in a video surveillance system, for example, when a network is not dedicated for transmission of a video stream, a video stream may not be transmitted without securing a bandwidth required for transmission. Further, in the video monitoring system, for example, when transmission of video streams from lower-layer bases overlaps, communication may not be performed due to insufficient bandwidth. In order to deal with these problems, there are cases where the number of function units operating to reduce the number of data received at the headquarters is manually reduced, or the data collection schedule is manually adjusted based on network configuration information.

  In the technology related to scheduling, there are, for example, a method in which a schedule is evaluated while rearranging the processing order of a plurality of devices to obtain a better schedule, and a method in which scheduling is performed based on a determined priority.

Japanese Patent Laid-Open No. 05-189448 JP 2008-90845 A JP 09-251302 A

  However, the conventional technique has a problem that efficient scheduling for data collection in the network cannot be performed. The conventional technique for evaluating a schedule is a schedule that does not reduce the operating rate of a specific device, and is applied to a video surveillance system that collects data from a plurality of bases in a network having a tree-like hierarchical structure. Is difficult. Similarly, it is difficult to apply the conventional technique for scheduling based on priority to a video surveillance system that collects data from a plurality of bases in a network having a tree-like hierarchical structure. Further, when manually adjusting the schedule, it is difficult to perform scheduling if the number of bases is enormous or the network configuration is complicated. Furthermore, in a system that preferably grasps the situation in real time, such as a video monitoring system, if the video stream cannot be transmitted, the meaning of monitoring will be reduced.

  Therefore, the technology disclosed in the present application has been made in view of the above. An object of the technology disclosed in the present application is to provide a data collection apparatus, a data collection program, and a data collection method capable of performing efficient scheduling for data collection in a network.

  In order to solve the above-described problems and achieve the object, a data collection device disclosed in the present application is a data collection device that collects data from a plurality of nodes via a network. A data collection unit that collects data once from each node, a success list that includes node information of nodes corresponding to the collection result that has been successfully collected from the collection result of data collection by the data collection unit, and normal data The first generation unit that generates a failure list including node information of the node corresponding to the collection result that could not be collected, the predetermined number of clocks obtained from the total number of nodes of the network and the number of nodes that can be collected in clock units. Each clock corresponding to a smaller number of clocks includes a node included in the failure list generated by the first generation unit. And a second generation unit for generating a collection list for allocating node information included in the success list generated by the first generation unit and determining a data collection order for the next cycle. .

  One aspect of the data collection device, the data collection program, and the data collection method disclosed in the present application has an effect that efficient scheduling for data collection in a network can be performed.

FIG. 1 is a diagram illustrating a system configuration example of a network including a data collection device. FIG. 2A is a diagram illustrating a format example of an IGMPv2 packet between the headquarters and the bases. FIG. 2B is a diagram illustrating a format example of a PIM-SM packet between sites and between sites. FIG. 3 is a diagram illustrating a configuration example of the data collection apparatus according to the first embodiment. FIG. 4 is a diagram illustrating an example of information held in the collection target list according to the first embodiment. FIG. 5A is a diagram illustrating an example of a collection target list at the beginning of processing by the data collection device. FIG. 5B is a diagram illustrating an example of a collection processing execution result in the first cycle, a success list, and a failure list. FIG. 5C is a diagram illustrating a relationship between the collection processing execution result and the clock in the first cycle. FIG. 5D is a diagram for explaining the recovery process in the first cycle. FIG. 5E is a diagram for explaining the first-cycle recovery process. FIG. 6A is a schematic diagram illustrating generation of a collection target list according to the first embodiment. FIG. 6B is a schematic diagram illustrating generation of a collection target list according to the first embodiment. FIG. 6C is a diagram illustrating generation of the collection target list according to the first embodiment. FIG. 6D is a schematic diagram illustrating generation of a collection target list according to the first embodiment. FIG. 6E is a schematic diagram illustrating generation of a collection target list according to the first embodiment. FIG. 6F is a schematic diagram illustrating generation of a collection target list according to the first embodiment. FIG. 6G is a schematic diagram illustrating generation of a collection target list according to the first embodiment. FIG. 6H is a schematic diagram illustrating generation of a collection target list according to the first embodiment. FIG. 6I is a diagram illustrating the relationship between the collection processing execution result in the second cycle and the clock. FIG. 6J is a diagram illustrating a relationship between the collection processing execution result in the third period and the clock. FIG. 7 is a flowchart illustrating an example of a flow of data collection processing according to the first embodiment. FIG. 8 is a flowchart illustrating an example of the flow of recovery processing according to the first embodiment. FIG. 9 is a flowchart illustrating an example of a flow of a collection target list generation process according to the first embodiment. FIG. 10 is a diagram illustrating an example of the IP address of the camera according to the second embodiment. FIG. 11A is a schematic diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11B is a schematic diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11C is a diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11D is a diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11E is a schematic diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11F is a schematic diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11G is a diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11H is a schematic diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11I is a schematic diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11J is a diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11K is a diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11L is a schematic diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11M is a diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11N is a diagram illustrating generation of a collection target list according to the second embodiment. FIG. 11O is a diagram illustrating the relationship between the collection processing execution result in the second cycle and the clock. FIG. 12A is a flowchart illustrating an example of a flow of a collection target list generation process according to the second embodiment. FIG. 12B is a flowchart illustrating an example of a flow of a collection target list generation process according to the second embodiment. FIG. 13 is a diagram illustrating a computer that executes a data collection program.

  Embodiments of a data collection device, a data collection program, and a data collection method disclosed in the present application will be described below with reference to the accompanying drawings. In addition, this invention is not limited by the following examples. In addition, the embodiments can be appropriately combined within a range that does not contradict the contents.

[System configuration]
A system configuration of a network including a data collection device disclosed in the present application will be described with reference to FIG. FIG. 1 is a diagram illustrating a system configuration example of a network including a data collection device. In the following, a video surveillance system including a data collection device arranged at the headquarters that collects video streams captured by cameras arranged at each site in a network having a tree-like hierarchical structure will be described as an example.

  For example, as shown in FIG. 1, the video surveillance system includes a headquarter in which a data collection device is arranged and a plurality of bases in which an arbitrary number of cameras are arranged in a network having a tree-like hierarchical structure. Each base has a relay device such as a router. In the example of FIG. 1, the camera 01-1 is arranged at the base 01. Similarly, at the base 02, a camera 02-1 and a camera 02-2 are arranged. Similarly, a camera 11-1, a camera 11-2, and a camera 11-3 are arranged at the base 11. Similarly, a camera 12-1 is arranged at the base 12. Similarly, a camera 13-1 and a camera 13-2 are arranged at the base 13. Similarly, a camera 21-1 is arranged at the base 21. Similarly, a camera 22-1 and a camera 22-2 are disposed at the base 22.

  The network in the video surveillance system is realized by an IP (Internet Protocol) network as one aspect. In such a video monitoring system, for example, there may be a restriction on the bandwidth of a line through which a video stream can be transmitted and received between the headquarters and the bases or between the bases and the bases. The bandwidth of the line shown in FIG. 1A can transmit and receive, for example, three video streams. The bandwidth of the lines shown in FIGS. 1B, 1C, and 1G can transmit and receive two video streams, for example. The bandwidths of the lines shown in (D), (E), and (F) of FIG. 1 can transmit and receive, for example, one video stream. Further, the data collection device may not have the configuration information such as the configuration of the line in the network and the line bandwidth.

  In the above-described configuration, the data collection device in the video monitoring system, for example, transmits an imaging request via a network to cameras arranged at each base, and collects the captured video stream. Specifically, the camera constantly transmits the captured video stream to a base such as a router. Each router transmits the video stream of the corresponding camera to the data collection device in response to the imaging request received from the data collection device. Here, for example, IGMP (Internet Group Management Protocol) v2 packets are used between the headquarters and the bases. In addition, for example, PIM-SM (Protocol Independent Multicast-Sparse Mode) packets are used between sites.

  FIG. 2A is a diagram illustrating a format example of an IGMPv2 packet between the headquarters and the bases. For example, the IGMPv2 packet format includes a “Type” field, a “Max Response Time” field, a “Check Sum” field, and a “Group Address” field. And are included. FIG. 2B is a diagram illustrating a format example of a PIM-SM packet between sites and between sites. For example, the PIM-SM packet format includes a “Version” field, a “Type” field, an “Address Area Length” field, a “Checksum” field, and a “Message” field. It is.

  Also, the data collection device generates a still image from the collected video stream, for example. Thereafter, the data collection device that has generated the still image transmits an imaging stop request to the cameras arranged at each base when, for example, the monitoring is ended. As a result, the data collection device in the video monitoring system outputs the generated still image and realizes monitoring of each site. Details of processing by the data collection device will be described later.

[Data collection device configuration]
Next, the configuration of the data collection device will be described with reference to FIG. FIG. 3 is a diagram illustrating a configuration example of the data collection apparatus according to the first embodiment. For example, as illustrated in FIG. 3, the data collection device 100 includes a storage unit 110 and a control unit 120.

  The storage unit 110 stores, for example, data required for various processes by the control unit 120 and various processing results by the control unit 120. In addition, the storage unit 110 includes, for example, a collection target list 111, a success list 112, a failure list 113, and a recovery target list 114. The storage unit 110 is, for example, a semiconductor memory device such as a RAM (Random Access Memory), a ROM (Read Only Memory), and a flash memory, or a storage device such as a hard disk or an optical disk.

  The collection target list 111 is used, for example, for collecting video streams of a certain period, and holds camera information of each camera in the collection order in which the video streams captured by the cameras are collected. The collection target list 111 is a list generated by the collection target list generation unit 125 described later, for example.

  FIG. 4 is a diagram illustrating an example of information held in the collection target list 111 according to the first embodiment. For example, as illustrated in FIG. 4, the collection target list 111 corresponds to “camera ID (identifier)”, “camera name”, “video stream address”, and “execution result” in the order of video stream collection. Keep it attached. For example, the collection target list 111 includes a camera ID “1”, a camera name “camera 01-1”, a video stream address “230.11.1.1:10101”, and an execution result “possible”. Are stored in association with each other. As another example, the collection target list 111 includes a camera ID “2”, a camera name “camera 02-1”, a video stream address “230.11.2.1: 10201”, and an execution result “impossible”. ”In association with each other. Among these pieces of information, the “execution result” is empty before the video stream is collected.

  The success list 112 holds, for example, camera information of each camera corresponding to a collection result in which a video stream can be normally collected from a camera at each site. Specifically, the success list 112 in the video surveillance system holds camera information of the camera that is the transmission source of the video stream when the video stream is successfully collected and the still image is generated. The success list 112 is a list generated by the collection result receiving unit 123 described later, for example.

  The failure list 113 holds, for example, camera information of each camera corresponding to the collection result in which the video stream cannot be normally collected from the camera at each base. Specifically, the failure list 113 in the video surveillance system holds camera information of a camera that is a transmission source of the video stream when the video stream collection or the still image generation fails. The failure list 113 is a list generated by the collection result receiving unit 123 described later, for example.

  For example, the recovery target list 114 holds the camera information of each camera corresponding to the collection result in which the video stream was not normally collected from the camera at each base or the still image could not be generated, as in the failure list 113. The recovery target list 114 is a list generated by the recovery processing unit 124 described later, for example.

  The control unit 120 includes, for example, a control program, a program defining various processing procedures, and an internal memory for storing required data. In addition, the control unit 120 includes, for example, a collection start instruction unit 121, collection units 122a to 122n (n is a natural number), a collection result reception unit 123, a recovery processing unit 124, and a collection target list generation unit 125. Have. The control unit 120 is, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array), or an electronic circuit such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit).

  For example, the collection start instructing unit 121 instructs the collection units 122a to 122n to start collecting video streams every cycle having a predetermined number of clocks in accordance with the collection order held in the collection target list 111. In addition, the collection start instruction unit 121 instructs the collection units 122a to 122n to start collecting the number of simultaneous collections for collecting data at the same time based on a predetermined number of collections. The collection start instruction unit 121 acquires a clock from, for example, a clock generation unit (not shown) that generates a clock.

  For example, the collection units 122a to 122n collect video streams from the camera every cycle having a predetermined number of clocks. Specifically, when the collection units 122a to 122n receive a video stream collection instruction from the collection start instruction unit 121, the collection units 122a to 122n transmit an imaging request to the corresponding camera by multicast. That is, as described above, the collection units 122a to 122n do not have to have the network configuration information in the data collection device 100, and thus transmit an imaging request to the corresponding camera by multicast.

  Then, the collecting units 122a to 122n collect video streams from the cameras, and generate still images from the collected video streams. At this time, when the collection units 122a to 122n are able to normally collect and generate a still image, the collection units 122a to 122n notify the collection result reception unit 123 together with the corresponding camera information that normal processing has been performed. Further, when the collecting units 122a to 122n cannot normally collect or generate a still image, the collecting units 122a to 122n notify the collection result receiving unit 123 that the processing has not been performed normally together with the corresponding camera information. Note that the collection units 122a to 122n also collect a video stream and generate a still image even when a collection instruction is received from the recovery processing unit 124. At this time, the collection units 122a to 122n do not notify the collection result reception unit 123.

  For example, the collection result receiving unit 123 generates the success list 112 and the failure list 113 from the collection results of the video streams notified from the collection units 122a to 122n. Specifically, the collection result receiving unit 123 generates the success list 112 based on the camera information of the camera that is the transmission source of the video stream that has been normally collected and generated by the collection units 122a to 122n. Further, the collection result receiving unit 123 generates the failure list 113 based on the camera information of the camera that is the transmission source of the video stream that has not been normally collected or generated by the collection units 122a to 122n. The collection result receiving unit 123 notifies the recovery processing unit 124 and the collection target list generation unit 125 that the generation is completed after the generation processing of the success list 112 and the failure list 113. In addition, the collection result receiving unit 123 also reflects the collection result held in the collection target list 111.

  For example, the recovery processing unit 124 collects a video stream from a camera that is a transmission source of the video stream that has not been normally collected or generated as a still image within the remaining clock after the collection of the video streams by the collection units 122a to 122n. Run the process again. Specifically, the recovery processing unit 124 receives a generation completion notification of the success list 112 and the failure list 113 from the collection result receiving unit 123. Then, when the list is generated in the failure list 113, the recovery processing unit 124 shifts the generated list to the recovery target list 114. If the list is not generated in the failure list 113, the recovery processing unit 124 ends the process without executing the recovery process.

  Subsequently, the recovery processing unit 124 obtains the number of video streams that can be collected simultaneously in the recovery process. The number of video streams that can be collected simultaneously is, for example, “S ′” for the number of simultaneous collections in the recovery process, “M” for the clock for the collection waiting time, “R” for the remaining number of cameras to be recovered, If the remaining clock number is “Q”, “S ′ = (M × R) ÷ Q” is obtained. When the calculation result of “S ′” is not divisible, the value of the calculation result is rounded up. As for rounding up after the decimal point, for example, “S ′ = ceil ((M × R) ÷ Q)” may be set by using a rounding function “ceil”. In addition, “S ′” is “S ′ ≦ S” where “S” represents the number of simultaneous collections in normal times that are not recovery processing, that is, the number of collection units 122a to 122n used for collecting video streams in normal times. As a condition. In short, since the recovery processing unit 124 may decrease “R” as the clock advances, the recovery processing unit 124 calculates “S ′” for each clock to obtain the minimum simultaneous collection number “S ′” at that time. .

  Subsequently, the recovery processing unit 124 alternately alternates video streams from the cameras corresponding to the camera information existing at the head side and the tail side of the recovery target list 114 for each number “S ′” at which video streams can be collected simultaneously. To collect. When collecting the video stream, the recovery processing unit 124 instructs the collection units 122a to 122n to start collecting the video stream, similarly to the collection start instruction unit 121.

  For example, the collection target list generation unit 125 generates the collection target list 111 based on the success list 112 and the failure list 113 generated by the collection result reception unit 123. Specifically, the collection target list generation unit 125 receives a generation completion notification of the success list 112 and the failure list 113 from the collection result reception unit 123. Then, the collection target list generation unit 125 obtains the minimum number of clocks when the video streams can be normally collected from all the cameras by collecting the video streams once from each camera included in the network. The shortest clock number is obtained by “C = A ÷ S”, for example, where “C” is the shortest clock number, “A” is the number of cameras, and “S” is the number of simultaneous collections. That is, the shortest clock number is a clock number less than one cycle having a predetermined clock number, which is obtained from the total number of nodes of the network and the number of nodes that can be collected in clock units. When the calculation result of “C” is not divisible, the value of the calculation result is rounded up. As for rounding up after the decimal point, for example, the function “ceil” of rounding up may be used to set “C = ceil (A ÷ S)”. The number of simultaneous collections indicates the number of collection units 122a to 122n used for collecting video streams.

  Subsequently, the collection target list generation unit 125 collects the cameras corresponding to the camera information included in the failure list 113 to each of the clocks within the determined minimum clock number and is used for video stream collection in the next period. A list 111 is generated. For distribution within the shortest number of clocks, the collection target list generation unit 125 obtains the number assigned to the collection units 122a to 122n from the success list 112 or the failure list 113 for each clock.

  Here, the number assigned from the failure list 113 is, for example, “L” for the number assigned from the failure list 113, “N” for the number of camera information included in the failure list 113, and “C” for the shortest clock number. L = N ÷ C ”. When the calculation result of “L” is not divisible, the value of the calculation result is rounded up. As for rounding up after the decimal point, for example, “L = ceil (N ÷ C)” may be used by using a rounding function “ceil”. On the other hand, if the number allocated from the success list 112 is, for example, “W”, the number of simultaneous collections is “S”, and the number allocated from the failure list 113 is “L”, then “W = S−L”. Is required. That is, the collection target list generation unit 125 normally collects a video stream from a camera corresponding to the camera information included in the failure list 113 to generate a still image, and suppresses the processing time. The number to be assigned with priority is obtained.

  Thereafter, the collection target list generation unit 125 repeatedly combines the “W” camera information from the success list 112 and the “L” camera information from the failure list 113 until either list becomes empty, A collection target list 111 is generated. Furthermore, when any list becomes empty, the collection target list generation unit 125 moves all the camera information remaining in any list to the collection target list 111, thereby creating a new collection target list 111. Is generated. Note that the collection target list generation unit 125 ends the process without generating the collection target list 111 when the list is not generated in the failure list 113.

[Recovery processing]
Next, the recovery process will be described with reference to FIGS. 5A to 5E. In the following, a video stream collection recovery process for a camera that cannot be normally collected or generated as a still image when a video stream is collected from each camera shown in FIG. 1 will be described. In the following, a case where one cycle is 10 clocks and the collection waiting time is 2 clocks (“M = 2”) will be described. In the following, a case where the number of simultaneous collections “S” is “S = 3” will be described. At this time, it is assumed that the collection units that execute the collection process are the collection units 122a to 122c. In the following description, for the convenience of explanation, only the camera name is shown for information held in the collection target list 111, the success list 112, the failure list 113, and the recovery target list 114.

  FIG. 5A is a diagram illustrating an example of the collection target list 111 at the beginning of processing by the data collection device 100. For example, as illustrated in FIG. 5A, the collection target list 111 at the beginning of the process includes a camera 01-1, a camera 02-1, a camera 02-2, a camera 11-1, a camera 11-2, and a camera 11-3 in this order. Keep information. In addition, the collection target list 111 at the beginning of the process includes the camera 12-1, the camera 13-1, the camera 13-2, the camera 21-1, the camera 22-1, and the camera 22-2 following the camera 11-3. Hold camera information in order. The data collection device 100 executes video stream collection processing from 12 cameras according to the collection target list 111 shown in FIG. 5A.

  FIG. 5B is a diagram illustrating an example of the collection processing execution result in the first cycle, the success list 112, and the failure list 113. FIG. 5C is a diagram illustrating a relationship between the collection processing execution result and the clock in the first cycle. For example, as shown in the left side of FIG. 5B or FIG. 5C, in the collection processing execution in the first cycle, collection of video streams or generation of still images corresponding to five cameras has failed over 6 clocks.

  Specifically, at the first clock, the video stream collection and the still image generation corresponding to the camera 01-1, the camera 02-1, and the camera 02-2 are successful without being affected by the network bandwidth limitation. . At the second clock, although three video streams are passed from the camera 11-1, the camera 11-2, and the camera 11-3 to the base 11, the line bandwidth between the base 01 and the base 11 is equivalent to two video streams. Collection is not possible due to a shortage, and the collection process cannot be terminated. At the 3rd clock, although the collection process for the same camera as the 2nd clock is continued, the status of the line bandwidth does not change, so the collection process for the corresponding camera is completed as the 2 clocks that are the collection waiting time elapse. is doing. That is, at the second to third clocks, video stream collection or still image generation corresponding to the camera 11-1, the camera 11-2, and the camera 11-3 has failed due to the influence of the line bandwidth.

  At the fourth clock, the video stream corresponding to the camera 12-1 is collected and the still image is successfully generated without being affected by the network bandwidth limitation. However, at the fourth clock, although two video streams are passed from the camera 13-1 and the camera 13-2 to the base 13, the line bandwidth between the base 10 and the base 13 is equivalent to one video stream. Cannot be collected, and the collection process cannot be completed. At the fifth clock, the video stream corresponding to the camera 21-1 is collected and the still image is successfully generated without being affected by the network bandwidth limitation. However, at the 5th clock, although the collection process is continued for the camera that could not collect the video stream at the 4th clock, the status of the line bandwidth does not change. The collection process for the camera to be completed has been completed. That is, at the fourth to fifth clocks, collection of video streams or still image generation corresponding to the cameras 13-1 and 13-2 has failed due to the influence of the line bandwidth.

  At the fifth clock, video streams corresponding to the cameras 22-1 and 22-2 are successfully collected and still images are generated without being affected by the network bandwidth limitation. On the left side of FIG. 5B, “◯” indicating “OK” is assigned to the successful camera information and “X” indicating “impossible” is assigned to the failed camera information as the collection processing result of the first cycle. . As a result, in the collection process in the first period, it takes 6 clocks to collect video streams for 5 cameras or to generate still images. Such a situation does not change as long as there is no change in the network bandwidth as long as the collection processing is executed in the order of the collection target list 111 shown in FIG. 5A.

  In addition, as illustrated on the right side of FIG. 5B, the data collection device 100 includes a camera 01-1, a camera 02-1, a camera 02-2, a camera 12-1, a camera 21-1, a camera 22-1, and a camera 22. A success list 112 including the camera information of -2. Similarly, as illustrated on the right side of FIG. 5B, the data collection device 100 includes a failure list including camera information of the camera 11-1, the camera 11-2, the camera 11-3, the camera 13-1, and the camera 13-2. 113 is generated.

  5D and 5E are diagrams for explaining the first-cycle recovery process. Note that FIG. 5D describes the execution of the recovery process at the seventh clock after the collection process ends at the sixth clock (FIG. 5C). FIG. 5E illustrates the execution of the recovery process at the eighth clock after the recovery process at the seventh clock (FIG. 5D).

  For example, the recovery processing unit 124 generates the recovery target list 114 from the camera information included in the failure list 113 as illustrated on the left side of FIG. 5D. Then, the recovery processing unit 124 obtains “S ′ = (2 × 5) ÷ 4”, that is, “S ′ = 3” based on “S ′ = (M × R) ÷ Q”. However, the recovery processing unit 124 rounds up “S ′” so that it cannot be divided, and determines that “S ′ ≦ S” (S = 3 in this example) is satisfied. Here, it is assumed that the collecting units to be collected simultaneously are collecting units 122a to 122c.

  Then, the recovery processing unit 124 instructs the collection unit 122a to collect the video stream from the camera 11-1 corresponding to the camera information held at the top of the recovery target list 114 as indicated by the arrow in FIG. 5D. . In addition, the recovery processing unit 124 instructs the collection unit 122b to collect the video stream from the camera 13-2 corresponding to the camera information held at the end of the recovery target list 114. In addition, the recovery processing unit 124 instructs the collection unit 122c to collect the video stream from the camera 11-2 corresponding to the camera information held second from the top of the recovery target list 114. As a result, at the seventh clock, as shown on the right side of FIG. 5D, video streams corresponding to the camera 11-1, the camera 13-2, and the camera 11-2 are collected without being affected by the network bandwidth limitation. And still image generation has been successful.

  Subsequently, the recovery processing unit 124 obtains “S ′ = (2 × 2) ÷ 3”, that is, “S ′ = 2” based on “S ′ = (M × R) ÷ Q”. However, the recovery processing unit 124 rounds up “S ′” so that it cannot be divided, and determines that “S ′ ≦ S” is satisfied. Here, it is assumed that the collecting units to be collected simultaneously are the collecting unit 122a and the collecting unit 122b.

  Thereafter, the recovery processing unit 124 collects the video stream from the camera 13-1 corresponding to the camera information held second from the end of the recovery target list 114 as indicated by the arrow in FIG. 5E. To instruct. In addition, the recovery processing unit 124 instructs the collection unit 122b to collect the video stream from the camera 11-3 corresponding to the camera information held third from the top of the recovery target list 114. As a result, at the 8th clock, as shown on the left side of FIG. 5E, the video stream collection and still image generation corresponding to the cameras 13-1 and 11-3 are not affected by the network bandwidth limitation. Has succeeded. In short, the recovery processing unit 124 alternately collects from the cameras corresponding to the camera information held at the beginning and the end of the recovery target list 114 so that the collection processing order is not the same as the collection failure at the time of normal collection processing. Execute the process. In this way, the recovery processing unit 124 repeatedly executes the recovery processing as described above until the number of remaining clocks in the corresponding cycle becomes “0” or the unprocessed camera in the recovery processing becomes “0”. To do.

[Generation of Collection Target List According to Embodiment 1]
Next, generation of the collection target list 111 according to the first embodiment will be described with reference to FIGS. 6A to 6J. 6A to 6H are diagrams illustrating generation of the collection target list 111 according to the first embodiment. FIG. 6I is a diagram showing the relationship between the collection processing execution result in the second cycle and the clock. FIG. 6J is a diagram illustrating the relationship between the collection processing execution result in the third period and the clock.

  Hereinafter, a collection target list generation process for generating the collection target list 111 based on the generated success list 112 and failure list 113 after collecting video streams from the respective cameras illustrated in FIG. 1 will be described. In the following, a case where one cycle is 10 clocks and the collection waiting time is 2 clocks (“M = 2”) will be described. In the following, a case where the number of simultaneous collections “S” is “S = 3” will be described. The collection units that execute the collection process at this time are assumed to be the collection units 122a to 122c. In the following description, for the convenience of explanation, only the camera name is shown for information held in the collection target list 111, the success list 112, the failure list 113, and the recovery target list 114.

  For example, as shown on the left side of FIG. 6A, the success list 112 includes the camera 01-1, the camera 02-1, the camera 02-2, the camera 12-1, the camera 21-1, the camera 22-1, and the camera 22-. 2 camera information is held. Similarly, the failure list 113 holds camera information of the camera 11-1, the camera 11-2, the camera 11-3, the camera 13-1, and the camera 13-2.

  Then, the collection target list generation unit 125 calculates “L = 5 × 3 ÷ 12”, that is, “L = 2” based on “L = N ÷ C (C = A ÷ S)”. However, the collection target list generation unit 125 calculates “L” by rounding it up because it cannot be divided. The collection target list generation unit 125 obtains “W = 3-2”, that is, “W = 1” based on “W = S−L”. Accordingly, the collection target list generation unit 125 generates the collection target list 111 for each number of simultaneous collections, “1” from the success list 112 and “2” from the failure list 113.

  Specifically, as illustrated in FIG. 6A, the collection target list generation unit 125 assigns the camera 01-1 to the new collection target list 111 from the camera information included in the success list 112. Further, as illustrated in FIG. 6B, the collection target list generation unit 125 assigns the camera information included in the failure list 113 to the new collection target list 111 as the camera 11-1 and the camera 11-2.

  Then, the collection target list generation unit 125 assigns the camera 02-1 to the new collection target list 111 from the camera information included in the success list 112 as illustrated in FIG. 6C. Further, as illustrated in FIG. 6D, the collection target list generation unit 125 assigns the camera 11-3 and the camera 13-1 to the new collection target list 111 from the camera information included in the failure list 113.

  Subsequently, the collection target list generation unit 125 assigns the camera 02-2 to the new collection target list 111 from the camera information included in the success list 112, as illustrated in FIG. 6E. Further, as illustrated in FIG. 6F, the collection target list generation unit 125 assigns the camera 13-2 from the camera information included in the failure list 113 to the new collection target list 111.

  Here, the collection target list generation unit 125 determines that the failure list 113 is empty. Thereby, as illustrated in FIG. 6G, the collection target list generation unit 125 selects the camera 12-1, the camera 21-1, the camera 22-1, and the camera 22-2 from the remaining camera information included in the success list 112. The new collection target list 111 is assigned. As a result, the collection target list generation unit 125 newly generates the collection target list 111 shown on the right side of FIG. 6G. The new collection target list 111 holds camera information in the order of the camera 01-1, the camera 11-1, the camera 11-2, the camera 02-1, the camera 11-3, and the camera 13-1. In addition, the new collection target list 111 includes the camera 13-2, the camera 02-2, the camera 13-2, the camera 12-1, the camera 21-1, the camera 22-1, and the camera 22-2 in this order. Holds camera information.

  Thereafter, as illustrated in FIG. 6H, the data collection device 100 executes video stream collection processing in the second period from 12 cameras in accordance with the collection target list 111 based on the first period. Here, in the video stream collection processing result in the second period, as shown in FIG. 6I, it takes 5 clocks and the video streams are collected for the three cameras 21-1, 22-1 and 22-2. Or still image generation has failed. That is, since the line bandwidth between the headquarters and the base 02 is two video streams, the data collection device 100 collects video streams corresponding to the camera 21-1, the camera 22-1, and the camera 22-2 due to insufficient bandwidth. Still image generation fails. Also, the sixth to seventh clocks shown in FIG. 6I show the results of executing the recovery process for the three cameras 21-1, the camera 22-1, and the camera 22-2. Here, the details of the recovery process for the camera 21-1, the camera 22-1, and the camera 22-2 are omitted.

  In addition, as illustrated in FIG. 6H, the data collection device 100 includes a camera 01-1, a camera 11-1, a camera 11-2, a camera 02-1, a camera 11-3, a camera 13-1, a camera 02-2, A success list 112 including camera information of the camera 13-2 and the camera 12-1 is generated. In addition, as illustrated in FIG. 6H, the data collection device 100 generates a failure list 113 including camera information of the camera 21-1, the camera 22-1, and the camera 22-2.

  Then, the collection target list generation unit 125 calculates “L = 3 × 3 ÷ 12”, that is, “L = 1” based on “L = N ÷ C (C = A ÷ S)”. However, the collection target list generation unit 125 calculates “L” by rounding it up because it cannot be divided. The collection target list generation unit 125 obtains “W = 3-1”, that is, “W = 2” based on “W = S−L”. Accordingly, the collection target list generation unit 125 generates the collection target list 111 for each number of simultaneous collections by “2” from the success list 112 and “1” from the failure list 113.

  Specifically, as illustrated in FIG. 6H, the collection target list generation unit 125 selects the camera 01-1 and the camera 11-1 from the success list 112, and the camera 21-1 from the failure list 113 based on the second period. Assigned to the collection target list 111 to be collected. Similarly, the collection target list generation unit 125 changes the success list 112 from the camera 11-2 and the camera 02-1 to the failure list 113 from the camera 22-1 to the collection target list 111 based on the second period. assign. Subsequently, the collection target list generation unit 125 similarly collects the camera 11-3 and the camera 13-1 from the success list 112, the camera 22-2 from the failure list 113, and the collection target list 111 based on the second period. Assign to.

  Here, the collection target list generation unit 125 determines that the failure list 113 is empty. As a result, the collection target list generation unit 125 collects the remaining cameras 02-2, 13-2, and 12-1 included in the success list 112 based on the second period as illustrated in FIG. 6H. Assign to the target list 111. As a result, the collection target list generation unit 125 newly generates the collection target list 111 based on the second period. The collection target list 111 based on the second period holds camera information in the order of the camera 01-1, the camera 11-1, the camera 21-1, the camera 11-2, the camera 02-1, and the camera 22-1. In addition, the collection target list 111 based on the second period is the camera 11-1, the camera 11-1, the camera 13-1, the camera 22-2, the camera 02-2, the camera 13-2, and the camera 22-1. The camera information is held in the order of 12-1.

  Thereafter, as illustrated in FIG. 6H, the data collection device 100 executes video stream collection processing in the third period from 12 cameras according to the collection target list 111 based on the second period. Here, in the video stream collection processing result in the third period, as shown in FIG. 6J, it takes 4 clocks, and the video stream collection and the still image generation are successful for all 12 cameras. As a result, the data collection apparatus 100 initially took only 6 clocks and could only collect video streams and generate still images for 5 cameras. However, in the 3rd cycle, the data collection apparatus 100 took 4 clocks and took all 12 cameras. Video stream collection and still image generation are successful. The 4 clocks is the shortest clock number “C = 12 ÷ 3” when the number of cameras is “12” and the simultaneous collection number is “3”.

[Data collection processing flow]
Next, the flow of data collection processing according to the first embodiment will be described with reference to FIG. FIG. 7 is a flowchart illustrating an example of a flow of data collection processing according to the first embodiment.

  For example, as illustrated in FIG. 7, when starting the data collection process (Yes in Step S101), the data collection device 100 executes a normal data collection process according to the collection target list 111 (Step S102). Note that when the data collection device 100 does not start the data collection processing (No at Step S101), the data collection device 100 waits for the data collection processing to start.

  Then, the data collection device 100 determines from the data collection processing result whether or not “N> 0” is satisfied for the number “N” of cameras for which data collection or still image generation has not been performed normally (Step S1). S103). At this time, when “N> 0” (Yes in step S103), the data collection device 100 generates a success list 112 including camera information of cameras that have successfully collected data and generated a still image (step S103). S104). In addition, when “N> 0” (Yes in step S103), the data collection device 100 generates the failure list 113 including the camera information of the cameras that could not normally collect data or generate a still image ( Step S104). Here, when “N = 0” (No in step S103), the data collection device 100 has successfully collected data from all the cameras and generated still images, and thus executes the process of step S107. .

  Subsequently, the data collection device 100 generates the recovery target list 114 from the failure list 113, and executes the recovery process for the cameras corresponding to the camera information included in the recovery target list 114 (step S105). Thereafter, the data collection device 100 executes a collection target list generation process using the success list 112 and the failure list 113 (step S106). In addition, when the data collection process is finished (Yes at Step S107), the data collection device 100 finishes the process, and when the data collection process is not finished (No at Step S107), the data collection device 100 executes the process of Step S102. That is, when the data collection device 100 is in operation in monitoring of a video monitoring system or the like, the data collection device 100 can perform continuous monitoring by repeatedly executing the processing shown in FIG.

[Recovery process flow]
Next, the flow of recovery processing according to the first embodiment will be described with reference to FIG. FIG. 8 is a flowchart illustrating an example of the flow of recovery processing according to the first embodiment. In addition, FIG. 8 demonstrates the flow of a process in step S105 shown in FIG.

  For example, as illustrated in FIG. 8, the recovery processing unit 124 generates the recovery target list 114 from the failure list 113 (step S201). At this time, the remaining number “R” of cameras to be recovered is the same as the number “N” of camera information included in the failure list 113. Then, the recovery processing unit 124 calculates the number of simultaneous collections in the current clock from the number of remaining clocks “Q” in the corresponding period, the number of remaining cameras “R” to be recovered, and the clock “M” of the collection waiting time. “S ′” is calculated (step S202). Note that “S ′” is “S ′ = (M × R) ÷ Q”, and when “S ′” is not divisible, it is obtained by rounding up.

  Subsequently, the recovery processing unit 124 sequentially turns to the collection units 122a to 122n alternately from the top side and the end side of the unprocessed camera information in the recovery target list 114 so that the number of simultaneous collections is “S ′”. The allocation is instructed to start collection (step S203). At this time, the collection units are used by the number of simultaneous collections “S ′”. Thereafter, the collection units 122a to 122n collect a video stream from the corresponding camera, and execute a process of generating a still image from the collected video stream (step S204). Here, the recovery processing unit 124 determines that the camera information of the recovery target list 114 corresponding to the camera for which the collection process has been completed has been processed (step S205).

  Then, when the clock generation unit advances one clock (step S206), the recovery processing unit 124 determines whether or not the remaining clock number “Q” is “Q> 0” (step S207). At this time, if “Q> 0” (Yes at Step S207), the recovery processing unit 124 determines whether or not the remaining number of cameras to be recovered is “R = 0” (Step S208). At this time, if “R = 0” (Yes in step S208), the recovery processing unit 124 ends the process because there is no camera to be recovered. Note that when “Q = 0” (No in step S207), the recovery processing unit 124 ends the process because a predetermined period has elapsed. Further, when “R> 0” (No in step S208), the recovery processing unit 124 executes the process of step S202 on the camera that has not been processed for the recovery process.

[Collection Target List Generation Processing Flow According to Embodiment 1]
Next, the flow of the collection target list generation process according to the first embodiment will be described with reference to FIG. FIG. 9 is a flowchart illustrating an example of a flow of a collection target list generation process according to the first embodiment. In addition, FIG. 9 demonstrates the flow of the process in step S106 shown in FIG.

  For example, as illustrated in FIG. 9, the collection target list generation unit 125 newly generates a framework of the collection target list 111 (step S301). The framework of the collection target list 111 indicates, for example, securing a memory area. At this time, the collection target list generation unit 125 calculates the number “L” allocated from the success list 112 to the collection target list 111 and the number “W” allocated from the failure list 113 to the collection target list 111.

  Then, the collection target list generation unit 125 extracts “W” pieces of camera information from the success list 112 and sequentially adds them to the collection target list 111 (step S302). Subsequently, the collection target list generation unit 125 extracts “L” pieces of camera information from the failure list 113 and sequentially adds them to the collection target list 111 (step S303). Thereafter, the collection target list generation unit 125 determines whether either the success list 112 or the failure list 113 is empty (step S304).

  When the collection target list generation unit 125 determines that one of the lists has become empty (Yes in step S304), the collection target list 111 includes the camera information remaining in the list in the success list 112 or the failure list 113. They are added in order (step S305). On the other hand, if the collection target list generation unit 125 determines that camera information remains in either list (No in step S304), the process in step S302 is performed to add the remaining camera information to the collection target list 111. Execute.

[Effects of Example 1]
As described above, the data collection device 100 generates the success list 112 indicating the node information that has been successfully collected from the nodes included in the network, and the failure list 113 that indicates the node information that has not been successfully collected. . Then, the data collection device 100 generates the collection target list 111 indicating the data collection order while distributing the node information included in the failure list 113 within the shortest number of clocks when data can be collected normally from all nodes. To do. As a result, the data collection device 100 can perform efficient scheduling for data collection even in a network having a tree-like hierarchical structure. In other words, the data collection device 100 can perform efficient scheduling for data collection in a network having a tree-like hierarchical structure without depending on network configuration information, line bandwidth information, or the like.

  In addition, the data collection device 100 generates the collection target list 111 for the purpose of keeping the number of clocks in the case where data can be collected normally from all nodes, so that the data collection process can be executed in a shorter time. Can do. In addition, since the above-described processing is dynamically executed, stable and efficient scheduling can be performed even if the number of data collection targets, the network configuration, and the line bandwidth status change. .

  In the first embodiment, the generation of the collection target list 111 has been described with respect to the case where the number that is unsuccessfully collected in the previous cycle is preferentially obtained and the number assigned to one clock is obtained and distributed within the shortest clock. The generation of the collection target list 111 can be further distributed within the shortest clock by assigning those that are far from each other for each number that can be collected simultaneously in one clock. Therefore, in the second embodiment, a case will be described in which the distances that are far from each other are allocated for each number that can be collected at the same time in one clock, and are distributed within the shortest clock.

[Collection target list according to embodiment 2]
The IP address of the camera according to the second embodiment will be described with reference to FIG. FIG. 10 is a diagram illustrating an example of the IP address of the camera according to the second embodiment. Each camera shown in FIG. 10 corresponds to each camera shown in FIG.

  For example, as shown in FIG. 10, the camera with the camera ID “1” and the camera name “camera 01-1” has an IP address “192.168.1.1” and an address value “0xC0A80101”. The address value is a value obtained by converting each octet into an IP address into a hexadecimal number and considering it as one number.

  Further, the camera with the camera ID “2” and the camera name “Camera 02-1” has an IP address “192.168.2.1” and an address value “0xC0A80201”. The camera with the camera ID “3” and the camera name “camera 02-2” has an IP address “192.168.2.2” and an address value “0xC0A80202”.

  The camera with the camera ID “4” and the camera name “camera 11-1” has an IP address “192.168.11.1” and an address value “0xC0A80B01”. The camera with the camera ID “5” and the camera name “camera 11-2” has an IP address “192.168.11.2” and an address value “0xC0A80B02”. The camera with the camera ID “6” and the camera name “11-3” has an IP address “192.168.11.3” and an address value “0xC0A80B03”.

  The camera with the camera ID “7” and the camera name “camera 12-1” has an IP address “192.168.12.1” and an address value “0xC0A80C01”. Further, the camera with the camera ID “8” and the camera name “camera 13-1” has an IP address “192.168.13.1” and an address value “0xC0A80D01”. The camera with the camera ID “9” and the camera name “camera 13-2” has an IP address of “192.168.13.2” and an address value of “0xC0A80D02”.

  The camera with the camera ID “10” and the camera name “camera 21-1” has an IP address “192.168.21.1” and an address value “0xC0A81501”. Further, the camera with the camera ID “11” and the camera name “camera 22-1” has an IP address “192.168.22.1” and an address value “0xC0A81601”. The camera with the camera ID “12” and the camera name “camera 22-2” has an IP address “192.168.22.2” and an address value “0xC0A81602”.

  As can be seen from FIG. 10, the IP address of each camera is assigned the same or continuous subnet within the same base due to the network configuration. Utilizing this situation, the data collection processing for the cameras existing in the continuous network area is not executed at the same time so that the used lines are distributed and the frequency of occurrence of insufficient bandwidth is further reduced. Hereinafter, generation of the collection target list 111 used for this purpose will be described.

[Generation of Collection Target List According to Embodiment 2]
The generation of the collection target list 111 according to the second embodiment will be described with reference to FIGS. 11A to 11O. 11A to 11N are diagrams illustrating generation of the collection target list 111 according to the second embodiment. FIG. 11O is a diagram showing the relationship between the collection processing execution result in the second cycle and the clock.

  Hereinafter, a collection target list generation process for generating the collection target list 111 based on the generated success list 112 and failure list 113 after collecting video streams from the respective cameras illustrated in FIG. 1 will be described. In the following, a case where one cycle is 10 clocks and the collection waiting time is 2 clocks (“M = 2”) will be described. In the following, a case where the number of simultaneous collections “S” is “S = 3” will be described. The collection units that execute the collection process at this time are assumed to be the collection units 122a to 122c. In the following description, for the convenience of explanation, only the camera name is shown for information held by the collection target list 111, the success list 112, and the failure list 113.

  For example, as shown on the left side of FIG. 11A, the success list 112 includes the camera 01-1, the camera 02-1, the camera 02-2, the camera 12-1, the camera 21-1, the camera 22-1, and the camera 22-. 2 camera information is held. Similarly, the failure list 113 holds camera information of the camera 11-1, the camera 11-2, the camera 11-3, the camera 13-1, and the camera 13-2.

  Then, the collection target list generation unit 125 calculates “L = 5 × 3 ÷ 12”, that is, “L = 2” based on “L = N ÷ C (C = A ÷ S)”. However, the collection target list generation unit 125 calculates “L” by rounding it up because it cannot be divided. The collection target list generation unit 125 obtains “W = 3-2”, that is, “W = 1” based on “W = S−L”. Accordingly, the collection target list generation unit 125 generates the collection target list 111 for each number of simultaneous collections, “1” from the success list 112 and “2” from the failure list 113. Further, at this time, the collection target list generation unit 125 uses the address value of each camera in order to combine the camera information whose distances are far from each other.

  Specifically, the collection target list generation unit 125 sorts the failure list 113 in ascending order of address values using the address values corresponding to the camera information included in the failure list 113. In the example of FIG. 11A, the order is the same as when the failure list 113 was generated. Then, as illustrated in FIG. 11A, the collection target list generation unit 125 moves the camera 01-1 from the camera information included in the success list 112 to the success / failure mixed buffer. Such a mixed success / failure buffer is a buffer that can hold camera information by the number of simultaneous collections, and camera information included in the success list 112 and the failure list 113 is transferred.

  Subsequently, the collection target list generation unit 125 corresponds to the sorted failure list 113 if there is camera information whose address value is smaller than the address value corresponding to the camera 01-1 included in the success / failure mixed buffer. The camera information is transferred to the mixed success / failure buffer. In the example of FIG. 11B, the collection target list generation unit 125 does not have camera information having an address value smaller than that of the camera 01-1 in the failure list 113. Migrate camera information with large difference. That is, as illustrated in FIG. 11B, the collection target list generation unit 125 succeeds / fails the camera 13-2 (address value “0xC0A80D02”) having the largest difference from the address value “0xC0A80101” of the camera 01-1. Move to mixed buffer.

  Thereafter, the collection target list generation unit 125 determines that there is no camera information having an address value smaller than the address value corresponding to the camera 01-1 included in the success / failure mixed buffer in the failure list 113. In addition, the collection target list generation unit 125 determines that there is no camera information having an address value larger than the address value corresponding to the camera 13-2 included in the success / failure mixed buffer in the failure list 113. Therefore, the collection target list generation unit 125 stores the camera information of the camera having the address value having the smallest difference from the intermediate value of each address value corresponding to the camera information included in the success / failure mixture buffer. Migrate to That is, as illustrated in FIG. 11C, the collection target list generation unit 125 performs the camera of the camera 11-1 having the smallest address value with the intermediate value “0xC0A80701” of the address values of the camera 01-1 and the camera 13-2. Information is transferred to a mixed success / failure buffer. At this time, since the success / failure mixture buffer is filled, the collection target list generation unit 125 sorts the camera information included in the success / failure mixture buffer in ascending order of the corresponding address values, as shown in FIG. 11D. To the new collection target list 111.

  Then, the collection target list generation unit 125 calculates “L = 3 × 3 ÷ 12”, that is, “L = 1” based on “L = N ÷ C (C = A ÷ S)”. However, the collection target list generation unit 125 calculates “L” by rounding it up because it cannot be divided. The collection target list generation unit 125 obtains “W = 3-1”, that is, “W = 2” based on “W = S−L”. Accordingly, the collection target list generation unit 125 generates the collection target list 111 for each number of simultaneous collections by “2” from the success list 112 and “1” from the failure list 113.

  Subsequently, as illustrated in FIG. 11E, the collection target list generation unit 125 sorts the cameras 02-1 and 02-2 from the camera information included in the success list 112 in ascending order of the address values and mixes success / failure Move to buffer. Thereafter, the collection target list generation unit 125 determines that there is no camera information having an address value smaller than the address value corresponding to the camera 02-1 included in the success / failure mixed buffer in the failure list 113. In addition, the collection target list generation unit 125 determines that there is no camera information that is intermediate between the address values corresponding to the camera information included in the success / failure mixed buffer. That is, as illustrated in FIG. 11F, the collection target list generation unit 125 obtains camera information of the camera 13-1 having an address value that is larger than the address value corresponding to the camera information 02-2 and has the largest difference. To the mixed success / failure buffer. At this time, since the success / failure mixture buffer is filled, the collection target list generation unit 125 sorts the camera information included in the success / failure mixture buffer in ascending order of the corresponding address values as shown in FIG. 11G. To the new collection target list 111.

  Thereafter, the collection target list generation unit 125 obtains “L = 2 × 3 ÷ 12”, that is, “L = 1” based on “L = N ÷ C (C = A ÷ S)”. However, the collection target list generation unit 125 calculates “L” by rounding it up because it cannot be divided. The collection target list generation unit 125 obtains “W = 3-1”, that is, “W = 2” based on “W = S−L”. Accordingly, the collection target list generation unit 125 generates the collection target list 111 for each number of simultaneous collections by “2” from the success list 112 and “1” from the failure list 113.

  Then, as illustrated in FIG. 11H, the collection target list generation unit 125 sorts the cameras 12-1 and the cameras 21-1 from the camera information included in the success list into the success / failure mixed buffer while sorting the cameras 12-1 in ascending order of address values. Transition. Subsequently, the collection target list generation unit 125 adds the camera information having the largest difference among the camera information having the address value smaller than the address value corresponding to the camera 12-1 included in the success / failure mixed buffer in the failure list 113. Is determined to exist. The corresponding camera is the camera 11-2. In addition, the collection target list generation unit 125 indicates that there is no camera information in the failure list 113 that is intermediate between the address values corresponding to the camera 12-1 and the camera 21-1 included in the success / failure mixed buffer. judge. Further, the collection target list generation unit 125 determines that there is no camera information having an address value larger than the address value corresponding to the camera 21-1 included in the success / failure mixed buffer in the failure list 113. That is, as illustrated in FIG. 11I, the collection target list generation unit 125 moves the camera information of the camera 11-2 included in the failure list 113 to the mixed success / failure buffer. At this time, since the success / failure mixture buffer is filled, the collection target list generation unit 125 sorts the camera information included in the success / failure mixture buffer in ascending order of the corresponding address values, as shown in FIG. 11J. To the new collection target list 111.

  Thereafter, the collection target list generation unit 125 obtains “L = 1 × 3 ÷ 12”, that is, “L = 1” based on “L = N ÷ C (C = A ÷ S)”. However, the collection target list generation unit 125 calculates “L” by rounding it up because it cannot be divided. The collection target list generation unit 125 obtains “W = 3-1”, that is, “W = 2” based on “W = S−L”. Accordingly, the collection target list generation unit 125 generates the collection target list 111 for each number of simultaneous collections by “2” from the success list 112 and “1” from the failure list 113.

  Then, as illustrated in FIG. 11K, the collection target list generation unit 125 sorts the cameras 22-1 and 22-2 from the camera information included in the success list in the ascending order of the address values into the success / failure mixed buffer. Transition. Subsequently, the collection target list generation unit 125 indicates that the camera information having the largest difference exists among the camera information having an address value smaller than that of the camera 22-1 included in the success / failure mixed buffer in the failure list 113. judge. The corresponding camera is the camera 11-3. In addition, the collection target list generation unit 125 indicates that there is no camera information in the failure list 113 that is intermediate between the address values corresponding to the cameras 22-1 and 22-2 included in the success / failure mixed buffer. judge. Further, the collection target list generation unit 125 determines that there is no camera information having an address value larger than the address value corresponding to the camera 22-2 included in the success / failure mixed buffer in the failure list 113. That is, as illustrated in FIG. 11L, the collection target list generation unit 125 moves the camera information of the camera 11-3 included in the failure list 113 to the mixed success / failure buffer. At this time, since the success / failure mixture buffer is filled, the collection target list generation unit 125 sorts the camera information included in the success / failure mixture buffer in ascending order of the corresponding address values as shown in FIG. 11M. To the new collection target list 111. At this time, the collection target list generator 125 uses the collection target list 111 shown on the right side of FIG. 11M for data collection in the next cycle because the success list 112 and the failure list 113 are empty. The target list 111 is assumed.

  As a result, the new collection target list 111 holds camera information in the order of the camera 01-1, the camera 11-1, the camera 13-2, the camera 02-1, the camera 02-2, and the camera 13-1. In addition, the new collection target list 111 includes the camera 13-1, the camera 11-2, the camera 12-1, the camera 21-1, the camera 11-3, the camera 22-1, and the camera 22-2 in this order. Holds camera information.

  Thereafter, as illustrated in FIG. 11N, the data collection device 100 executes video stream collection processing in the second period from 12 cameras in accordance with the collection target list 111 based on the first period. Here, as shown in the right side of FIG. 11N and FIG. 110, the video stream collection result in the second period took 4 clocks and successfully collected the video stream and generated still images for all 12 cameras. is doing.

[Collection Target List Generation Processing Flow According to Second Embodiment]
Next, the collection target list generation process according to the second embodiment will be described with reference to FIGS. 12A and 12B. 12A and 12B are flowcharts illustrating an example of a flow of a collection target list generation process according to the second embodiment.

  For example, as illustrated in FIGS. 12A and 12B, the collection target list generation unit 125 newly generates a framework of the collection target list 111 and the success / failure mixed buffer, and acquires the camera information included in the failure list 113. Sort in ascending order of address values (step S401). The framework of the collection target list 111 and the mixed success / failure buffer indicates, for example, securing a memory area.

  Then, the collection target list generation unit 125 calculates the number “L” allocated to the success / failure mixed buffer from the success list 112 and the number “W” allocated to the success / failure mixed buffer from the failure list 113 (step S402). . “L” and “W” are calculated based on the number “N” of camera information included in the failure list 113.

  Subsequently, the collection target list generation unit 125 moves to the success / failure mixed buffer while sorting “W” items from the success list 112 in ascending order of the address values (step S403). Here, when “W = 0”, the success / failure mixed buffer is empty. Thereafter, the collection target list generation unit 125 determines whether or not the failure list 113 includes camera information having an address value smaller than the address value corresponding to the first camera information in the success / failure mixed buffer (step) S404). In addition, when it is determined that the collection target list generation unit 125 exists, the collection target list generation unit 125 extracts the camera information having the address value having the largest difference from the address value corresponding to the first camera information and the difference between the address values. (Step S404).

  Then, the collection target list generation unit 125 sets a pointer indicating the camera information in the success / failure mixture buffer at the head of the success / failure mixture buffer (step S405). Subsequently, the collection target list generation unit 125 determines whether or not the pointer position of the success / failure mixed buffer is the end (step S406). At this time, if the position of the pointer is not the end (No in step S406), the collection target list generation unit 125 sets the address value of the camera information indicated by the pointer as the minimum address value, and sets the address value of the next camera information as the maximum address. A value is set (step S407). In addition, the collection target list generation unit 125 determines whether or not camera information having an address value within the range exists in the failure list 113 within the range of the minimum / maximum address value (step S407). In addition, the collection target list generation unit 125, when camera information having an address value within the range exists in the failure list 113, collects the camera information having the address value having the smallest difference from the intermediate value of the minimum / maximum address values. Extract and hold (step S407). Further, the collection target list generation unit 125 obtains and holds the difference between the smaller address value of both ends of the range and the address value having the smallest difference between the intermediate values (step S407). Thereafter, the collection target list generation unit 125 advances the pointer of the mixed success / failure buffer by one (step S408).

  In addition, when the pointer position is the end (Yes in step S406), the collection target list generation unit 125 determines whether camera information having an address value larger than the address value corresponding to the end camera information exists in the failure list 113. It is determined whether or not (step S409). In addition, the collection target list generation unit 125 extracts the camera information in the failure list 113 having an address value having the largest difference from the address value corresponding to the last camera information when the collection target list generation unit 125 exists in the failure list 113. Hold (step S409). Furthermore, the collection target list generation unit 125 calculates and holds the difference between the address value corresponding to the last camera information and the address value having the largest difference (step S409).

  After that, the collection target list generation unit 125 succeeds / fails the camera information of the failure list 113 held in S404 to S409, one camera information at a time in descending order of the obtained difference value, and up to “L” pieces of camera information. The process proceeds to the mixing buffer (step S410). Then, the collection target list generation unit 125 determines whether or not there is a vacancy in the success / failure mixed buffer (step S411). At this time, when there is a vacancy (Yes at Step S411), the collection target list generation unit 125 determines whether or not all the camera information of the held failure list 113 has been transferred to the success / failure mixed buffer (Step S412). ). If the collection target list generation unit 125 determines that all of the success / failure mixing buffers have been transferred (Yes in step S412), the camera information included in the success / failure mixing buffer is sorted in ascending order of address values. (Step S413). The collection target list generation unit 125 executes the process of step S404 after the process of step S413. The collection target list generation unit 125 executes the process of step S410 when it is determined that all of the success / failure mixed buffers have not been transferred (No in step S412).

  Further, when there is no space (No in step S411), the collection target list generation unit 125 sorts the camera information included in the success / failure mixed buffer in ascending order of address values, and moves to the end of the collection target list 111 ( Step S414). Then, the collection target list generation unit 125 determines whether either the success list 112 or the failure list 113 is empty (step S415). At this time, when the collection target list generation unit 125 determines that one of them is empty (Yes in step S415), the collection target list 111 acquires camera information of the remaining list of either the success list 112 or the failure list 113. (Step S416). Here, when the collection target list generation unit 125 determines that neither the success list 112 nor the failure list 113 is empty (No in step S415), the collection target list generation unit 125 executes the process of step S402.

  As described above, in the processing example according to the second embodiment, data can be normally collected from the nodes included in the network at a cycle shorter than the data collection processing result according to the first embodiment illustrated in FIG. 6J.

[Effects of Example 2]
As described above, since the data collection device 100 generates the collection target list 111 by combining nodes that are far from each other in the network, it is possible to distribute the network lines used during data collection. . In other words, the data collection device 100 executes the data collection process according to the collection target list 111 in which the lines of the network are distributed, so that more efficient data collection scheduling can be performed.

  Although the embodiments of the transmission device 100 disclosed in the present application have been described so far, the present invention may be implemented in various different forms other than the embodiments described above. Therefore, different embodiments in (1) recovery processing, (2) data collection, (3) configuration, and (4) program will be described.

(1) Recovery process In the above-described embodiment, the case where the recovery process is executed within the remaining clock in one cycle has been described. However, the recovery process may not be executed. That is, the recovery process is executed after data collection from the camera corresponding to the camera information included in the collection target list 111, and performs recovery for the camera that could not be normally collected by data collection. Does not affect the collection process. In other words, the recovery process is a process executed to collect data that could not be collected by data collection using the collection target list 111 within the same period. For example, in the video monitoring system, the recovery process is performed within the same period. Optional processing to increase the number. For this reason, as described above, in the recovery process, when one cycle elapses, no further recovery is performed even if data collection by the recovery process cannot be normally performed at that time. In the above embodiment, the case has been described in which node information is alternately acquired from the head side and the tail side of the failure list and data is collected. However, if the order of acquiring node information from the failure list is arbitrarily determined good.

(2) Data collection In the second embodiment, the case where the address value of the IP address is used is described as one mode of the process of collecting data from nodes that are far from each other. However, the address value is not used. May be. For example, the data collection device 100 stores an actual distance between nodes in advance, and executes a process related to data collection from nodes far from each other using each distance.

(3) Configuration In addition, for information including the processing procedure, control procedure, specific name, various data, parameters, etc. shown in the above document or drawing (for example, information held in the collection target list 111) It can be changed arbitrarily unless otherwise specified.

  Each component of the illustrated data collection device 100 is functionally conceptual and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various burdens or usage conditions. Can be integrated. For example, since the collection units 122a to 122n have exemplified the video monitoring system in the above-described embodiment, the collection units 122a to 122n have been described as executing up to the still image generation process. That is, the collection units 122a to 122n may only execute a process of collecting data from a predetermined node included in the network. In other words, the data collection device 100 is not only applied to the video surveillance system, but may be applied to anything that collects arbitrary data.

(4) Program The various processes of the data collection apparatus described in the above embodiments can be realized by executing a program prepared in advance on a computer system such as a personal computer or a workstation. In the following, an example of a computer that executes a data collection program having the same function as the data collection apparatus described in the above embodiment will be described with reference to FIG. FIG. 13 is a diagram illustrating a computer that executes a data collection program.

  As illustrated in FIG. 13, the computer 1000 includes a CPU (Central Processing Unit) 1100, a ROM (Read Only Memory) 1200, an HDD (Hard Disk Drive) 1300, and a RAM (Random Access Memory) 1400. These units 1000 to 1400 are connected via a bus 1500.

  The ROM 1200 includes a data collection program that exhibits the same functions as the collection start instruction unit 121, the collection units 122a to 122n, the collection result reception unit 123, and the collection target list generation unit 125 described in the first embodiment. Stored in advance. That is, the ROM 1200 stores a data collection program 1200a as shown in FIG. Note that the data collection program 1200a may be separated as appropriate. Then, the CPU 1100 reads the data collection program 1200a from the ROM 1200 and executes it. The HDD 1300 is provided with a success list 1300a and a failure list 1300b. The success list 1300a corresponds to the success list 112 shown in FIG. The failure list 1300b corresponds to the failure list 113 shown in FIG.

  Then, the CPU 1100 reads out the success list 1300a and the failure list 1300b and stores them in the RAM 1400. Further, the CPU 1100 executes a data collection program using the success list data 1400a and the failure list data 1400b stored in the RAM 1400. Note that all the data stored in the RAM 1400 may not always be stored in the RAM 1400, and only the data required for processing may be stored in the RAM 1400. Note that the data collection program does not necessarily have to be stored in the ROM 1200 from the beginning.

  For example, the program is stored in a “portable physical medium” such as a flexible disk (FD), a CD-ROM, a DVD disk, a magneto-optical disk, or an IC card inserted into the computer 1000. Then, the computer 1000 may read and execute the program from these. Further, the program is stored in “another computer (or server)” connected to the computer 1000 via a public line, the Internet, a LAN, a WAN, or the like. Then, the computer 1000 may read and execute the program from these.

DESCRIPTION OF SYMBOLS 100 Data collection device 110 Storage part 111 Collection target list 112 Success list 113 Failure list 114 Recovery target list 120 Control part 121 Collection start instruction part 122a-122n Collection part 123 Collection result reception part 124 Recovery process part 125 Collection target list production | generation part

Claims (7)

In a data collection device that collects data from a plurality of nodes via a network,
A data collection unit that collects data from each node once for each cycle having a predetermined number of clocks;
From the collection result of the data collection by the data collection unit, the success list including the node information of the node corresponding to the collection result that has been successfully collected, and the node information of the node that corresponds to the collection result that has not been successfully collected. A first generation unit for generating a failure list including:
Node information included in the failure list generated by the first generation unit in each clock corresponding to the number of clocks smaller than the predetermined number of clocks obtained from the total number of nodes of the network and the number of nodes that can be collected in clock units. And a second generation unit that allocates node information included in the success list generated by the first generation unit and generates a collection list that determines the data collection order of the next cycle. Characteristic data collection device.   The second generation unit is a result of dividing the number of node information included in the failure list by the number of clocks smaller than the predetermined number of clocks obtained from the total number of nodes of the network and the number of nodes that can be collected in clock units. The division result value is set as the number of node information to be allocated in clock units from the failure list, and the subtraction result value obtained by subtracting the division result value from the number of nodes that can be collected in clock units is set in clock units from the success list. The data collection apparatus according to claim 1, wherein the collection list is generated as the number of node information to be allocated. The network is a network having a tree-like hierarchical structure,
The data collection apparatus according to claim 2, wherein the second generation unit generates the collection list by combining nodes that are distant from each other and assigning them to the same clock.   The data re-collection unit further collects data from a node corresponding to the node information included in the failure list within a remaining clock after data collection by the data collection unit within each period. The data collection device according to any one of 1 to 3.   The data recollection unit alternately acquires node information held at the head side and the tail side of the failure list for each number of nodes that can be collected in the clock unit, and from the node corresponding to the acquired node information The data collection apparatus according to claim 4, wherein data collection is performed again. In a data collection program that collects data from multiple nodes over a network,
Collect data once from each node once per cycle with a given number of clocks,
From the collection results of data collection, a success list including node information of nodes corresponding to the collection results that were successfully collected and a failure list including node information of nodes corresponding to the collection results that were not successfully collected. Generate
The node information included in the failure list is distributed and allocated to each clock corresponding to the number of clocks smaller than the predetermined number of clocks, which is obtained from the total number of nodes of the network and the number of nodes that can be collected in units of clocks, and A data collection program that causes a computer to execute a process of generating a collection list that assigns node information included in a success list and determines a data collection order of the next period. In a data collection method for collecting data from a plurality of nodes via a network, which is executed by a computer,
Collect data once from each node once per cycle with a given number of clocks,
From the collection results of data collection, a success list including node information of nodes corresponding to the collection results that were successfully collected and a failure list including node information of nodes corresponding to the collection results that were not successfully collected. Generate
The node information included in the failure list is distributed and allocated to each clock corresponding to the number of clocks smaller than the predetermined number of clocks, which is obtained from the total number of nodes of the network and the number of nodes that can be collected in units of clocks, and A data collection method characterized by generating a collection list that assigns node information included in a success list and determines a data collection order of a next period.

Images