JP6747704B1 - Allocation control device, allocation control system, and allocation control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an allocation control device, an allocation control system, and an allocation control method for reducing the load on a processing unit and smoothly utilizing imaging data for various purposes. An allocation control device includes an allocation processing unit connected to a camera, and a plurality of specialized processing units connected to the allocation processing unit and corresponding to any of a plurality of preset types. The allocation processing unit is connected to the camera and has a specification processing unit that specifies the type of the main object included in the image pickup data input from the camera as a type and outputs the image pickup data together with the type data indicating the specified type. .. Further, the allocation processing unit has an allocation unit that outputs the imaging data output together with the type data to the specialized processing unit corresponding to the type indicated by the type data output from the specific processing unit. [Selection diagram] Fig. 13

Description

The present invention relates to an allocation control device, an allocation control system, and an allocation control method.

2. Description of the Related Art Conventionally, there is known a system that uses image data captured by a camera for monitoring or the like (for example, see Patent Document 1). The security system of patent document 1 switches the camera operated according to the position change of a moving object, and utilizes the image pick-up data by one camera for a display on a monitor. However, the image data captured by the camera includes various information according to the time zone and season in addition to the installation environment of the camera, and is used for various purposes other than display as well as display for various purposes. be able to.

JP-A-6-44470

However, if the usage is changed according to the information included in the imaged data, the load on the processing unit such as the processor becomes very large. In addition, recently, the amount of data to be communicated is increasing, including the increase in the amount of image data captured by cameras, and the load on the processing unit is also increasing accordingly. When 5G (fifth generation mobile communication system) becomes widespread in the future, communication of a large amount of data will be performed more than ever, and the processing load on the processing unit will be significantly increased. That is, if it is attempted to switch the usage of the imaged data according to the included information with the conventional configuration, the load on the processing unit increases, and thus there is a problem that the processing cannot proceed smoothly. Furthermore, as 6G (sixth generation mobile communication system), 7G (seventh generation mobile communication system), and next-generation mobile communication system are sequentially realized, it becomes difficult to perform various processes with the conventional configuration. Expected to be.

The present invention has been made to solve the above problems, and provides an allocation control device and an allocation control system that reduce the load on a processing unit and smoothly utilize imaging data for various purposes. To aim.

An allocation control device according to an aspect of the present invention includes an allocation processing unit connected to a camera, a plurality of specialized processing units connected to the allocation processing unit and corresponding to any of a plurality of preset types. , And the allocation processing unit is connected to the camera, specifies the type of the main object included in the image pickup data input from the camera as a type, and outputs the image pickup data together with the type data indicating the specified type. The specific processing unit and the allocation unit that outputs the imaging data output together with the type data to the specialized processing unit corresponding to the type indicated by the type data output from the specific processing unit.

An allocation control device according to an aspect of the present invention includes an allocation processing unit connected to a camera, a plurality of specialized processing units connected to the allocation processing unit and corresponding to any of a plurality of preset types. , And the allocation processing unit is connected to the camera, and specifies the main age group of the person included in the imaging data input from the camera as a type, and outputs the imaging data together with the type data indicating the specified type. The specific processing unit and the special processing unit corresponding to the type indicated by the type data output from the specific processing unit, and the allocation unit that outputs the imaging data output together with the type data.

Further, an allocation control device according to an aspect of the present invention is the allocation control device described above, wherein the specialized processing unit, when the imaging data is output from the allocation unit, the imaging data is a specific processing unit of an output source. It is determined which of the plurality of attributes in the type specified in 1 is specified, and the processing corresponding to the determined attribute is executed.
An allocation control system according to one aspect of the present invention includes the allocation control device and a display device, and the specialized processing unit causes the display device to display a moving image or a still image according to the determined attribute. is there.

In the allocation control method according to one aspect of the present invention, the identification processing unit connected to the camera identifies the type of the main object included in the image capture data input from the camera as the type of the image capture data, and identifies the identified type. The imaging data is output together with the type data shown, and the allocation unit connected to the plurality of specialized processing units corresponding to any of the plurality of types indicates the type data output from each of the plurality of specialized processing units. The imaging data output together with the type data is output to the specialized processing unit corresponding to the type.

According to the present invention, the plurality of specific processing units connected to the camera identify the type of the imaged data, and the allocation unit outputs the imaged data to the specialized processing unit corresponding to the type. The load can be distributed to different parts. Therefore, it is possible to reduce the load on each processing unit and smoothly utilize the imaged data for various purposes.

It is a block diagram which illustrated the composition of the allocation control device and the allocation control system concerning Embodiment 1 of the present invention. It is a block diagram which shows the functional structural example of the special processing part of FIG. It is a block diagram which shows the application example of the allocation control system of FIG. It is an external view which illustrates the flying body which mounts the allocation control system of FIG. It is a block diagram which shows the example which mounted the allocation control system of FIG. 1 in the flying body. It is explanatory drawing which shows the example which applied the allocation control system of FIG. 1 to facilities, such as a stadium. It is a block diagram which matched the allocation control system of FIG. 1 with the situation of FIG. 8 is a block diagram showing a configuration example in which the allocation control device of FIG. 1 is applied to the selection unit of FIG. 7. It is explanatory drawing which shows the example which applied the allocation control system of FIG. 1 to the street. It is a block diagram which matched the allocation control system of FIG. 1 with the situation of FIG. 3 is a flowchart illustrating an operation of the allocation control device of FIG. 1. It is a block diagram which illustrated the composition of the allocation control device and the allocation control system concerning the modification 1-1 of Embodiment 1 of the present invention. It is a block diagram which illustrated the composition of the allocation control device and the allocation control system concerning the modification 1-2 of Embodiment 1 of the present invention. It is explanatory drawing which shows the example which applied the allocation control system of FIG. 13 to the intersection etc. in a town. It is a block diagram which matched the allocation control system of FIG. 13 with the situation of FIG. 14 is a flowchart showing an example of an operation performed by a specialized processing unit of the allocation control device of FIG. 13. 14 is a flowchart showing another example of the operation by the specialized processing unit of the allocation control device of FIG. 13. It is a block diagram which illustrated the composition of the allocation control device and the allocation control system concerning Embodiment 2 of the present invention. FIG. 19 is a block diagram showing an application example of the allocation control system of FIG. 18. 19 is a flowchart illustrating an operation of the allocation control device of FIG. 18. It is a block diagram which illustrated the composition of the allocation control device and the allocation control system concerning the modification 2-1 of Embodiment 2 of the present invention. It is a block diagram which illustrated the composition of the allocation control device and the allocation control system concerning the modification 2-2 of Embodiment 2 of the present invention. It is a block diagram which illustrated the composition of the allocation control system concerning Embodiment 3 of the present invention.

Embodiment 1.
The configurations of the allocation control device and the allocation control system according to the first embodiment will be described with reference to FIG. The allocation control system 100 includes a plurality of cameras A1 to An (n is a natural number of 2 or more), the allocation control device 10, and an actuator unit 90. The actuator unit 90 is composed of a plurality of actuators X1 to Xk (k is a natural number of 2 or more) to be controlled by the allocation control device 10. When a plurality of cameras A1 to An are referred to without distinction, they are referred to as "camera A". When the plurality of actuators X1 to Xk are referred to without distinction, they are referred to as “actuator X”.

The camera A has an image sensor such as a CCD (Charge Coupled Devices) or a CMOS (Complementary Metal Oxide Semiconductor), and captures a moving image or a still image. The camera A outputs the captured moving image or still image to the allocation control device 10 as captured image data. The camera A of the first embodiment outputs the identification information indicating itself to the allocation control device 10.

The actuator X is a device that operates based on an electric signal output from the allocation control device 10. As the actuator X, for example, a light emitting unit including a light source such as an LED (light emitting diode), a notification unit including a speaker, or a driving unit such as a motor is assumed.

The allocation control device 10 includes an allocation processing unit 20 connected to a plurality of cameras A1 to An, and a plurality of specialized processing units C1 to Cm (m is 2 or more) corresponding to any of a plurality of preset types. Natural number) and. The plurality of specialized processing units C1 to Cm are connected to the allocation processing unit 20. When a plurality of specialized processing units C1 to Cm are referred to without distinction, they are referred to as “specialized processing unit C”.

The allocation processing unit 20 has a plurality of specific processing units B1 to Bn connected to each of the plurality of cameras A1 to An on a one-to-one basis. Each of the plurality of specification processing units B1 to Bn specifies which of the plurality of types the imaging data input from the camera A of the connection destination is, and generates type data indicating the specified type. Then, each of the plurality of identification processing units B1 to Bn outputs the type data together with the imaging data relating to the identification of the type to the allocation processing unit 20. When a plurality of specific processing units B1 to Bn are referred to without distinction, they are referred to as "specific processing unit B".

The identification processing unit B is equipped with AI (artificial intelligence) based on learning such as deep learning, and identifies the type of imaging data by estimating what the main object included in the imaging data is. The specific processing unit B has a communication unit, a storage unit, and a control unit, which are not shown, like the special processing unit C described later. The control unit includes, for example, a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). Since the processing amount of the specific processing unit B is smaller than that of the special processing unit C, the specific processing unit B can be configured by a member having lower performance than the special processing unit C.

Further, the allocation processing unit 20 outputs the imaging data output together with the type data to the specialized processing unit C corresponding to the type indicated by the type data output from each of the plurality of specific processing units B1 to Bn. Has 30. The allocation unit 30 has m routes (options) for allocating the data input from one specific processing unit B to any of the plurality of specialized processing units C1 to Cm, as shown by a broken line in FIG. ing. That is, the allocation unit 30 has n×m routes (options) that allocate the data input from the plurality of specific processing units B1 to Bn to any of the plurality of specialized processing units C1 to Cm. The allocation unit 30 is configured by a programmable logic device such as FPGA (field-programmable gate array). The allocation program, which is the logic program of the allocation unit 30, can be appropriately changed according to the use case.

Here, the type is the type of the main object included in the imaging data, and the main object includes an inanimate object such as an electric wire or a utility pole and a moving object such as an animal, a vehicle, or a flying object. .. For example, the type may be set in the dimensions of “person” and “dog”. Further, for example, the type may be set for each person based on a concept such as “age group” or “movement state”. Furthermore, for each type, a plurality of attributes that further subdivide the type are set in advance. For example, attributes such as pigeons, crows, and bats are set for the bird type.

The specialized processing section C performs image analysis specialized for any of a plurality of types. More specifically, the specialized processing unit C is equipped with an AI based on learning such as deep learning, and estimates which attribute in the type the specialized object is, the main object included in the imaging data.

That is, when the imaging data is output from the allocating unit 30, the specialized processing unit C determines which of the plurality of attributes in the type identified by the identification processing unit B that is the output source, the imaging data. To do. Then, the specialized processing unit C executes processing according to the result of the determination, that is, processing according to the determined attribute. The content of the processing performed by the specialized processing unit C according to the attribute is preset in association with the attribute in the operation program of the specialized processing unit C, and can be changed as appropriate. The specialized processing unit C outputs, for example, a control signal according to the determined attribute to one or more actuators X.

Here, a specific configuration example of the specialized processing unit C will be described with reference to FIG. As shown in FIG. 2, the specialized processing unit C includes a first communication unit 41, a second communication unit 42, a storage unit 43, and a control unit 44. The first communication unit 41 is an input interface that receives data output from the allocation unit 30 and transfers the data to the control unit 44. The second communication unit 42 is an interface for the control unit 44 to communicate with the actuator unit 90, a device connected to the network N, or a device connected wirelessly.

The storage unit 43 has a memory unit 43a and a storage unit 43b. The memory unit 43a is a volatile storage device configured by, for example, a RAM (Random Access Memory) and temporarily storing data. The storage unit 43b is configured by a flash memory, an eMMC (embedded Multi Media Card), an SSD (Solid State Drive), or the like. In the example of FIG. 2, a learning program 431, an analysis processing program 432, and an update program 433 are illustrated as the operation programs of the specialized processing unit C. The analysis processing program 432 is a program for executing processing relating to attribute determination.

The control unit 44 has a learning processing unit 44a, an analysis processing unit 44b, and an update processing unit 44c. In other words, the control unit 44 functions as the learning processing unit 44a by reading the learning program 431, functions as the analysis processing unit 44b by reading the analysis processing program 432, and updates processing unit by reading the update program 433. It functions as 44c. The control unit 44 is configured by a SoC (System-on-a-chip) including, for example, a CPU or a GPU.

The learning processing unit 44a generates the analysis processing program 432 by learning such as deep learning using the learning data. The analysis processing unit 44b determines which of the plurality of attributes in the type identified by the identification processing unit B, the imaging data output from the allocation unit 30 and executes processing according to the determined attribute. To do. The analysis processing unit 44b outputs the imaging data input from the allocation unit 30 and the analysis information including the determination result to the update processing device 80 via the second communication unit 42. In the first embodiment, the analysis processing unit 44b is configured to store the image pickup data and the analysis information input from the allocation unit 30 in the storage unit 43b.

The update processing unit 44c updates the analysis processing program 432 by learning based on the imaging data and the analysis information accumulated in the storage unit 43b. Further, the update processing means 44c updates the analysis processing program 432 in the storage means 43b by the update analysis processing program 432 transmitted from the update processing device 80. However, the update processing unit 44c may perform any of the update processing of the analysis processing program 432 based on its own learning and the update processing of the analysis processing program 432 in cooperation with the update processing device 80. ..

In the first embodiment, the plurality of specialized processing units C1 to Cm are connected to the update processing device 80 via a network N such as the Internet. The update processing device 80 is, for example, a cloud server based on cloud computing. Each of the plurality of specialized processing units C1 to Cm is configured to transmit the imaging data and the analysis information input from the allocation unit 30 to the update processing device 80.

The update processing device 80 stores the analysis processing program 432 of each specialized processing section C. The update processing device 80 has a function of updating the analysis processing program 432 by learning based on the imaging data and the analysis information transmitted from the specialized processing unit C. The update processing device 80 updates the analysis processing program 432 at every predetermined time or when the accumulated amount of data reaches a certain amount. Then, the update processing device 80 transmits the updated analysis processing program 432 to the specialized processing unit C at a preset timing. As a result, each specialized processing unit C can update the analysis processing program 432. However, the update processing device 80 is not limited to the cloud server and may be a physical server such as a Web server. The update processing device 80 may be a desktop PC (Personal Computer), a notebook PC, a tablet PC, or the like. The specialized processing unit C and the update processing device 80 may perform wired communication, wireless communication, or a combination of wired and wireless communication.

That is, the update processing device 80 updates the respective analysis processing programs 432 of the plurality of specialized processing units C based on the imaging data and the analysis information transmitted from the plurality of specialized processing units C. Then, the update processing device 80 transmits the updated analysis processing program 432 to each specialized processing unit C. Therefore, when the updated analysis processing program 432 is transmitted from the update processing device 80, the specialized processing section C can update the analysis processing program 432 in the storage section 43 by the analysis processing program 432.

Next, an application example of the allocation control system 100 will be described with reference to FIG. Here, an example in which the allocation control system 100 is incorporated in the monitoring and management of a rice field or a field will be described. In FIG. 3, the allocation control system 100 has four cameras A, the allocation control device 10 has four specific processing units B and five specialized processing units C, and the actuator unit 90 has five actuators X. An example is shown.

The specialized processing units C1 and C2 correspond to the type “bird”, the specialized processing units C3 and C4 correspond to the type “terrestrial animal”, and the specialized processing unit C5 corresponds to the type “person”. It shall be. The actuator X1 is a light emitting unit that emits light, the actuators X2 and X3 are notification units that generate a sound that animals dislike, and the actuator X4 is a sound wave generation unit that generates an ultrasonic wave. The actuator X5 is assumed to be a notification means for generating a sound or voice for warning.

In FIG. 3, the specific processing unit B1 sets the type of the imaging data from the camera A1 to “terrestrial animals”, the specific processing unit B2 sets the type of the imaging data from the camera A2 to “birds”, and the specific processing unit B3 to the camera. An example is shown in which the type of imaging data from A3 is "bird" and the type of the imaging data from the camera A4 is "person" by the specific processing unit B4.

In the case of this example, the specialized processing unit C1 determines the attribute of the type “bird” of the imaged data input from the allocation unit 30. As an example, the specialized processing unit C1 outputs a control signal for instructing the actuator X1 to emit light when it is determined that the attribute of the imaging data is “dove”. When the specialized processing unit C1 determines that the attribute of the imaging data is “crow”, the specialized processing unit C1 outputs a control signal instructing at least one of the actuators X2 and X3 to generate a warning sound. At that time, the specialized processing unit C1 may grasp the position from the identification information of the camera A2 and output the control signal to one of the actuators X2 and X3 that is closer to the camera A2.

The specialized processing unit C3 determines the attribute of the type “terrestrial animal” of the imaging data input from the allocation unit 30. As an example, when the specialized processing unit C3 determines that the attribute of the captured image data is “raccoon dog”, the specialized processing unit C3 outputs a control signal for instructing the actuator X2 or X3 to call a crying wolf. Similarly, when the specialized processing unit C3 determines that the attribute of the imaged data is “Hakubishin”, it outputs a control signal instructing the generation of ultrasonic waves to the actuator X4.

The specialized processing unit C5 determines the attribute of the type “person” of the imaging data input from the allocation unit 30. As an example, when the specialized processing unit C3 determines that the attribute of the imaged data is “suspicious person”, the specialized processing unit C3 outputs a control signal for instructing output of a warning sound or voice to the actuator X5. The processing performed by the specialized processing units C2 and C4 is similar to the above.

However, the processing by the specialized processing unit C is an example, and does not indicate an optimum example. The actuator unit 90 may have an actuator X other than the above, such as an odor generating unit that generates a specific odor and a blowing unit that blows air on an object. That is, the processing by the specialized processing unit C and the actuator X can be appropriately changed. In addition, the specialized processing unit C may output a control signal to the plurality of actuators X, such as outputting a control signal to both the actuator X1 and the actuator X2, to simultaneously operate the plurality of actuators X.

Further, the allocation control system 100 may cause the update processing device 80 to learn the reaction of the moving object with respect to the operation of the actuator X and the like from the imaging data or the analysis information of each specialized processing unit C. Then, the allocation control system 100 may upload the analysis processing program 432 of the specialized processing unit C using the update processing device 80 so that the specialized processing unit C appropriately selects the optimal repulsion method.

Subsequently, an example in which the allocation control system 100 is applied to an air vehicle will be described with reference to FIGS. 4 and 5. 4 and 5, a drone having six propellers P mounted on a main body 510 is illustrated as an example of the flying object 500. The allocation control system 100 has four cameras A, the allocation control device 10 has four specific processing units B and three specialized processing units C, and the actuator unit 90 has nine actuators X. ing.

The specialized processing unit C1 corresponds to the type “bird”, the specialized processing unit C2 corresponds to the type “person”, and the specialized processing unit C3 corresponds to the type “obstacle”. The actuators Xa1 to Xa4 are light emitting means for emitting light, the actuators Xb1 to Xb4 are notification means for generating sound or voice, and the actuator X5 is a driving means for controlling the operation of the motor M. Here, for convenience, it is assumed that the actuators Xb1 to Xb4 have a function of generating ultrasonic waves. In FIG. 5, the power supply device and various sensors are omitted.

In the example of FIG. 5, the specialized processing unit C1 is connected to the actuators Xa1 to Xa4, the actuators Xb1 to Xb4, and the actuator X5. The specialized processing unit C2 is connected to the actuators Xb1 to Xb4 and the actuator X5. The specialized processing unit C3 is connected to the actuator X5.

It is assumed that the specific processing unit B1 sets the type of the image pickup data from the camera A1 to “bird” and outputs the type data indicating the type “bird” to the allocation unit 30. Then, the allocation unit 30 outputs the imaging data to the specialized processing unit C1 based on the type data from the specific processing unit B1. The specialized processing unit C1 determines whether the attribute of the imaging data is pigeon, crow, bat, or the like. For example, when the specialized processing unit C1 determines that the attribute of the imaging data is “bat”, it selects the actuator Xb1 that is the notification means closest to the camera A1 from the identification information of the camera A1, and selects the selected actuator Xb1. A control signal for instructing the generation of ultrasonic waves is output.

It is assumed that the specific processing unit B2 sets the type of the image pickup data from the camera A2 as “obstacle” and outputs the type data indicating the type “obstacle” to the allocation unit 30. Then, the allocation unit 30 outputs the imaging data to the specialized processing unit C3 based on the type data from the specific processing unit B2. The specialized processing unit C3 determines whether the attribute of the imaging data is an electric wire, a utility pole, or the like. Then, for example, the specialized processing unit C3 analyzes the imaging data to obtain the distance from the obstacle, and based on the obtained distance and the speed of the flying object 500, avoids the obstacle as necessary. The control signal is output to the actuator X5. As a result, it is possible to avoid a situation where the flying object 500 collides with an obstacle.

It is assumed that the specific processing unit B3 sets the type of the imaging data from the camera A3 to “person” and outputs the type data indicating the type “person” to the allocation unit 30. Then, the allocation unit 30 outputs the imaging data to the specialized processing unit C2 based on the type data from the specific processing unit B3. The specialized processing unit C2 determines whether the attribute of the imaged data is “soccer”, “baseball”, “kiting”, or the like. Then, the specialized processing unit C2 analyzes, for example, the movement of the ball or the like included in the imaging data, and selects the actuator Xb3 that is the notification means closest to the camera A3 from the identification information of the camera A3. Next, the specialized processing unit C2 outputs a control signal for instructing the selected actuator Xb3 to output a caution sound or the like, if necessary.

It is assumed that the specific processing unit B4 sets the type of the imaging data from the camera A3 to “bird” and outputs the type data indicating the type “bird” to the allocation unit 30. Then, the allocation unit 30 outputs the imaging data to the specialized processing unit C1 based on the type data from the specific processing unit B4, and the specialized processing unit C1 determines what the attribute of the imaging data is. For example, when the specialized processing unit C1 determines that the attribute of the imaging data is “dove”, the specialized processing unit C1 selects the actuator Xb4, which is the notification means closest to the camera A4, from the identification information of the camera A4, and selects the selected actuator Xb4. A control signal for instructing light emission is output.

Next, an example in which the allocation control system 100 is applied to a facility such as a stadium will be described with reference to FIGS. 6 and 7. 6 and 7, the allocation control system 100 has six cameras A, the allocation control device 10 has six specific processing units B and six specialized processing units C, and the actuator unit 90 has six actuators. An example with X is shown. The plurality of cameras A1 to A6 are fixed-point cameras, and the plurality of actuators X1 to X6 are movable cameras that capture a moving image or a still image and generate captured image data.

The specialized processing units C1 to C3 correspond to the type “ball”, and the specialized processing units C4 to C6 correspond to the type “person”. Each specialized processing unit C controls the operation of one or more actuators X based on the analysis result of the imaging data. The specialized processing units C1 to C3 determine the position and movement of the ball as attributes by analyzing the imaged data. That is, the specialized processing units C1 to C3 select one or a plurality of actuators X to be controlled, based on the position and movement of the ball included in the imaging data. Then, the specialized processing units C1 to C3 generate, for example, a control signal instructing an operation that follows the movement of the ball, and output the control signal to the selected one or a plurality of actuators X.

The specialized processing units C4 to C6 analyze the imaging data to determine the number, position, movement, etc. of people as attributes. That is, the specialized processing units C4 to C6 select one or a plurality of actuators X to be controlled, based on the number, position, movement, etc. of persons included in the imaged data. The specialized processing units C4 to C6 indicate a specific person such as "crowded" indicating a state in which people are gathered, "falling" indicating a state in which people are collapsing, or an ace as an attribute in the type "person". The “spot” and the “warning” indicating the situation where the referee holds the card are identified. These attributes may be set for each competition or event.

The specialized processing unit C may use the identification information of the camera A to select one or a plurality of actuators X. That is, for example, a ball exists in the area R of FIG. 6, the imaged data of the camera A1 is output to the specialized processing unit C1 via the identification processing unit B1 and the allocation unit 30, and the imaged data of the camera A6 is identified. It is assumed that the data is output to the specialized processing unit C2 via the unit B6 and the allocating unit 30 (see the broken line arrow in FIG. 7). The actuators X1, X2, and X3 are associated with the camera A1, and the actuators X4, X5, and X6 are associated with the camera A6. At this time, the specialized processing unit C1 may select at least one of the actuators X1, X2, and X3 based on the identification information of the camera A1. Similarly, the specialized processing unit C2 may select at least one of the actuators X4, X5, and X6 based on the identification information of the camera A6.

In addition, a plurality of specialized processing units C may cooperate with each other so that the control target of each specialized processing unit C is switched each time depending on the situation. It should be noted that one of the plurality of specialized processing units C may be provided with a function of controlling the selection of the actuator X in a centralized manner. You may provide the processing part which controls it. Alternatively, if one or a plurality of actuators X are associated with each camera A in advance and a plurality of control signals are simultaneously output to the actuators X, the actuator X determines the operation content from the plurality of control signals. May be. That is, in the example of FIG. 6, the actuators X6 and X1 for the camera A1, the actuators X1 and X2 for the camera A2, the actuators X2 and X3 for the camera A3, the actuators X3 and X4 for the camera A4, the actuators X4 and X5 for the camera A5, and the camera A6. It is assumed that the actuators X5 and X6 are associated with. Then, in the situation of the dashed arrow in FIG. 7, control signals are output to both the specialized processing unit C1 and the specialized processing unit C2 to the actuator X6. In such a situation, the actuator X6 may operate by selecting either the control signal of the specialized processing unit C1 or the control signal of the specialized processing unit C2 based on a predetermined priority or the like. Alternatively, the actuator X6 may operate based on both the control signal of the specialized processing unit C1 and the control signal of the specialized processing unit C2. That is, when a plurality of control signals are output, the actuator X may perform a predetermined arithmetic process on the plurality of control signals to determine the degree of operation.

In the configuration of FIG. 7, the allocation control system 100 constitutes an image display system 400 together with the cameras A1 to A6, the actuator section 90, the display control device 95, and the display device 99. The display control device 95 includes a selection unit 96 and a display processing unit 98. The selection unit 96 analyzes each of the image pickup data output from the actuators X1 to X6 and selects the image pickup data for display on the display device 99. The selection unit 96 has analysis processing units Y1 to Y6 connected to the actuators X1 to X6 in a one-to-one relationship, and a selection processing unit 97 connected to the analysis processing units Y1 to Y6. When the analysis processing units Y1 to Y6 are referred to without distinction, they are referred to as “analysis processing unit Y”.

The analysis processing unit Y analyzes the image pickup data output from the actuator X and obtains an evaluation index indicating the importance of the image pickup data. Then, the analysis processing unit Y outputs the imaging data used for the analysis together with the obtained evaluation index to the selection processing unit 97. The analysis processing unit Y has a communication unit, a storage unit, and a control unit, although not shown in the figure, like the specialized processing unit C described above. The control unit is composed of, for example, a CPU or a GPU. The calculation algorithm of the evaluation index may be selectively determined by the analysis processing unit Y according to the content of the imaging data, or may be set in advance according to the competition held at the facility.

The selection processing unit 97 selects one or a plurality of imaging data based on the evaluation index output from each analysis processing unit Y and outputs the selected imaging data to the display processing unit 98. That is, the selection processing unit 97 selects one piece of imaging data when one screen is set to display one screen on the display unit of the display device 99, and causes the display unit of the display device 99 to display a plurality of screens. Multiple image data is selected when multi-screen is set.

The display processing unit 98 causes the display device 99 to display an image based on the one or a plurality of image pickup data output from the selection processing unit 97. In the case of multi-screen setting, the priority is set for each screen. Therefore, the display processing unit 98 sequentially assigns the imaging data with a high evaluation index to the screen with a high priority. Here, the display processing unit 98 may acquire the image pickup data of the camera A from the specialized processing unit C via the network N. Then, the display processing unit 98 may switch the image displayed on the display device 99 to the image of the camera A in a predetermined situation such as a scene in which a person or a ball does not move. The display device 99 includes, for example, a liquid crystal display (Liquid Crystal Display), and displays various images on the display unit.

Incidentally, the image display system 400 may be configured by incorporating the allocation control device 10 into the selection unit 96 of the display control device 95, as shown in FIG. FIG. 8 partially shows the configuration of the image display system 400. In the case of such a configuration, the specific processing units B7 to B12 specify the type such as “person” or “ball”, and the specialized processing units C7 to C12 specify one type such as “person” or “ball”. Execute specialized processing. That is, in the configuration of FIG. 8, the specific processing unit B and the specialized processing unit C execute the same processing as the selection processing unit 97. Therefore, the load can be distributed to the specific processing unit B and the specialized processing unit C, and the load per processing unit can be reduced.

Although the example in which the camera faces the inside of the field is shown in FIG. 6, the present invention is not limited to this, and at least one of the plurality of cameras A and the plurality of actuators X may be directed to a passenger seat or a bench. .. 6 to 8 illustrate the image display system 400 including the six cameras A and the six actuators X, the number of cameras A and the number of actuators X can be arbitrarily changed without being limited to this. .. However, the actuator unit 90 may include an actuator X other than the movable camera.

Next, an example in which the allocation control system 100 is applied to a street will be described with reference to FIGS. 9 and 10. 9 and 10, the allocation control system 100 has six cameras A, the allocation control device 10 has six specific processing units B and six specialized processing units C, and the actuator unit 90 has nine actuators. An example with X is shown. The specialized processing units C1 and C2 correspond to the type "automobile", the specialized processing units C3 and C4 correspond to the type "person", the specialized processing unit C5 corresponds to the type "bicycle", and the specialized processing unit C6 corresponds to the type " "Ball". The actuators X1, X2, X7, X8, and X9 are electronic display boards including a light source such as an LED, and the actuators X3 to X6 are notification means including a speaker.

In the example of FIG. 9, the camera A1 and the actuator X1 are provided near the traffic signal 160A. The display plate X1a of the actuator X1 is arranged so that it can be visually recognized from a vehicle approaching the traffic light 160A. The camera A2 and the actuator X2 are provided near the traffic signal 160B. The display board X2a of the actuator X2 is arranged so that it can be visually recognized from a vehicle approaching the traffic light 160B. The traffic signal 161A and the traffic signal 161B are arranged at each end of the pedestrian crossing and face each other.

The monitoring/warning unit 150A includes a housing 151 that houses the camera A3, the camera A4, the actuator X3, and the actuator X4, a leg portion 152 that supports the housing 151, and an actuator X7 provided on the upper portion of the housing 151. have. The actuator X7 has a display board X7a and a display board X7b.

The monitoring and warning unit 150B supports the first housing 151a that houses the camera A5 and the actuator X5, the second housing 151b that houses the camera A6 and the actuator X6, and the first housing 151a and the second housing 151b. And a leg portion 152. Further, the monitoring/warning unit 150B has actuators X8 and X9 provided above the first casing 151a and the second casing 151b. The actuator X8 has a display board X8a, and the actuator X9 has a display board X9a.

In such a configuration, when the specialized processing units C1 and C2 acquire the imaging data from the allocation unit 30, the specialized processing units C1 and C2 determine the attribute of the type “automobile” of the imaging data. Attributes of the type “car” include “speed attention”, “looking aside”, “while driving”, “meandering”, “no lights at night”, and the like. For example, when the specialized processing unit C1 determines that the attribute of the image data captured by the camera A1 is “no lights at night”, the specialized processing unit C1 issues a control signal to the actuator X1 to display a caution message such as “Turn on lights”. Output. As a result, the actuator X1 causes the display board X1a to display a warning letter or the like. Hereinafter, the control signal for instructing the audio output of the alert is also referred to as the alert signal.

When the specialized processing units C3 and C4 acquire the imaging data from the allocation unit 30, the specialized processing units C3 and C4 determine the attribute of the type “person” of the imaging data. Attributes of the type “person” include “child”, “person with a cane”, “wheelchair”, “person running”, “smartphone”, “group”, “suspicious person”, and the like. For example, when the specialized processing unit C4 determines that the attribute of the image data captured by the camera A3 is “child”, the specialized processing unit C4 outputs to the actuator X3 a caution signal instructing the output of voice such as “Be careful of the car”. As a result, the actuator X3 outputs a warning sound.

When the specialized processing unit C5 acquires the imaging data from the allocation unit 30, the specialized processing unit C5 determines the attribute of the type “bicycle” of the imaging data. Attributes of the type “bicycle” include “walkway”, “smartphone”, “speed attention”, “meandering driving”, and “two-seater”. The attributes relating to speed may be set in stages such as “high speed”, “medium speed”, and “low speed”. Here, it is assumed that the sidewalking by bicycle is prohibited in the place of FIG. 9, and the information to that effect is set in each specialized processing unit C. In such a situation, the specialized processing unit C5, which has determined that the attribute of the image data captured by the camera A4 is "sidewalk", has a voice such as "Let's run in the bicycle lane" "Running on the sidewalk is prohibited". A caution signal for instructing output is output to actuator X4. As a result, the actuator X4 outputs a warning sound.

When the specialized processing unit C6 acquires the imaging data from the allocation unit 30, the specialized processing unit C6 determines the attribute of the type “ball” of the imaging data. The attributes of the type “ball” are “bound”, which indicates that the person is bouncing the ball with his/her hand, “dribble”, which indicates that the person is kicking the ball with his/her foot, and the person who throws the ball by hand. There is a "throw" etc. that indicates the situation. For example, when the specialized processing unit C6 determines that the attribute of the image data captured by the camera A5 is “bound”, the actuator issues a caution signal instructing the output of voice such as “Danger. Please hold it firmly with your hand.” Output to X5. As a result, the actuator X5 outputs a warning sound.

However, each specialized processing unit C may output a control signal to the plurality of actuators X. By the way, it is considered that the main object included in the image pickup data of the cameras A1 and A2 is usually an automobile. Therefore, when the type of the image data captured by the camera A1 or A2 is specified as “person” or “ball”, special caution is required. Therefore, when the specialized processing unit C other than the type “automobile” receives the image pickup data from the camera A1 or A2, it is preferable to reflect the fact in the processing of the actuator X. For example, when the specialized processing unit C3 acquires the imaged data from the camera A2, the specialized processing unit C3 may output a caution signal to the actuator X6 to instruct the output of a voice such as "It is dangerous. Please return to the sidewalk." At that time, the specialized processing unit C3 may also output a control signal for instructing the display of a caution such as “danger!” to the actuator X9. Similar to the above, special processing may be set for the image data of the other cameras A, depending on the arrangement and the like.

Further, as illustrated in FIG. 10, each specialized processing unit C acquires signal data indicating the status of each traffic signal from the traffic signal controller 600 connected to the network N, and reflects the acquired signal data in the processing. You may let me. For example, when the traffic lights 161A and 161B are blue and the camera A2 captures a situation where the car is heading toward the pedestrian crossing at a tremendous speed, the specialized processing unit C1 or C2 instructs the output of a voice such as "slow down!" It is preferable to output a caution signal to the actuator X2. However, the plurality of specialized processing units C may control the actuator unit 90 in cooperation with each other.

Although FIG. 9 shows the monitoring and warning unit 150A or 150B including one or more cameras A and one or more actuators X, these are merely examples, and the arrangement, number, and shape of the cameras A and actuators X, etc. Is not limited to the example of FIG. The same applies to the other figures. The monitoring/warning unit 150A may be configured such that the actuator X7 can be automatically or manually rotated about a straight line along the leg portion 152 as an axis. Further, the monitoring/warning unit 150B may be configured such that the angle formed by the actuators X8 and X9 in the horizontal direction can be automatically or manually changed.

Next, the allocation control method according to the first embodiment will be described with reference to the flowchart in FIG. First, the plurality of specific processing units B1 to Bn each input image pickup data from the camera A that is a connection destination (step S101). Next, the plurality of specifying processing units B1 to Bn respectively specify which of the plurality of types the input imaged data is (step S102). Then, each of the plurality of specification processing units B1 to Bn outputs the type data indicating the specified type and the imaging data related to the type specification to the allocation unit 30 (step S103).

The allocating unit 30 allocates imaging data to the specialized processing unit C based on the type data. That is, the allocation unit 30 outputs the image pickup data output together with the type data to the specialized processing unit C corresponding to each type data output from each of the plurality of specific processing units B1 to Bn (step S104). ..

The specialized processing unit C to which the imaging data is output from the allocating unit 30 determines which of the plurality of attributes in the type identified by the identification processing unit B that is the output source, the imaging data (step S105). ). Then, the specialized processing unit C executes processing according to the determined attribute. That is, the specialized processing unit C selects, for example, one or a plurality of actuators X according to the determined attribute and outputs a control signal to the selected one or a plurality of actuators X (step S106).

As described above, in the allocation control device 10 according to the first embodiment, the plurality of specific processing units B connected to the camera A specify the type of the imaging data, and the allocation unit 30 is specified by the specific processing unit B. The imaging data is output to the specialized processing unit C corresponding to the type. Therefore, according to the allocation control device 10, the load can be distributed to the specific processing unit B and the specialized processing unit C, so that the load on each processing unit can be reduced and the captured data can be smoothly utilized for various purposes. be able to. That is, the allocation control device 10 can efficiently use the imaging data for a purpose according to the information included in the imaging data.

Further, when the imaging data is output from the allocating unit 30, the specialized processing unit C determines which of the plurality of attributes the imaging data has, and executes the process according to the determined attribute. In this way, the specialty processing unit C may mainly perform image analysis on the image data regarding the type of specialty. Therefore, by reducing the processing amount of the specialized processing unit C, it is possible to speed up the processing of the allocation control device 10 as a whole. That is, according to the allocation control device 10, by distributing a plurality of processes to each processing unit and reducing the load on each processing unit, it is possible to speed up the entire process, reduce the number of high-performance processing units, and reduce the cost. It is possible to reduce. Further, the specialized processing unit C outputs a control signal according to the attribute of the imaged data based on the determination process to at least one of the plurality of actuators X to be controlled, and the actuator X operates according to the control signal. To execute. Therefore, according to the allocation control device 10, it is possible to realize a quick operation of the actuator X according to the attribute of the imaging data.

Here, the plurality of specialized processing units C1 to Cm may include specialized processing units C corresponding to different types as described above. That is, the allocation control device 10 may include specialized processing units C corresponding to different types. For example, as shown in FIG. 3, the plurality of specialized processing units C1 to Cm may include two or more specialized processing units C corresponding to the same type. That is, in the allocation control device 10, some of the specialized processing units C1 to Cm may correspond to a common type. In this way, the allocation control device 10 changes the combination of the specialized processing units C according to the installation environment of the camera A, such as relatively increasing the specialized processing units C corresponding to the objects that frequently appear in the imaging data. Since it can be changed, it is possible to reduce a state in which one or a plurality of allocation control devices 10 is not functioning and avoid a processing delay. Further, all of the plurality of specialized processing units C1 to Cm may correspond to different types. That is, in the allocation control device 10, all of the plurality of specialized processing units C1 to Cm may correspond to different types. Such a configuration is effective, for example, in a situation in which the object included in the imaging data changes regularly or irregularly.

On the other hand, the plurality of specialized processing units C1 to Cm may all correspond to the same type. That is, in the allocation control device 10, all of the plurality of specialized processing units C1 to Cm may correspond to the same type. Such a configuration is effective, for example, in a situation where various birds hinder the flight of the flying object 500. In such a situation, the allocation control device 10 may include only the specialized processing unit C that specializes in birds as the plurality of specialized processing units C. Similarly, in a field or the like, in a situation where damage by terrestrial animals is the main, the allocation control device 10 may have only the specialized processing unit C that specializes in terrestrial animals as the plurality of specialized processing units C. Good. In addition, it is also assumed that only people are desired to be monitored at an event site, a station yard, a storage place for important confidentiality or cash of a company or a financial institution, a department store, a convenience store, a parking lot, a park, a street corner, an intersection, or the like. In such a case, the allocation control device 10 may include only the specialized processing units C that specialize in people as the plurality of specialized processing units C. In this way, if only a plurality of specialized processing units C corresponding to the same type are included in the allocation control device 10 in accordance with the installation environment of the camera A, there is a specialized processing unit C that hardly functions. Since the situation can be avoided, waste can be eliminated and cost can be reduced.

The allocation control system 100 according to the first embodiment includes an allocation control device 10 and a plurality of actuators X to be controlled by the allocation control device 10. Then, the specialized processing unit C outputs, for example, a control signal according to the determined attribute to at least one actuator X as a process according to the result of the determination regarding the imaged data. Therefore, according to the allocation control system 100, it is possible to realize the quick operation of the actuator X according to the attribute of the imaging data.

By the way, although the allocation control system 100 has been described above as including a plurality of cameras A1 to An, the present invention is not limited to this, and the allocation control system 100 may include a plurality of cameras A1 to An. Good. That is, the plurality of cameras A1 to An may be an external configuration of the allocation control system 100. Further, the allocation control system 100 may include the update processing device 80, or may be configured without including the update processing device 80. In addition, the specialized processing unit C may have a function of determining whether or not an operation command to the actuator X is necessary by analyzing the imaged data. That is, when it is determined that the operation instruction to the actuator X is unnecessary, the specialized processing unit C may not output the control signal.

<Modification 1-1>
As illustrated in FIG. 12, the allocation control device 10 according to the present modification 1-1 has a central update unit 50 that updates the analysis processing program 432 of each specialized processing unit C in cooperation with the update processing device 80. ing. Therefore, the update processing unit 44c of each specialized processing unit C functions as a relay unit between the centralized updating unit 50 and the storage unit 43b. The integrated update unit 50 and the update processing device 80 may perform wired communication, wireless communication, or wired and wireless combination.

That is, the allocation control device 10 of the present modification 1-1 further includes the centralized updating unit 50 connected to the plurality of specialized processing units C. Here, the specialized processing unit C has an analysis processing program 432 relating to attribute determination, and outputs the imaging data and the analysis information to the overall update unit 50. The integrated update unit 50 transmits the imaging data and analysis information output from each of the plurality of specialized processing units C to the update processing device 80. The update processing device 80 updates the respective analysis processing programs 432 of the plurality of specialized processing units C based on the imaging data and the analysis information transmitted from the general updating unit 50, and centrally updates the updated analysis processing programs 432. It is transmitted to the unit 50. When the updated analysis processing program 432 is transmitted from the update processing device 80, the centralized updating section 50 transfers it to the specialized processing section C corresponding to the attribute of the analysis processing program 432. That is, the update processing device 80 transmits the updated analysis processing program 432 to each specialized processing unit C via the centralized updating unit 50. Therefore, when the specialized processing unit C acquires the updated analysis processing program 432 from the centralized updating unit 50, the specialized processing unit C can update the analysis processing program 432 in the storage unit 43 by the analysis processing program 432.

As described above, according to the assignment control device 10 of the present modification 1-1, a part of the load of each specialized processing unit C is assigned to the centralized updating unit 50, and the processing amount of each specialized processing unit C can be reduced. Therefore, the processing of the allocation control device 10 can be speeded up. Further, since the specifications of each specialized processing unit C can be lowered, the cost can be reduced.

<Modification 1-2>
In the above description, an example in which the allocation control system 100 has a plurality of cameras A is shown. However, the allocation control system 100 of the present modification 1-2 has one camera A as shown in FIG. There is. That is, only one camera A is connected to the allocation control device 10 of Modification 1-2.

More specifically, the allocation control device 10 includes an allocation processing unit 20 connected to the camera A and a plurality of specialized processing units C1 to Cm corresponding to any one of a plurality of types. The plurality of specialized processing units C1 to Cm are connected to the allocation processing unit 20. The allocation processing unit 20 has a specific processing unit B connected to the camera A. The identification processing unit B identifies which of the plurality of types the imaged data input from the camera A is, and outputs the imaged data together with the type data indicating the identified type to the allocation unit 30. The allocation unit 30 outputs the imaging data output together with the type data to the specialized processing unit C corresponding to the type data output from the specific processing unit B.

The other configurations of the allocation control device 10 and the allocation control system 100 and the content of the allocation control method are the same as those described with reference to FIGS. 1 to 11. Further, the configuration of Modification 1-1 described above may be applied to the configuration of Modification 1-2.

As described above, since the load can be distributed to the specific processing unit B and the specialized processing unit C also by the allocation control device 10 of the present modification 1-2, the load of each processing unit can be reduced, and the imaging data can be obtained. Can be used smoothly for various purposes. That is, according to the allocation control device 10, the usage of the image pickup data can be flexibly changed according to the information contained in the image pickup data, which changes every moment.

Here, with reference to FIGS. 14 to 17, an example in which the allocation control system 100 of the present modification 1-2 is applied to an intersection or the like in a city will be described. 14 and 15 exemplify a display device including a display processing unit 90a and a display unit 90b as the actuator X forming the actuator unit 90. The imaging range of the camera A includes a waiting area on the side opposite to the traffic light 161 in the pedestrian crossing. A person H standing in the waiting area can visually recognize the display portion 90b of the actuator X.

First, an example of processing by the allocation control device 10 will be described with reference to the flowchart in FIG. In the example of FIG. 16, the identification processing unit B is premised on identifying, as the type, the main age group of the people included in the imaging data, that is, the age group to which the people included in the imaging data most belong. Therefore, the identification processing unit B identifies the age group such as 20s, 30s, 40s, 50s, 60s, 70s, 80s or older as a type. Therefore, each specialized processing unit C is configured to correspond to any one of a plurality of age groups. However, the age group is not limited to the unit of 10 years as described above, and the age range may be narrowed or widened. Further, the age range may be individually set for each age group, and the age groups having different age ranges may be combined.

When the specialized processing unit C acquires the imaging data from the allocation unit 30 (step S111), the specialized processing unit C determines whether or not the acquired imaging data includes a relatively large number of walking people (step S112). When it is determined that a relatively large number of walking people are included (step S112/Yes), the professional processing unit C causes the actuator X to display a music moving image corresponding to the specified age group (step S113). On the other hand, when it is determined that the number of people waiting for a signal is relatively large (step S112/No), the specialty processing unit C causes the actuator X to display a moving image advertisement according to the specified age group (step S114). ..

Next, another example of the processing by the allocation control device 10 will be described with reference to the flowchart in FIG. The same steps as those in FIG. 16 are denoted by the same reference numerals and the description thereof will be omitted. In the example of FIG. 17, each specialized processing unit C has a condition as to whether a relatively large number of walking people are included and a condition as to whether or not the number of people included in the imaging data is equal to or more than a threshold value. A plurality of attributes are set by the combination of.

That is, when the specialized processing unit C determines that a relatively large number of walking people are included (step S112/Yes), whether the number of people included in the imaging data is equal to or greater than a preset threshold value. Is determined (step S121). If the number of people is equal to or more than the threshold value (step S121/Yes), the specialized processing unit C causes the actuator X to display a music moving image of relatively high importance (step S123). On the other hand, when the number of people is less than the threshold value (step S121/No), the specialty processing unit C causes the actuator X to display a music moving image of relatively low importance (step S124).

When it is determined that the number of walking people is not relatively large (step S112/No), the professional processing unit C determines whether the number of people included in the imaging data is equal to or larger than a preset threshold value. The determination is made (step S122). If the number of people is equal to or greater than the threshold value (step S122/Yes), the specialized processing unit C causes the actuator X to display a moving image advertisement having a relatively high degree of importance (step S125). On the other hand, when the number of people is less than the threshold value (step S122/No), the specialty processing unit C causes the actuator X to display a moving image advertisement having a relatively low degree of importance (step S126). The threshold value of step S121 and the threshold value of step S122 may have the same value or different values. However, the threshold for the number of people may be set more finely.

In the description using FIG. 16 and FIG. 17, the setting of the music video and the video advertisement associated with each condition is merely an example. That is, the allocation control device 10 may cause the actuator X to display various moving images or still images according to settings that match the preferences of the sponsor. In addition, the specialized processing unit C may select a moving image or a still image with a classifier generated by learning such as deep learning, and may display the selected moving image or a still image on the actuator X.

Further, not limited to the examples of FIGS. 16 and 17, for example, the identification processing unit B may identify, as the type, whether the imaging data includes a relatively large number of walking people. In this case, each specialized processing unit C corresponds to either the type “walking” or the type “waiting for a signal”, and determines the age group as an attribute. For example, after determining the age group, the specialized processing unit C refers to table information that associates a plurality of age groups with the moving images to be displayed, selects a moving image or a still image, and selects the selected moving image or still image as an actuator. It may be displayed on X. Alternatively, the specialized processing unit C may select a moving image or a still image with a classifier generated by learning such as deep learning and display the selected moving image or still image on the actuator X. However, each specialized processing unit C may determine the attribute that combines the condition of the age group and the condition of whether or not the number of people included in the imaging data is equal to or more than the threshold value.

Embodiment 2.
The configurations of the allocation control device and the allocation control system according to the second embodiment will be described with reference to FIG. The same components as those in the first embodiment described above are designated by the same reference numerals and the description thereof will be omitted.

The allocation control system 200 includes a plurality of cameras A1 to An, an allocation control device 110, and an actuator section 90. The allocation control device 110 has an allocation processing unit 20 including a plurality of specific processing units B1 to Bn and an allocation unit 30, and a plurality of specialized processing units C1 to Cm.

The allocation control device 110 includes an attribute allocation unit 70 connected to the plurality of specialized processing units C1 to Cm, and a plurality of attribute processing units D1 to Dk corresponding to any of the plurality of attributes. .. The plurality of attribute processing units D1 to Dk are connected to the attribute assigning unit 70. When the plurality of attribute processing units D1 to Dk are referred to without distinction, they are referred to as "attribute processing unit D".

When the imaging data is output from the allocating unit 30, the specialized processing unit C determines which of the plurality of attributes in the type identified by the identification processing unit B that is the output source, the imaging data. Then, the specialized processing unit C of the first embodiment outputs the determination result, that is, the attribute data indicating the determined attribute to the attribute assigning unit 70.

As shown by the broken line in FIG. 18, the attribute assigning unit 70 assigns the attribute data input from the plurality of specialized processing units C1 to Cm to any of the plurality of attribute processing units D1 to Dk, and has m×k routes. Have (choices). The attribute assigning unit 70 outputs the attribute data output from each of the plurality of specialty processing units C to the attribute processing unit D corresponding to the attribute indicated by the attribute data. The attribute assigning unit 70 is composed of, for example, an FPGA. The attribute allocation program, which is a logic program of the attribute allocation unit 70, can be appropriately changed according to the use case.

The attribute processing unit D is associated with one attribute, that is, one attribute data, and when the attribute data is output from the attribute assigning unit 70, the attribute processing unit D executes a process according to the attribute data. The attribute processing unit D has a communication unit, a storage unit, and a control unit, which are not shown in the figure, like the specialized processing unit C described above. The control unit is composed of, for example, a CPU or a GPU. Information indicating the processing associated with each attribute data is stored in the storage unit. In the second embodiment, when the attribute data is output from the attribute assigning unit 70, the attribute processing unit D outputs a control signal indicating a process previously associated with the attribute data to the actuator X as a connection destination. ..

Here, the specialized processing unit C may output attribute data including additional information indicating detailed control contents of the actuator X. The additional information is, for example, information indicating volume/speech when the actuator X is a notification unit, and information indicating brightness/emission color when the actuator X is a light emitting unit. In this case, the attribute processing unit D generates a control signal based on the attribute data and outputs the generated control signal to the actuator X of the transmission destination. The internal storage device of the attribute processing unit D may store table information in which each of the plurality of pieces of additional information is associated with the detailed control content of the actuator X. In this case, the attribute processing unit D generates a control signal by comparing the additional information included in the attribute data with the table information.

By the way, the specialized processing unit C may have a function of determining the necessity of an operation command to the actuator X by analyzing the imaged data. In this case, when it is determined that the operation command to the actuator X is unnecessary, the specialized processing unit C may not output the attribute data. Alternatively, the specialty processing unit C may output attribute data including additional information indicating that control of the actuator X is unnecessary.

Next, an application example of the allocation control system 200 will be described with reference to FIG. Here, an example in which the allocation control system 200 is incorporated into monitoring around a pedestrian crossing will be described. In FIG. 19, the allocation control system 200 has four cameras A, the allocation control device 110 has four specific processing units B, five specialized processing units C, and k attribute processing units D, and an actuator unit. An example in which 90 has k actuators X is shown. In the example of FIG. 19, the traffic signal controller 600 is connected to the network N, and each specialized processing unit C can acquire the signal data indicating the status of the traffic signal (not shown) from the traffic signal controller 600. Has become.

Here, the specialized processing units C1 and C2 correspond to the type “person”, the specialized processing unit C3 corresponds to the type “bicycle”, the specialized processing unit C4 corresponds to the type “ball”, and the specialized processing unit It is assumed that C5 corresponds to the type “car”. Further, it is assumed that the attribute processing unit Dk corresponds to operation control via the network N, and the actuator Xk is a notification unit arranged in a vehicle passing through the monitoring area by the camera A and generating sound or voice. To do.

In the case of this example, when the specialized processing units C1 and C2 acquire the imaging data from the allocation unit 30, the specialized processing units C1 and C2 determine the attribute of the type “person” of the imaging data. For example, in a situation where the attribute processing unit D3 corresponds to the attribute “suspicious person” indicating a person having a suspicious movement, the specialized processing unit C1 outputs the attribute data indicating the attribute “suspicious person” to the attribute assigning unit 70, It is assumed that the attribute assignment unit 70 outputs the attribute data to the attribute processing unit D3. Here, it is assumed that the actuator X2 is a notification unit capable of outputting a sound and the actuator X3 is an electronic display board. Then, it is assumed that the attribute data indicating the attribute “suspicious person” is associated with the process of causing the actuator X2 to output a caution sound and the process of displaying the warning character on the actuator X3. In this case, the attribute processing unit D3 outputs a warning signal to the actuator X2 and also outputs a control signal for instructing the display of warning characters to the actuator X3.

When the specialized processing unit C3 acquires the imaging data from the allocation unit 30, the specialized processing unit C3 determines the attribute of the type “bicycle” of the imaging data. For example, in a situation in which the attribute processing unit D4 supports the attribute “while smartphone”, the specialized processing unit C3 outputs the attribute data indicating the attribute “bicycle” to the attribute allocation unit 70, and the attribute allocation unit 70 outputs the attribute data. Is output to the attribute processing unit D4. Here, it is assumed that the actuator X1 is a notification means, and the attribute data indicating the attribute “while being a smartphone” is associated with the process of causing the actuator X1 to output a warning sound. In this case, the attribute processing unit D4 outputs a control signal instructing the output of the warning sound to the actuator X1.

When the specialized processing unit C4 acquires the imaging data from the allocation unit 30, the specialized processing unit C4 determines the attribute of the type “ball” of the imaging data. For example, when the attribute processing unit D1 supports the attribute “dribble”, the specialized processing unit C4 outputs the attribute data indicating the attribute “dribble” to the attribute assigning unit 70, and the attribute assigning unit 70 outputs the attribute data. It is assumed that the data is output to the attribute processing unit D1. Here, it is assumed that the actuator X3 is an electronic display board, and the attribute data indicating the attribute “dribble” is associated with the process of displaying the warning character on the actuator X3. In this case, the attribute processing unit D1 outputs a control signal for instructing the display of warning characters to the actuator X3.

When the specialized processing unit C5 acquires the imaging data from the allocation unit 30, the specialized processing unit C5 determines the attribute of the type “automobile” of the imaging data. For example, in a situation where the attribute processing unit D2 corresponds to the attribute “meandering driving”, the specialized processing unit C4 outputs the attribute data indicating the attribute “meandering driving” to the attribute assigning unit 70, and the attribute assigning unit 70 outputs the attribute. It is assumed that the data is output to the attribute processing unit D2. Here, it is assumed that the actuator X4 is a lightning display panel attached around the traffic light, and the attribute data indicating the attribute “meandering operation” is associated with the process of displaying information recommending a break on the actuator X4. To do. In this case, the attribute processing unit D2 outputs a control signal for instructing the display of information recommending a break to the actuator X4.

Here, the specialized processing unit C may acquire signal data from the traffic signal controller 600, and further subdivide the attributes based on the acquired signal data. For example, under the situation where a pedestrian tries to cross a pedestrian crossing when the traffic light on the side of the pedestrian is red, the specialized processing unit C sets the attribute of the imaging data to “danger”. The attribute determination process may be performed in cooperation with the traffic signal controller 600.

Alternatively, the plurality of specialized processing units C may cooperate with each other to perform the determination process. For example, the specialized processing unit C of the type “person” transmits current affair information indicating that the attribute “child” is determined to the specialized processing unit C of the type “bicycle”, and the specialized processing unit C of the type “bicycle” Cooperation such as using current affairs information for attribute determination is possible. Further, a specialized processing unit C of the type “traffic signal” is provided in association with the camera A whose signal range is included in the imaging range, and the specialized processing unit C outputs current affairs information indicating the state of the traffic signal to another specialized processing unit C. You can In this way, each specialized processing unit C can perform the determination process utilizing the current affairs information from the specialized processing unit C of the type “traffic signal” without acquiring the signal data from the traffic signal controller 600. it can. In addition, the specialized processing unit C that has determined that a pedestrian has jumped out onto the road may transmit the current affairs information to that effect to the specialized processing unit C of the type “automobile”. Then, the specialized processing unit C of the type “automobile” may output a control signal for instructing the generation of sound or voice to the actuator Xk via the network through the attribute allocation unit 70 and the attribute processing unit Dk.

Next, the allocation control method according to the second embodiment will be described with reference to the flowchart in FIG. The same steps as those in FIG. 11 are denoted by the same step numbers and description thereof is omitted.

First, the allocation processing unit 20 executes the processing of steps S101 to S104. Next, the specialized processing unit C to which the imaging data is output from the allocation unit 30 determines which of the plurality of attributes in the type specified by the specification processing unit B of the output source the imaging data is ( Step S105). Then, the specialized processing unit C outputs the attribute data indicating the determination result to the attribute assigning unit 70 (step S201).

When the specialty processing unit C outputs the attribute data, the attribute assigning unit 70 assigns the attribute data to any of the plurality of attribute processing units D1 to Dk. That is, the attribute allocation unit 70 outputs the attribute data to the attribute processing unit D corresponding to the attribute indicated by the attribute data output from each of the plurality of specialized processing units C (step S202).

The attribute processing unit D, when the attribute data is output from the attribute assigning unit 70, executes a process according to the attribute data. The attribute processing unit D according to the second embodiment generates a control signal corresponding to the attribute data and outputs the generated control signal to at least one actuator X as a process according to the attribute data. The attribute processing unit D outputs, for example, a control signal for instructing light emission to the actuator X which is a light emitting means (step S203).

As described above, in the allocation control device 110 according to the second embodiment, the plurality of specification processing units B connected to the camera A specify the type of the imaging data, and the allocation unit 30 is specified by the specification processing unit B. The imaging data is output to the specialized processing unit C corresponding to the type. Then, the specialized processing unit C determines the attribute of the imaging data and outputs the attribute data indicating the determination result to the attribute assigning unit 70. Then, the attribute assigning unit 70 outputs the attribute data to the attribute processing unit D corresponding to the attribute data, and the attribute processing unit D executes processing according to the attribute data. Therefore, according to the allocation control device 110, the load can be distributed to the specific processing unit B, the specialized processing unit C, and the attribute processing unit D, so that the load of each processing unit can be reduced and the imaging data can be used for various purposes. It can be used smoothly. That is, the specialized processing unit C according to the second embodiment only needs to perform image analysis for the specialized type on the imaged data, and the attribute processing unit D can leave the instruction to the actuator X. Then, by further reducing the processing amount of the specialized processing unit C and distributing the load, it is possible to speed up the processing of the allocation control device 10 as a whole.

Further, when the attribute data is output from the attribute assigning unit 70, the attribute processing unit D outputs a control signal corresponding to the attribute data to at least one of the plurality of actuators X to be controlled, and the actuator X executes the operation according to the control signal. Therefore, according to the allocation control device 110, it is possible to realize a quick operation of the actuator X according to the attribute of the imaging data.

The allocation control system 200 according to the second embodiment includes an allocation control device 110 and a plurality of actuators X to be controlled by the allocation control device 110. Then, when the attribute data is output from the attribute assigning unit 70, the attribute processing unit D outputs a control signal according to the attribute data to at least one actuator X, and the actuator X executes an operation according to the control signal. To do. Therefore, according to the allocation control system 200, it is possible to realize the quick operation of the actuator X according to the attribute of the imaging data.

The configurations of the alternative configuration and the modified example 1-2 described in the first embodiment can be applied to the allocation control device 110 and the allocation control system 200 of the second embodiment. Other effects are similar to those of the first embodiment.

<Modification 2-1>
Next, a configuration example of the allocation control device and the allocation control system according to the modification 2-1 will be described with reference to FIG. The specialized processing unit C in the present modification 2-1 outputs the image pickup data related to the determination to the attribute assigning unit 70 together with the attribute data indicating the determination result. When the attribute data and the imaging data are output from the specialized processing unit C, the attribute allocation unit 70 in the present modification 2-1 allocates the imaging data to the attribute processing unit D corresponding to the attribute indicated by the attribute data. That is, the attribute allocation unit 70 outputs the image pickup data output together with the attribute data to the attribute processing unit D corresponding to the attribute indicated by the attribute data output from each of the plurality of specialized processing units C.

The attribute processing unit D of the present modification 2-1 has approximately the same performance as the specialized processing unit C, and has the same functional configuration as the specialized processing unit C shown in FIG. The attribute processing unit D has a function of performing image analysis specialized for any of a plurality of attributes in a specific type. That is, the attribute processing unit D is equipped with an AI based on learning such as deep learning, and extracts the feature amount of the main object included in the imaging data to classify the attribute more finely. When the image capturing data is output from the attribute assigning unit 70, the attribute processing unit D analyzes the image capturing data and extracts a feature amount. Then, the attribute processing unit D generates a control signal according to the extracted feature amount, and outputs the generated control signal to at least one actuator X.

The update processing device 80 of the present modification 2-1 stores the control program of each attribute processing unit D. The update processing device 80 updates the control program by learning based on the imaging data, the analysis result, and the like transmitted from the attribute processing unit D, and transmits the updated control program to the attribute processing unit D at a preset timing. As a result, each attribute processing unit D can update the control program.

By the way, the attribute assignment unit 70 may output the attribute data to the attribute processing unit D together with the imaging data. Then, the attribute processing unit D may use the attribute data for the feature amount extraction processing.

As described above, the assignment control device 110 of the present modification 2-1 makes the attribute processing unit D have the image analysis function. Therefore, since the attribute processing unit D can issue a finer instruction to at least one actuator X according to the feature amount extracted from the imaging data, the actuator unit 90 can be caused to perform a more significant operation. it can. Further, the update processing device 80 can utilize the analysis result of the specialized processing section C and the analysis result of the attribute processing section D for the update processing of the respective control programs of the specialized processing section C and the attribute processing section D. Therefore, the accuracy of updating the control program can be improved, and the efficiency of updating the control programs of the specialized processing unit C and the attribute processing unit D can be improved.

<Modification 2-2>
Next, with reference to FIG. 22, a configuration example of the allocation control device and the allocation control system according to Modification 2-2 will be described. That is, although the allocation control device 110 in which each component is mounted on one board or the allocation control device 110 configured by one chip is assumed in FIGS. 18, 19, and 21, the present invention is not limited to this. Not done.

The allocation control device 10 of the present modification 2-2 adopts a configuration in which each component is mounted on two boards or a configuration with two chips. More specifically, the allocation control device 110 includes a first control device 110A having an allocation processing unit 20 and a plurality of specialized processing units C1 to Cm, an attribute allocation unit 70, and a plurality of attribute processing units D1 to Dk. And a second control device 110B.

As described above, the allocation control device 10 of the present modification 2-2 is composed of two parts, the first control device 110A and the second control device 110B. Therefore, the degree of freedom of arrangement is increased, and maintenance and management are facilitated, which improves user convenience. However, you may apply the structure of the modification 2-1 to the 2nd control apparatus 110B.

Embodiment 3.
A configuration example of the allocation control system according to the third embodiment of the present invention will be described with reference to FIG. The allocation control system 300 according to the third embodiment can have a configuration including a plurality of allocation control devices 10 or 110, or a combination of the allocation control device 10 and the allocation control device 110. FIG. 23 shows a configuration example in which the allocation control system 300 includes two allocation control devices 10, two allocation control devices 110, or the allocation control device 10 and the allocation control device 110. Hereinafter, when the allocation control device 10 and the allocation control device 110 are collectively referred to, the reference numerals are omitted and the allocation control device is described.

In the allocation control device constituting the allocation control system 300, each specialized processing unit C is connected to the update processing device 80 via the network N. In the case of the allocation control device 110 of modification 2-1, each attribute processing unit D is also connected to the update processing device 80 via the network N. That is, in the allocation control system 300, the imaging data output from the plurality of allocation control devices and the analysis results of the specialized processing units C and the like can be collected in the update processing device 80.

The update processing device 80 according to the third embodiment can analyze the imaging data input from the plurality of allocation control devices and reflect the analyzed result in the update of the control program of the specialized processing unit C or the attribute processing unit D. it can. That is, the update processing device 80 updates each analysis processing program 432 of each of the plurality of specialized processing units C based on the imaging data and the analysis information transmitted from each of the plurality of specialized processing units C of each allocation control device. To do. Then, the update processing device 80 transmits the updated analysis processing program 432 to each allocation control device. In the configuration of FIG. 21, the update processing device 80 uses the plurality of attribute processing units D based on the imaging data and the analysis result information transmitted from the plurality of attribute processing units D of each allocation control device. Each analysis processing program of is updated. Then, the update processing device 80 transmits the updated analysis processing program to each allocation control device.

Therefore, the allocation control system 300 has a larger amount of imaging data analyzed by the update processing device 80 than a configuration including only one allocation control device, and thus the update accuracy of the control program by the update processing device 80 is improved. .. Therefore, the control program of the specialized processing unit C or the attribute processing unit D can be efficiently updated. Then, in the allocation control system 300, since a plurality of allocation control devices are connected to the update processing device 80 via the network N, for example, imaging data and imaging data from a plurality of allocation control devices whose installation locations greatly differ from each other. The analysis result and the like can be collected and utilized in the update processing device 80.

Incidentally, although FIG. 23 shows an example in which the allocation control system 300 includes the actuator unit 90, the present invention is not limited to this, and the actuator unit 90 may be an external configuration of the allocation control system 300. Other configurations and alternative configurations are the same as those of the first and second embodiments described above, and other effects are also the same as those of the first and second embodiments.

The embodiments described above are preferred specific examples of the allocation control device and the allocation control system, and the technical scope of the present invention is not limited to these aspects. For example, FIGS. 4 and 5 have been described on the assumption that the actuators Xb1 to Xb4 can output not only sound or voice but also ultrasonic waves according to the control signal from the specialized processing unit C, but the present invention is not limited to this. Not done. The actuator unit 90 includes, for example, one or more actuators X specialized in sound output, one or more actuators X specialized in sound output, one or more actuators X specialized in ultrasonic output, and the like. May be separately provided.

4 and 5, the actuators Xa1 to Xa4 have been described on the assumption that they are light emitting means of the same quality, but the present invention is not limited to this. The actuator unit 90 may separately include, for example, one or a plurality of actuators X including a light source such as an LED and one or a plurality of actuators X including a strobe.

4 and 5, a drone equipped with six motors M and six propellers P is illustrated as the flying object 500, but the present invention is not limited to this. The air vehicle 500 may be a drone equipped with five or less or seven or more motors M and propellers P. However, the number and arrangement of each actuator X, the connection relationship between each specialized processing unit C and each actuator X, and the like are not limited to the configurations of FIGS. 4 and 5, and can be appropriately changed. Further, the flying body 500 may be an airplane, a helicopter, an airship, a balloon, or the like.

18, 19, and 22 show an example in which the attribute processing unit D and the actuator X are associated with each other in a one-to-one relationship, but the present invention is not limited to this, and the attribute processing unit D controls a plurality of actuators X. It may be configured to output a signal. By doing so, it is possible to increase the variation of the output of the actuator unit 90 according to the attribute of the imaged data, such as performing a combination of sound or voice notification and light emission.

The update processing device 80 acquires the type data from each specialized processing unit C, and sets the control program of the specialized processing unit C of the type with low allocation frequency to the control program of the type with high allocation frequency according to the appearance frequency of the type. It may be rewritten to. With this configuration, for example, when the camera A connected to the allocation control device is arranged in a place where there are many birds, the specialized processing units C of the type “bird” can be increased afterwards. Further, even when the object included in the imaging data changes according to the day of the week, the time of day, etc., an appropriate combination of the plurality of specialized processing units C can be realized in a timely manner.

In each embodiment, one or a plurality of cameras A are illustrated as devices that output data to the allocation control device, but the present invention is not limited to this, and the allocation control device may use various sensing devices as devices that output data. The device may be connected. In this case, the allocation control device may adopt a configuration including one or a plurality of control units that perform arithmetic processing using the sensing data output from the sensing device. The control unit includes, for example, a CPU or GPU. That is, the allocation control device controls the operation of the one or more actuators X using the imaging data and the sensing data. Further, the allocation control device may be configured to control the operation of one or more actuators X in cooperation with an external device to which various sensing devices are connected. The external device is for performing arithmetic processing using sensing data output from various sensing devices.

Here, the allocation control device can also be applied to, for example, an automatic driving system of an automobile. In this case, each of the plurality of cameras A is arranged so as to image the inside or the outside of the automobile. Therefore, the plurality of cameras A output imaging data including information such as traffic signs, traffic lights, white lines on the road surface, vehicles ahead or behind, oncoming vehicles, motorcycles, bicycles, and pedestrians to the allocation control device. The allocation control device controls the operation of various actuators X that compose the automobile, based on the imaging data output from each of the plurality of cameras A. The allocation control device is connected to a sensing device such as a millimeter wave radar or LiDAR (Light Detection And Ranging) as a device that outputs data, and performs arithmetic processing using sensing data from these sensing devices. May be. In this case, the allocation control device controls the operation of various actuators X that compose the automobile, using the imaging data and the sensing data. Further, the allocation control device may control the operation of various actuators X constituting the automobile in cooperation with an external device to which a sensing device such as a millimeter wave radar or LiDAR is connected. However, the allocation control device may have a configuration capable of performing processing using GPS (Global Positioning System), and cooperates with an external device that performs processing using GPS to operate the various actuators X constituting the automobile. You may control.

By the way, the plurality of specific processing units B constituting the allocation control device may be formed as one chip, for example, as the SoC. Further, in the allocation control device, the allocation processing unit 20 may be formed by one chip, or the plurality of specialized processing units C may be formed by one chip. Further, the allocation control device may be entirely formed by one chip. 18, FIG. 19, FIG. 21, and FIG. 22, the allocation control device may form the plurality of attribute processing units D by one chip, and the attribute allocation unit 70 and the plurality of attribute processing units may be formed. You may combine with D and may form one chip. The control unit of the specific processing unit B may be formed as an SoC including a CPU or GPU. Further, the specific processing unit B may be mounted inside the camera A. In each embodiment, the names of types and attributes are illustrated in parentheses, but these are for convenience of description, and the names are not limited. The line of caution can also be changed as appropriate.

10, 110 allocation control device, 20 allocation processing unit, 30 allocation unit, 41 first communication unit, 42 second communication unit, 43 storage unit, 43a memory means, 43b storage means, 431 learning program, 432 analysis processing program, 433 Update program, 44 control unit, 44a learning processing unit, 44b analysis processing unit, 44c update processing unit, 50 integrated update unit, 70 attribute assignment unit, 80 update processing device, 90 actuator unit, 90a display processing unit, 90b display unit, 95 display control device, 96 selection unit, 97 selection processing unit, 98 display processing unit, 99 display device, 100, 200, 300 allocation control system, 110A first control device, 110B second control device, 150A, 150B monitoring warning unit , 151 housing, 151a first housing, 151b second housing, 152 leg, 160A, 160B, 161, 161A, 161B traffic light, 400 image display system, 500 air vehicle, 510 main body, 600 traffic signal controller, A, A1 to An camera, B, B1 to Bn specific processing unit, C, C1 to Cm specialized processing unit, D, D1 to Dk attribute processing unit, M motor, N network, P propeller, X, X1 to Xk, Xa1 -Xa4, Xb1-Xb4 actuator, X7a, X7b, X8a, X9a display board, Y, Y1-Y6 analysis processing part.

Claims (14)

An assignment processing unit connected to the camera,
A plurality of specialized processing units connected to the allocation processing unit and corresponding to any of a plurality of preset types;
The allocation processing unit,
A specification processing unit that is connected to the camera, specifies the type of a main object included in the image pickup data input from the camera as the type, and outputs the image pickup data together with the type data indicating the specified type.
An allocation control device comprising: a specialized processing unit corresponding to the type indicated by the type data output from the specific processing unit; and an allocation unit that outputs the imaging data output together with the type data. An assignment processing unit connected to the camera,
A plurality of specialized processing units connected to the allocation processing unit and corresponding to any of a plurality of preset types;
The allocation processing unit,
A specification processing unit that is connected to the camera, identifies the main age group of the person included in the imaging data input from the camera as the type, and outputs the imaging data together with the type data indicating the identified type.
An allocation control device comprising: a specialized processing unit corresponding to the type indicated by the type data output from the specific processing unit; and an allocation unit that outputs the imaging data output together with the type data. The specialized processing unit,
When the imaging data is output from the allocation unit, the imaging data is discriminated as to which of the plurality of attributes in the type specified by the specification processing unit that is the output source, and the determined attribute The allocation control device according to claim 2, wherein the allocation control device executes a process according to. The specialized processing unit,
The allocation control device according to claim 3, wherein it is determined whether or not a relatively large number of walking people are included in the imaging data as the attribute. The specialized processing unit,
When it is determined that the imaged data includes a relatively large number of people walking, it is determined whether or not the number of people included in the imaged data is equal to or more than a first threshold value, and it is different depending on the determination result. The allocation control device according to claim 4, which executes a process. The specialized processing unit,
When it is determined that the imaged data does not include a relatively large number of people walking, it is determined whether the number of people included in the imaged data is equal to or more than a second threshold value, and according to the determination result. The allocation control device according to claim 3 or 4, which executes different processes. An assignment processing unit connected to the camera,
A plurality of specialized processing units connected to the allocation processing unit and corresponding to any of a plurality of preset types;
The allocation processing unit,
It is connected to the camera and is specified as the type whether or not the imaging data input from the camera includes a relatively large number of people walking, and the imaging data is acquired together with the type data indicating the specified type. A specific processing unit to output,
An allocation control device comprising: a specialized processing unit corresponding to the type indicated by the type data output from the specific processing unit; and an allocation unit that outputs the imaging data output together with the type data. The specialized processing unit,
When the imaging data is output from the assigning unit, the imaging data is discriminated as to which of the plurality of attributes in the type specified by the specification processing unit that is the output source, and the determined attribute The allocation control device according to claim 7, which executes a process according to the above. The specialized processing unit,
The allocation control device according to claim 8, wherein the attribute is to determine a main age group of a person included in the imaging data. The specialized processing unit,
After determining the main age group of the persons included in the imaged data, it is determined whether the number of persons included in the imaged data is equal to or more than a threshold value, and different processing is executed according to the result of the determination. The allocation control device according to claim 8. An allocation control device according to any one of claims 3 to 6 and 8 to 10, and a display device,
The specialized processing unit,
An allocation control system for displaying a moving image or a still image according to the determined attribute on the display device. The specific processing unit connected to the camera
The type of the main object included in the imaging data input from the camera is specified as the type of the imaging data,
Output the imaging data together with the type data indicating the identified type,
The allocation unit connected to the plurality of specialized processing units corresponding to any of the plurality of types,
An allocation control method, which outputs the imaging data output together with the type data to the specialized processing unit corresponding to the type indicated by the type data output from the specific processing unit. The specific processing unit connected to the camera
The main age group of the person included in the imaging data input from the camera is specified as the type of the imaging data,
Output the imaging data together with the type data indicating the identified type,
The allocation unit connected to the plurality of specialized processing units corresponding to any of the plurality of types,
An allocation control method, which outputs the imaging data output together with the type data to the specialized processing unit corresponding to the type indicated by the type data output from the specific processing unit. The specific processing unit connected to the camera
Whether or not a relatively large number of people walking are included in the imaging data input from the camera is specified as a type of the imaging data,
Output the imaging data together with the type data indicating the identified type,
The allocation unit connected to the plurality of specialized processing units corresponding to any of the plurality of types,
An allocation control method, which outputs the imaging data output together with the type data to the specialized processing unit corresponding to the type indicated by the type data output from the specific processing unit.

Images