JP7402779B2 - Program, information processing method, and information processing system - Google Patents

Description

The present disclosure relates to a program, an information processing method, and an information processing system.

2. Description of the Related Art A management system for a waste treatment plant is known that guides garbage trucks, so-called garbage collection trucks, to a garbage treatment facility. In the waste treatment plant management system described in Patent Document 1, the waste treatment facility is provided with a device such as a hopper for receiving garbage and a parking space for a garbage truck. A garbage truck loaded with garbage enters the garbage treatment facility and stops at a parking space. The garbage loaded on the garbage truck is received into the hopper. A maximum number of garbage trucks that can enter a garbage processing facility is set, for example, the same number as the number of parking spaces. The management system allows garbage trucks to enter the garbage treatment facility when the number of garbage trucks parked in the parking space is less than the upper limit number. The management system does not allow garbage trucks to enter the garbage treatment facility when the number of garbage trucks parked in the parking space is at the upper limit.

Japanese Patent Application Publication No. 2018-167963

In the above-mentioned waste treatment plant management system, it is possible to prevent more than the upper limit number of garbage trucks from entering the garbage treatment facility. However, it is assumed that a person brings garbage to a waste treatment facility in a vehicle other than a garbage truck such as a general vehicle, that is, a vehicle other than a garbage truck enters the garbage treatment facility. Since garbage discharge methods may differ depending on the type of vehicle, for example, a garbage truck or a general vehicle, parking spaces may be provided depending on the type of vehicle being parked. If the parking spaces differ depending on the type of vehicle, for example, if the parking space for garbage trucks and the parking space for general vehicles are different, determine whether to allow the vehicle to enter the garbage processing facility. When doing so, it is necessary to consider the type of vehicle.

An object of the present disclosure is to provide a program and the like that can determine whether or not to permit entry of a vehicle into a waste disposal facility, depending on the type of vehicle.

A program according to one aspect of the present disclosure provides, in a waste processing facility where multiple types of vehicles including garbage trucks enter, first image data of the vehicle entering the waste processing facility, and an image of the inside of the waste processing facility. acquires vehicle image data including at least one of the second image data in which the vehicle is imaged, and inputs the acquired vehicle to a learning model that is trained to output the type of vehicle when the vehicle image data is input. The type of vehicle is acquired by inputting image data, and based on the relationship between the acquired type of vehicle and the parking space provided in the garbage treatment facility according to the type of vehicle, the garbage treatment facility A computer is caused to execute a process of determining whether or not entry of the vehicle is permitted.

An information processing method according to an aspect of the present disclosure provides, in a garbage treatment facility where multiple types of vehicles including garbage trucks enter, first image data of the vehicle entering the garbage treatment facility; Vehicle image data including at least one of second image data in which the vehicle in the facility is imaged is acquired, and the acquired vehicle image data is input to a learning model that is trained to output the type of vehicle when the vehicle image data is input. The vehicle image data is input to obtain the vehicle type, and based on the relationship between the vehicle type and the parking space provided in the garbage processing facility according to the vehicle type, the garbage A computer is caused to execute a process of determining whether or not to permit the vehicle to enter the processing facility.

An information processing system according to an aspect of the present disclosure provides, in a garbage processing facility where multiple types of vehicles including garbage trucks enter, first image data in which the vehicle entering the garbage processing facility is imaged; and the garbage processing facility. an image data acquisition unit that acquires vehicle image data including at least one of second image data in which the vehicle in the facility is imaged; and a learning unit that is trained to output the type of vehicle when vehicle image data is input. a type acquisition unit that inputs the acquired vehicle image data into the model and acquires the type of the vehicle; and a determination unit that determines whether or not to permit the vehicle to enter the waste treatment facility based on the relationship with the parking space.

According to the present disclosure, it is possible to determine whether or not to permit entry of a vehicle into a waste disposal facility depending on the type of vehicle.

1 is a schematic diagram of a garbage treatment plant having a garbage treatment facility. FIG. 1 is a block diagram showing a configuration example of an information processing system. FIG. 3 is a schematic diagram showing an example of vehicle image data. FIG. 2 is an explanatory diagram illustrating the configuration of a learning model. 12 is a flowchart illustrating a determination process performed by a control unit to determine whether or not to permit entry of a vehicle. 7 is a flowchart illustrating a determination process performed by a control unit according to a second embodiment. FIG. 3 is a schematic diagram showing a garbage disposal plant according to Embodiment 3. It is a conceptual diagram which shows the example of the content of detection part DB. 12 is a flowchart illustrating a determination process performed by a control unit in Embodiment 3. FIG. 3 is a schematic diagram showing a garbage disposal plant according to Embodiment 4. 12 is a flowchart illustrating a determination process performed by a control unit in Embodiment 4. FIG. 3 is a block diagram illustrating a configuration example of an information processing system according to a fifth embodiment. FIG. 2 is a schematic diagram showing an example of in-facility image data. 12 is a flowchart illustrating a notification process performed by a control unit in Embodiment 5 to notify of the approach of a person and a vehicle or a specific device. 12 is a block diagram showing a configuration example of an information processing system according to a modification of the fifth embodiment. FIG. FIG. 12 is a block diagram showing a configuration example of an information processing system according to a sixth embodiment. FIG. 2 is a conceptual diagram showing an example of contents of a garbage type information DB. 12 is a flowchart illustrating a notification process performed by a control unit in Embodiment 6 to notify the type of garbage loaded on a vehicle scheduled to enter. FIG. 7 is a schematic diagram of a garbage disposal plant according to Embodiment 7. 12 is a flowchart illustrating a determination process performed by a control unit according to a seventh embodiment.

(Embodiment 1)
Embodiments of the present invention will be described below based on the drawings. FIG. 1 is a schematic diagram of a waste treatment plant 100 having a waste treatment facility 1. As shown in FIG. Illustration of the ceiling of the waste treatment facility 1 is omitted. The waste treatment facility 1 receives, for example, household waste, non-burnable waste, or plastic waste. A plurality of types of vehicles 2 run inside the garbage disposal site 100. The plurality of types of vehicles 2 include a garbage truck that collects garbage, a general vehicle, a forklift, a wheel loader, and a hand truck. In this embodiment, an example will be described in which two types of vehicles 2, a garbage truck and a general vehicle, are used. Note that the types of vehicles 2 may be three or more types. For example, the type of vehicle 2 may be a garbage truck, a general vehicle, or another vehicle. Other vehicles are vehicles 2 other than garbage trucks and general vehicles, such as forklifts, wheel loaders, and hand trucks.

The vehicle 2 is loaded with the above garbage. The garbage treatment plant 100 includes an information processing device 3 that determines whether or not to permit the vehicle 2 to enter the garbage treatment facility 1. Note that a method for determining whether or not to permit entry of the vehicle 2 will be described later. When the vehicle 2 is permitted to enter the garbage treatment facility 1, the vehicle 2 enters the garbage treatment facility 1 to carry in garbage.

A temporary stop line 11 is provided in the waste disposal facility 100 at a position outside the waste disposal facility 1 and close to the entrance 10 of the waste disposal facility 1. In FIG. 1, one vehicle 2 is temporarily stopped and waiting at a stop line 11 in order to enter the waste treatment facility 1. One vehicle 2 is waiting behind the vehicle 2 waiting at a stop line 11. In other words, two vehicles 2 are waiting in a line near the entrance 10. In the example of FIG. 1, the vehicle 2 waiting at the stop line 11 is a garbage truck. The vehicle 2 behind the vehicle 2 is a general vehicle.

The garbage disposal site 100 is provided with a first camera 5 that captures an image of a vehicle 2 waiting at a stop line 11. The vehicle 2 waiting at the stop line 11 is the vehicle 2 entering the garbage treatment facility 1. For example, the first camera 5 is a surveillance camera and continuously captures images. In FIG. 1, the first camera 5 is attached to the outer wall of the waste treatment facility 1. The first camera 5 only needs to be able to image the vehicle 2 entering the waste disposal facility 1, and may be attached to a location other than the outer wall of the waste disposal facility 1. The orientation of the first camera 5 is fixed, and the range imaged by the first camera 5 is also fixed. Hereinafter, the image data captured by the first camera 5 of the vehicle 2 entering the waste disposal facility 1 will also be referred to as first image data. The first image data may be a still image of one frame or a moving image of several frames. Note that the first camera 5 only needs to be able to image the vehicle 2 entering the waste disposal facility 1, and the orientation of the first camera 5 and the range imaged by the first camera 5 may not be fixed. For example, the first camera 5 may be changeable in direction.

An entrance 10 of the waste disposal facility 1 is provided with a regulating device for regulating entry of the vehicle 2 into the waste disposal facility 1 . In the example of FIG. 1, the regulation device displays a display indicating whether or not the vehicle 2 is allowed to enter the waste disposal facility 1 based on the result of the determination as to whether or not to permit the vehicle 2 to enter the waste disposal facility 1, which will be described later. This is part 4. The display unit 4 is provided at a location visible from the waiting position of the vehicle 2 entering the waste disposal facility 1, for example, on the outer wall of the waste disposal facility 1 near the entrance 10. For example, the display unit 4 is a signal light that can be lit in blue and red. The display unit 4 lit in blue indicates that the vehicle 2 is permitted to enter. The display unit 4 lit in red indicates that the vehicle 2 is not permitted to enter. When the display unit 4 is lit in red, entry of the vehicle 2 is restricted. Note that the color that the signal light lights up is not limited to blue and red. The display unit 4 may display whether or not entry is permitted by turning on and off a signal light. For example, the display unit 4 may be configured to turn on when indicating that the vehicle 2 is not permitted to enter, and turn off when indicating that the vehicle 2 is permitted to enter.

The driver of the vehicle 2 entering the waste treatment facility 1 looks at the display section 4 and confirms whether the vehicle 2 is allowed to enter. If the vehicle 2 is permitted to enter, the driver causes the vehicle 2 to enter the waste treatment facility 1. If the vehicle 2 is not permitted to enter, the driver causes the vehicle 2 to wait until the vehicle 2 is permitted to enter. Note that the display unit 4 is not limited to a signal light, and may be a display, for example. The display displays symbols such as "〇" and "x" or characters such as "permitted" and "not permitted" depending on whether or not entry is permitted. The regulating device is not limited to the display unit 4, and may be a barrier or a shutter, for example. A barrier or shutter opens or closes depending on whether or not entry is permitted. When the barrier or shutter is in the closed state, entry of the vehicle 2 is restricted.

The waste treatment facility 1 has an entrance 10 and an exit 12 located opposite to it. A plurality of types of parking spaces 6 are provided in the garbage processing facility 1 according to the types of vehicles 2. In this embodiment, the number of parking spaces 6 is four. Garbage trucks stop at three of the four parking spaces 6. General vehicles stop at the remaining one parking space 6. Therefore, the upper limit of the number of garbage trucks that can enter the garbage processing facility 1 is three. The upper limit of the number of general vehicles that can enter the waste treatment facility 1 is one. For example, the upper limit of the number of vehicles 2 that can enter these garbage processing facilities is stored in the information processing device 3. Note that the number of parking spaces 6 is not limited to four. A plurality of parking spaces 6 for general vehicles may be provided.

The four parking spaces 6 are arranged side by side in the direction in which the entrance 10 and the exit 12 face each other. Hereinafter, among the four parking spaces 6, the parking space 6 closest to the entrance 10 in the above direction is the first parking space 6a, the second parking space 6 closest to the entrance 10 is the second parking space 6b, and the third parking space 6 is the entrance 10. The parking space 6 closest to is also referred to as a third parking space 6c. Among the four parking spaces 6, the parking space 6 furthest from the entrance 10 in the above direction is also referred to as a fourth parking space 6d. In FIG. 1, a first parking space 6a, a second parking space 6b, and a third parking space 6c are parking spaces 6 where garbage trucks stop. The fourth parking space 6d is a parking space 6 where general vehicles stop. In FIG. 1, the vehicle 2 is parked in the third parking space 6c and the fourth parking space 6d. The vehicle 2 parked in the third parking space 6c is a garbage truck. The vehicle 2 parked in the fourth parking space 6d is a general vehicle. Note that the parking spaces 6 do not have to be arranged side by side.

The waste treatment facility 1 is provided with a hopper 13 for receiving waste. In FIG. 1, the hopper 13 is arranged near a first parking space 6a, a second parking space 6b, and a third parking space 6c. When the vehicle 2 is a garbage truck, after entering the garbage treatment facility 1, the vehicle 2 stops at one of the first parking space 6a, the second parking space 6b, and the third parking space 6c. The garbage loaded on the vehicle 2 is received into a hopper 13. After the garbage is received in the hopper 13, the vehicle 2 exits the garbage treatment facility 1 via the exit 12.

The waste treatment facility 1 is provided with a dumping box 14 for throwing waste into a hopper 13. In FIG. 1, the dumping box 14 is located near the hopper 13 and near the fourth parking space 6d. If the vehicle 2 is a general vehicle, after entering the waste treatment facility 1, it stops at the fourth parking space 6d. The driver of the vehicle 2 places the garbage loaded on the vehicle 2 in the dumping box 14. The dumping box 14 is driven by an operation by a worker of the waste treatment facility 1 and throws the placed waste into the hopper 13. After the driver places the garbage in the dumping box 14, the vehicle 2 exits the garbage treatment facility 1 via the exit 12. Note that an occupant of the vehicle 2 other than the driver may place the garbage in the dumping box 14.

For example, the garbage processing facility 1 may be provided with a container for storing garbage instead of or in addition to the dumping box 14. For example, the container is placed near the fourth parking space 6d. The driver of a regular vehicle throws the garbage into the container.

Although the inlet 10 and the outlet 12 are provided separately, one of the inlet 10 and the outlet 12 may also serve as the other of the inlet 10 and the outlet 12. For example, the vehicle 2 may enter and exit the waste disposal facility 1 via the entrance 10. At this time, the outlet 12 may not be provided.

The waste treatment facility 1 is provided with a second camera 15 that captures an image of the vehicle 2 inside the garbage treatment facility 1 . For example, the second camera 15 is a surveillance camera and continuously captures images. For example, the second camera 15 is attached to the inner wall of the waste treatment facility 1. Hereinafter, the image data captured by the second camera 15 of the vehicle 2 inside the waste disposal facility 1 will also be referred to as second image data. Like the first image data, the second image data may be a still image of one frame or a moving image of several frames.

The second camera 15 is fixed at a position and orientation that allows it to capture images of the vehicles 2 parked in the four parking spaces 6 at once. Therefore, the range imaged by the second camera 15 is fixed. The second image data includes at least one of the first parking space 6a and the vehicle 2 stopping at the first parking space 6a, and at least one of the second parking space 6b and the vehicle 2 stopping at the second parking space 6b. . Furthermore, the second image data includes at least one of the third parking space 6c and the vehicle 2 stopping at the third parking space 6c, and at least one of the fourth parking space 6d and the vehicle 2 stopping at the fourth parking space 6d. Take a picture. Note that the second camera 15 only needs to be able to image the vehicle 2 in the waste disposal facility 1, and may have a configuration in which the position and orientation of the second camera 15, and the range imaged by the second camera 15 are not fixed. For example, the direction of the second camera 15 may be changeable.

Note that the type of vehicle 2 imaged in the second image data may be only a garbage truck or only a general vehicle among a garbage truck and a general vehicle. Moreover, the type of vehicle 2 imaged in the second image data may be both a garbage truck and a general vehicle.

In FIG. 1, an information processing device 3 is provided within a waste treatment facility 1. Note that the information processing device 3 may be provided outside the waste treatment facility 1. FIG. 2 is a block diagram showing a configuration example of the information processing system S. As shown in FIG. The information processing system S includes an information processing device 3 and determines whether or not to permit the vehicle 2 to enter the waste treatment facility 1.

The information processing device 3 is connected to a first camera 5 , a second camera 15 , a display section 4 , and a speaker 16 provided in the garbage disposal facility 1 . Note that the speaker 16 is not shown in FIG. 1. In FIG. 1, connection lines connecting the information processing device 3, the first camera 5, the second camera 15, and the display unit 4 are not shown. The information processing device 3 acquires first image data from the first camera 5 and acquires second image data from the second camera 15. The information processing device 3 includes a learning model M1, which will be described later, that is trained to output the type of vehicle 2 when vehicle image data including at least one of first image data and second image data is input.

The information processing device 3 inputs the acquired first image data into the learning model M1 to acquire the type of vehicle 2 scheduled to enter the waste treatment facility 1. A vehicle 2 scheduled to enter the garbage disposal facility 1 is a vehicle 2 waiting at a temporary stop line 11. The information processing device 3 inputs the acquired second image data into the learning model M1 to acquire the type of vehicle 2 in the waste treatment facility 1.

Although the information processing device 3 has the learning model M1 in this embodiment, a device other than the information processing device 3 may have the learning model M1. For example, each of the first camera 5 and the second camera 15 may have the learning model M1. In this case, the first camera 5 inputs the captured first image data to the learning model M1 to obtain the type of vehicle 2 scheduled to enter, and outputs the obtained type of vehicle 2 scheduled to enter to the information processing device 3. do. The second camera 15 inputs the captured second image data into the learning model M1 to acquire the type of vehicle 2 in the garbage treatment facility 1, and the information processing device inputs the acquired type of vehicle 2 in the garbage treatment facility 1. Output to 3. The information processing device 3 acquires the type of vehicle 2 scheduled to enter from the first camera 5, and acquires the type of vehicle 2 in the waste treatment facility 1 from the second camera 15.

For example, a cloud server (not shown) that can communicate with the information processing device 3 may have the learning model M1. In this case, the information processing device 3 outputs the acquired first image data and second image data to the cloud server. The cloud server acquires the output first image data and second image data. The cloud server inputs the acquired first image data into the learning model M1 to acquire the type of vehicle 2 scheduled to enter, and outputs the acquired type of vehicle 2 scheduled to enter to the information processing device 3. The cloud server also inputs the acquired second image data into the learning model M1 to acquire the type of vehicle 2 in the garbage treatment facility 1, and inputs the acquired type of vehicle 2 in the garbage treatment facility 1 to the information processing device 3. Output to. The information processing device 3 acquires the type of vehicle 2 scheduled to enter and the type of vehicle 2 in the waste treatment facility 1 from the cloud server.

Although the details will be described later, the information processing device 3 specifies the type of the parking space 6 in which the vehicle 2 is not parked, among the four parking spaces 6, based on the acquired type of the vehicle 2 in the garbage processing facility 1. Hereinafter, the parking space 6 where the vehicle 2 is not parked will also be referred to as an empty parking space. The information processing device 3 determines whether or not to permit entry based on the type of the specified empty parking space and the acquired type of the vehicle 2 scheduled to enter the garbage disposal facility 1.

When permitting the vehicle 2 to enter the garbage disposal facility 1, the information processing device 3 outputs information indicating that entry is permitted to the display unit 4. Hereinafter, information indicating that entry is permitted will also be referred to as entry permission information. The display unit 4 receives the admission permission information and lights up in blue. The driver of the vehicle 2 scheduled to enter can be notified that entry is permitted.

When permitting the vehicle 2 to enter the waste disposal facility 1, the information processing device 3 outputs entry permission information including audio data of a voice such as "Vehicle is entering" to the speaker 16, and outputs the voice to the speaker 16. It may also be output to the speaker 16. It is possible to notify the workers of the waste treatment facility 1 that the vehicle 2 scheduled to enter is permitted to enter. For example, the above audio may be output from the speaker 16 at a volume that can be heard by the driver of the vehicle 2 scheduled to enter. The driver can be notified by voice that entry is permitted. For example, a waste disposal facility may be provided with two speakers 16, one for notifying the workers and the other for notifying the driver. Note that the information processing device 3 may output admission permission information to both the display unit 4 and the speaker 16.

The information processing device 3 may output the entry permission information to a terminal (computer), not shown, which is carried by a worker of the waste disposal facility 1 and can communicate with the information processing device 3. The terminal carried by the worker is, for example, a terminal dedicated to the waste treatment facility 1. For example, a worker at the waste treatment facility 1 carries the terminal with him when working. Hereinafter, the terminal owned by the worker will also be referred to as a facility terminal. The facility terminal acquires the output entry permission information and displays a sentence such as "Vehicle will enter" on a display unit such as a display provided on the facility terminal.

The information processing device 3 may output admission permission information including the above audio data to the facility terminal. For example, a facility terminal has a speaker. When the information processing device 3 outputs the above admission permission information to the facility terminal, the facility terminal acquires the output admission permission information and transmits the audio of the audio data included in the acquired admission permission information to the speaker of the facility terminal. Output. It is possible to notify the worker of the entry of the vehicle 2. For example, the facility terminal may be connected to earphones worn by workers. In this case, the facility terminal outputs the audio of the audio data to the earphone.

When the information processing device 3 does not permit the vehicle 2 to enter the waste treatment facility 1, it outputs information indicating that entry is not permitted to the display unit 4. Hereinafter, information indicating that entry is not permitted will also be referred to as entry non-permission information. The display unit 4 receives the entry denial information and lights up in red. The driver of the vehicle 2 scheduled to enter can be notified that entry is not permitted.

When not allowing the vehicle 2 to enter the waste disposal facility 1, the information processing device 3 outputs entry denial information including audio data such as "Vehicles are not allowed to enter" to the speaker 16, and outputs the audio to the speaker 16. It may also be output to the speaker 16. Further, the information processing device 3 may output entry denial information to the facility terminal.

The display unit 4 (regulatory device) and the speaker 16 are included in a result notification unit that notifies the driver of the vehicle 2 or the driver and the worker of the vehicle 2 of the determination result of whether or not to permit entry. . When entry permission information or entry non-permission information is output to the facility terminal as described above, the facility terminal is included in the result notification section. The speaker 16 corresponds to an audio output section. Voices such as “vehicles will enter” and “vehicles are not allowed to enter” correspond to voices related to determination results. As a method for notifying the determination result, a method that can be visually and audibly perceived by a person as described above is used, but the method for notifying the determination result is not limited to the above-mentioned example. The method of notifying the determination result may be, for example, a method in which a person perceives it using five senses other than visual and auditory senses.

The information processing device 3 is a dedicated or general-purpose computer. The information processing device 3 may be a server computer. The information processing device 3 includes a control section 30, a storage section 31, and an input/output section 32. The control section 30, the storage section 31, and the input/output section 32 are connected to each other. The input/output unit 32 is an input/output interface connected to the first camera 5 , second camera 15 , display unit 4 , and speaker 16 .

The control unit 30 includes arithmetic processing devices such as one or more CPUs (Central Processing Units) or GPUs (Graphics Processing Units). It also has a memory such as ROM (Read Only Memory) or RAM (Random Access Memory). The control unit 30 reads out a program P, which will be described later, and executes various processes. The control section 30 communicates with the first camera 5, the second camera 15, the display section 4, and the speaker 16 via the input/output section 32. For example, the information processing device 3 includes a wireless communication unit (not shown) for performing wireless communication, and connects the control unit 30 to at least one of the first camera 5, second camera 15, display unit 4, and speaker 16 via the wireless communication unit. A configuration may also be adopted in which both communicate with each other.

The storage unit 31 includes, for example, a hard disk or a nonvolatile memory such as an SSD (Solid State Drive). The storage unit 31 stores programs and data that the control unit 30 refers to, including the program P, and a learning model M1. Note that the storage unit 31 may be composed of a plurality of storage devices, or may be an external storage device connected to the information processing device 3. The program P and the learning model M1 stored in the storage unit 31 may be recorded on the recording medium A in a computer-readable manner. The storage unit 31 stores the program P and the learning model M1 read from the recording medium A by a reading device (not shown). The program P and the learning model M1 stored in the storage unit 31 are obtained by downloading the program P and the learning model M1 from an external computer (not shown) connected to a communication network (not shown) and storing them in the storage unit 31. Good too. Further, the number of information processing devices 3 is not limited to one. Processing may be performed in a distributed manner by a plurality of devices.

The control unit 30 functions as a first image data acquisition unit that acquires the first image data from the first camera 5 by executing the program P stored in the storage unit 31, and acquires the second image data from the second camera 15. It functions as a second image data acquisition unit that acquires data. The control unit 30 functions as an image data acquisition unit that acquires vehicle image data. The control unit 30 acquires first image data from the first camera 5 and acquires second image data from the second camera 15 via the input/output unit 32 .

By executing the program P stored in the storage unit 31, the control unit 30 inputs the first image data into the learning model M1 to acquire the type of vehicle 2 scheduled to enter the waste disposal facility 1. Functions as a type acquisition unit. Further, the control unit 30 functions as a second type acquisition unit that inputs the second image data into the learning model M1 and acquires the type of vehicle 2 in the waste treatment facility 1. The control unit 30 functions as a type acquisition unit that inputs vehicle image data into the learning model M1 and acquires the type of vehicle 2. The control unit 30 executes the program P stored in the storage unit 31 to permit the vehicle 2 to enter the waste treatment facility 1 based on the obtained type of the vehicle 2 and the relationship with the parking space 6. It functions as a determination unit that determines whether or not to do so.

Hereinafter, acquisition of the type of vehicle 2 by the control unit 30 will be explained. In acquiring the type of vehicle 2, the control unit 30 acquires vehicle image data and inputs the acquired vehicle image data to the learning model M1. A detection target in the vehicle image data is detected by the learning model M1 based on the vehicle image data. In this embodiment, the detection target is the vehicle 2. The detected type of vehicle 2 is output from the learning model M1. The control unit 30 acquires the type of vehicle 2 output from the learning model M1. Note that the detection target is not limited to the vehicle 2.

FIG. 3 is a schematic diagram showing an example of vehicle image data. FIG. 4 is an explanatory diagram illustrating the configuration of the learning model M1. FIG. 3 shows an example of second image data as vehicle image data. Two vehicles 2 are shown in the vehicle image data of FIG. Therefore, the number of detection targets in the vehicle image data of FIG. 3 is two. In FIG. 3, the vehicle 2 on the right side is a garbage truck. Vehicle 2 on the left is a general vehicle.

In the learning model M1, the relationship between the vehicle image data, the region of the detection target in the vehicle image data, and the type of the detection target is learned in advance. The detection target area is an area where the detection target is captured in the vehicle image data.

The learning model M1 is a model generated by machine learning including deep learning, and is a discriminator that outputs the type of detection target according to input of vehicle image data. The learning model M1 is configured by a neural network such as a convolutional neural network (CNN). For example, the learning model M1 uses RCNN (Regions with Convolutional Neural Network), which is a type of CNN.

When vehicle image data is input, the learning model M1 outputs a detection target area and a detection target type. Specifically, the learning model M1 outputs the region of the vehicle 2 and the type of the vehicle 2 when vehicle image data is input. For example, the learning model M1 outputs the probability that the detected vehicle 2 is a garbage truck and the probability that the detected vehicle 2 is a general vehicle in the output of the type of vehicle 2. When the probability that the vehicle is a garbage vehicle is greater than the probability that the vehicle is a general vehicle, the type of vehicle 2 is a garbage vehicle. When the probability that the vehicle is a general vehicle is greater than the probability that the vehicle is a garbage truck, the type of vehicle 2 is a general vehicle. When the detection target is not detected based on the vehicle image data, the learning model M1 outputs information indicating that the detection target is not detected. For example, the learning model M1 may output only the type of detection target among the areas and types of detection targets.

The learning model M1 is generated and trained in advance in the information processing device 3 or an external device (not shown). As described above, the learning model M1 is configured by a neural network. The learning model M1 includes an input layer that receives input of vehicle image data, more specifically, pixel values constituting the vehicle image data, an intermediate layer that extracts image features of the vehicle image data, and a detection target area and a detection target type. and an output layer that outputs. For example, when the learning model M1 is a CNN, the intermediate layer includes a convolution layer and a pooling layer. One or more neurons exist in the input layer, intermediate layer, and output layer.

The learning model M1 is not limited to RCNN. Any object detection algorithm such as YOLO (You Only Look Once), Fast RCNN, Faster RCNN, SSD (Single Shot Multibox Detector), or segmentation network may be used for the learning model M1. In this embodiment, the learning model M1 will be described as a neural network, but the learning model M1 is not limited to a neural network, and may be a SVM (Support Vector Machine), a Bayesian network, a regression tree, or a U-net. The learning model M1 may be constructed using a learning algorithm.

The learning model M1 is generated using a large amount of training data. The training data associates vehicle image data with the area and type of detection target shown in the vehicle image data. For example, the training data includes vehicle image data, area data indicating a region of a detection target depicted in the vehicle image data, and type data indicating the type of detection target depicted in the vehicle image data. When the number of detection targets is two or more, each of the plurality of types indicated by the type data is associated with a plurality of regions indicated by the region data.

In generating the learning model M1, the information processing device 3 or an external device inputs the vehicle image data included in the training data to the learning model M1, and acquires the region and type of the detection target output from the learning model M1. The information processing device 3 or the external device evaluates each of the acquired detection target regions and types based on the region indicated by the region data and the type indicated by the type data. The information processing device 3 or the external device updates the values of parameters such as the weighting coefficient and bias of the learning model M1 based on the evaluation result. For example, the information processing device 3 or the external device calculates the adjustment amount of the parameter using the error backpropagation method, and updates the value of the parameter based on the calculated adjustment amount.

Hereinafter, the determination of whether or not to permit the vehicle 2 to enter the waste treatment facility 1, which is performed by the control unit 30, will be described. The control unit 30 acquires second image data from the second camera. The control unit 30 inputs the acquired second image data to the learning model M1. The learning model M1 outputs the area of the vehicle 2 in the waste treatment facility 1 and the type of the vehicle 2. The control unit 30 acquires the type of vehicle 2 in the waste treatment facility 1 output from the learning model M1. For example, when the second image data of FIG. 3 is input to the learning model M1, two vehicles 2 are detected. The types of each of the two detected vehicles 2 are output from the learning model M1. The control unit 30 acquires the types of each of the two output vehicles 2.

The control unit 30 specifies the number of garbage trucks and the number of general vehicles in the waste disposal facility 1 based on the number of detection targets and the types of vehicles 2 in the waste disposal facility 1. In the example of FIG. 3, the number of garbage trucks and the number of general vehicles in the garbage processing facility 1 is one each.

Furthermore, the control unit 30 specifies the type of vacant parking space. As described above, the garbage processing facility 1 is provided with three parking spaces 6 where garbage trucks can park and one parking space 6 where general vehicles can park. It is assumed that the vehicle 2 in the waste treatment facility 1 stops at the parking space 6.

In the above example, the number of garbage trucks in the garbage treatment facility 1 is one, and the number of parking spaces 6 for garbage trucks is three, so there are two empty parking spaces for garbage trucks in the garbage treatment facility 1. do. Since the number of general vehicles in the waste treatment facility 1 is one and the parking space 6 for general vehicles is one, there is no vacant parking space for general vehicles in the garbage treatment facility 1. Therefore, the control unit 30 specifies that the type of vacant parking space is a vacant parking space related to a garbage truck.

For example, when both of the two vehicles 2 are specified as garbage trucks, the control unit 30 specifies the types of empty parking spaces as empty parking spaces for garbage trucks and empty parking spaces for general vehicles. When three vehicles 2 are detected and all of the three vehicles 2 are identified as garbage trucks, the control unit 30 changes the type of vacant parking space to a vacant parking space for general vehicles. Identify. When four vehicles 2 are detected, and among the four vehicles 2, one vehicle 2 is identified as a general vehicle and the remaining three vehicles 2 are identified as garbage trucks. , the control unit 30 specifies that there is no vacant parking space.

The control unit 30 specifies the type of vehicle 2 that can enter (hereinafter referred to as the vehicle type that can enter) based on the type of the specified vacant parking space. When the type of vacant parking space is a vacant parking space for a garbage truck, the control unit 30 specifies that the type of vehicle that can enter is a garbage truck. When the type of vacant parking space is a vacant parking space for a general vehicle, the control unit 30 specifies that the vehicle type that can enter is a general vehicle. When the types of vacant parking spaces are vacant parking spaces for garbage trucks and vacant parking spaces for general vehicles, the control unit 30 specifies that the types of vehicles that can enter are garbage trucks and general vehicles. If there is no vacant parking space, the control unit 30 specifies that there are no available vehicle types.

The control unit 30 acquires first image data from the first camera 5. As described above, the range imaged by the first camera 5 is fixed. For example, when the first camera 5 images a vehicle 2 scheduled to enter the garbage disposal facility 1, even if there is a vehicle 2 behind the vehicle 2 scheduled to enter, It is provided in a position where it does not image 2. Therefore, even if a plurality of vehicles 2 are waiting near the entrance 10, only the vehicle 2 scheduled to enter among the plurality of vehicles 2 is captured in the first image data.

The control unit 30 inputs the acquired first image data to the learning model M1. The learning model M1 outputs the area of the vehicle 2 scheduled to enter and the type of the vehicle 2. The control unit 30 acquires the type of vehicle 2 scheduled to enter, which is output from the learning model M1.

In the present embodiment, even if a plurality of vehicles 2 are waiting near the entrance 10, only the vehicle 2 scheduled to enter among the plurality of vehicles 2 is imaged in the first image data. Although an example will be described, a plurality of vehicles 2 may be imaged in the first image data. Therefore, the type of vehicle 2 imaged in the first image data may be both a garbage truck and a general vehicle. Moreover, the type of vehicle 2 imaged in the first image data may be only a garbage truck or only a general vehicle among a garbage truck and a general vehicle. For example, when a plurality of vehicles 2 are imaged in the first image data, the control unit 30 determines which vehicle 2 is scheduled to enter from among the plurality of vehicles 2 in the first image data, and uses the first image data and the learning model. The type of vehicle 2 scheduled to enter is acquired using M1.

For example, the first image data shows a plurality of vehicles 2 waiting in a line near the entrance 10. In this first image data, the first vehicle 2 in the row is the vehicle 2 scheduled to enter. For example, the first camera 5 captures images of the plurality of vehicles 2 waiting in a line so that the first vehicle 2 in the line is photographed closest to the front. By specifying the vehicle 2 that is closest to the front among the plurality of vehicles 2 that appear in the first image data, it is possible to determine which vehicle 2 is scheduled to enter. The vehicle 2 scheduled to enter may be determined by identifying the vehicle 2 that appears largest among the plurality of vehicles 2 that appear in the first image data. The vehicle 2 that appears the largest is, for example, the vehicle 2 in the largest (widest) region among the plurality of regions output when the above first image data is input to the learning model M1.

The control unit 30 allows the vehicle 2 to enter the garbage disposal facility 1 based on the vehicle type that can enter, which has been identified based on the type of vacant parking space as described above, and the type of vehicle 2 scheduled to enter, which has been acquired as described above. Determine whether or not to permit entry. Specifically, the control unit 30 determines whether or not the type of vehicle 2 scheduled to enter is one that allows entry. When the type of vehicle 2 scheduled to enter is an entry-permitted vehicle type, the control unit 30 allows the vehicle 2 to enter. If the type of vehicle 2 scheduled to enter is not an entry-permitted vehicle type, the control unit 30 does not permit the vehicle 2 to enter.

When permitting entry of the vehicle 2, the control section 30 outputs entry permission information to the display section 4 via the input/output section 32. When the control unit 30 does not permit the vehicle 2 to enter, the control unit 30 outputs entry non-permission information to the display unit 4 via the input/output unit 32.

FIG. 5 is a flowchart illustrating a determination process performed by the control unit 30 to determine whether or not to permit entry of the vehicle 2. For example, the control unit 30 starts the following process based on an operation by a worker of the waste disposal facility 1 on an operation unit (not shown) such as a keyboard or a touch panel included in the information processing device 3.

The control unit 30 acquires second image data from the second camera 15 (S11). As described above, the control unit 30 inputs the acquired second image data to the learning model M1 (S12), and acquires the type of vehicle 2 in the waste treatment facility 1 output from the learning model M1 (S13).

As described above, the control unit 30 specifies the number of garbage trucks and the number of general vehicles in the garbage treatment facility 1 based on the acquired types of vehicles 2 in the garbage treatment facility 1. Further, the control unit 30 specifies the type of vacant parking space based on the number of garbage trucks and the number of general vehicles in the specified garbage processing facility 1. The control unit 30 specifies the vehicle type that can enter based on the type of vacant parking space (S14).

The control unit 30 acquires first image data from the first camera 5 (S15). As described above, the control unit 30 inputs the acquired first image data to the learning model M1 (S16), and acquires the type of vehicle 2 scheduled to enter that is output from the learning model M1 (S17).

The control unit 30 determines whether the type of vehicle 2 scheduled to enter is an entry-permitted vehicle type (S18). If the type of vehicle 2 scheduled to enter is an entry-permitted vehicle type (S18: YES), the control unit 30 outputs admission permission information to the display unit 4 (S19), and ends the determination process. The display unit 4 receives the admission permission information and lights up in blue to display that admission is permitted.

If the type of vehicle 2 scheduled to enter is not an entry-permitted vehicle type (S18: NO), the control unit 30 outputs entry non-permission information to the display unit 4 (S181), and ends the determination process. The display unit 4 receives the entry non-permission information and lights up in red to display that entry is not permitted.

For example, the control unit 30 may return to the process of S11 after outputting the admission permission information or the admission non-permission information. Further, when the learning model M1 into which the first image data has been input outputs information indicating that no detection target has been detected, the control unit 30 may terminate the determination process or return to the process of S11. good.

For example, the control unit 30 performs a process for specifying the types of vehicles that can enter, including processes in S11, S12, S13, and S14, and a process for acquiring the type of vehicle 2 scheduled to enter, including processes in S15, S16, and S17. You can also do this in parallel. Moreover, after the process for acquiring the type of vehicle 2 scheduled to enter, the control unit 30 may perform a process for specifying the type of vehicle that can enter.

The information processing device 3 uses the first image data and the learning model M1 to acquire the type of vehicle 2 scheduled to enter, and uses the second image data and the learning model M1 to determine the type of vehicle 2 in the waste disposal facility 1. get. Based on the obtained relationship between the type of vehicle 2 and the parking space 6, the information processing device 3 determines whether or not to permit the vehicle 2 scheduled to enter. In determining whether or not to permit entry, the information processing device 3 identifies vehicle types that can enter based on the type of vacant parking space that is specified based on the type of vehicle 2 in the waste disposal facility 1. The information processing device 3 determines whether the acquired type of the vehicle 2 scheduled to enter is a vehicle type that is allowed to enter. Therefore, it can be determined whether or not to permit entry of the vehicle 2, taking into account the type of vehicle 2. Since entry of the vehicles 2 is permitted based on the type of vehicle 2 scheduled to enter and the type of vacant parking space, an excessive number of vehicles 2 will not enter the waste treatment facility 1. Therefore, occurrence of contact and collision between the vehicles 2 can be suppressed.

For example, when the type of vehicle 2 is a garbage truck, general vehicle, or other vehicle, the number of other vehicles does not need to be specified. For example, it is not determined whether or not other vehicles are permitted to enter. Other vehicles can enter the waste treatment facility 1 at will.

The information processing device 3 outputs information on whether or not to permit entry, including admission permission information and entry non-permission information, to the display unit 4 in accordance with the result of the determination as to whether or not to permit entry. The display unit 4 receives information as to whether or not entry is permitted, and displays whether or not entry is permitted. Since the display unit 4 is visible from the position where the vehicle 2 waiting to enter the waste treatment facility 1 is allowed to enter, it can inform the driver of the vehicle 2 waiting at the above position whether or not the vehicle 2 is allowed to enter.

For example, the information processing device 3 may be configured to output only one of the admission permission information and the admission non-permission information. In this case, the information processing device 3 outputs admission permission information when permitting admission, and outputs neither admission permission information nor admission non-permission information when not permitting admission. Alternatively, the information processing device 3 outputs neither admission permission information nor admission non-permission information when permitting entry, but outputs admission non-permission information when not permitting entry.

For example, sensors for detecting passage of the vehicle 2 may be provided at the entrance 10 and the exit 12. The sensor can communicate with the information processing device 3 and outputs the detection result to the information processing device 3. Since the vehicles 2 passing through the entrance 10 and the exit 12 are detected, the information processing device 3 can detect the total number of vehicles 2 inside the waste treatment facility 1 . For example, when the total number of vehicles 2 in the waste disposal facility 1 is four, that is, when the vehicle is full, the information processing device 3 performs the determination process after detecting the exit of one vehicle.

The detection target may include empty parking spaces. For example, the learning model M1 calculates, when vehicle image data is input, the probability that the detection target is a garbage truck, the probability that the detection target is a general vehicle, the probability that the detection target is an empty space related to the garbage truck, and the probability that the detection target is a garbage truck. Outputs the probability that the space is empty for general vehicles. The control unit 30 can input the second image data into the learning model M1 and acquire the type of vacant parking space from the learning model M1.

Since the range imaged by the second camera 15 is fixed as described above, it is possible to specify in advance the image coordinates indicating the area where the vehicle 2 stopping at each parking space 6 in the second image data is imaged. . The image coordinates indicating the above area are specified for each parking space 6. For example, the image coordinates indicating the above area are stored in the storage unit 31 in advance. The information processing device 3 acquires the area of the vehicle 2 in the garbage treatment facility 1 from the learning model M1, and determines whether the acquired area of the vehicle 2 in the garbage treatment facility 1 matches the above-mentioned area. You may. If the area of the vehicle 2 in the waste treatment facility 1 and the above area match, it is presumed that the vehicle 2 is parked in the parking space 6. If the area of the vehicle 2 inside the waste disposal facility 1 does not match the above-mentioned area, it is presumed that the vehicle 2 is traveling inside the waste disposal facility 1.

For example, the information processing device 3 acquires the second image data from one second camera 15, but may acquire the second image data from a plurality of second cameras 15. That is, the number of second cameras 15 is not limited, and may be one or more. An example in which two second cameras 15 are provided will be described below. For example, the second image data captured by one of the second cameras 15 includes at least one of the first parking space 6a and the vehicle 2 stopping at the first parking space 6a, and the second parking space 6b and the second parking space 6b. At least one of the vehicles 2 parked at is shown. The second image data captured by the other second camera 15 includes at least one of the third parking space 6c and the vehicle 2 stopping at the third parking space 6c, and the fourth parking space 6d and the vehicle 2 stopping at the fourth parking space 6d. At least one of the vehicles 2 is photographed. The information processing device 3 acquires two pieces of second image data: second image data captured by one of the second cameras 15 and second image data captured by the other second camera 15 . The information processing device 3 acquires the type of vehicle 2 in the waste treatment facility 1 using the acquired two second image data and the learning model M1. Even if one second camera cannot image the entire interior of the garbage disposal facility 1, the use of a plurality of second cameras makes it possible to image the entire interior of the garbage treatment facility 1.

For example, the learning model M1 may be trained to output the type of vehicle 2 that stops in the parking space 6 related to the above region when image data of the above region in the second image data is input. The types of vehicles 2 that stop in the parking space 6 related to the above area are, for example, garbage trucks, general vehicles, and no vehicles. The information processing device 3 can input the image data of the above region in the second image data into the learning model M1 to obtain the type of vehicle 2 in the parking space 6 regarding the above region.

For example, the storage unit 31 may store two learning models M1. One of the learning models M1 is trained to output the area of the vehicle 2 in the waste disposal facility 1 and the type of the vehicle 2 when the second image data is input. For learning one of the learning models M1, training data including second image data is used as vehicle image data. The other learning model M1 is trained to output the region of the vehicle 2 scheduled to enter and the type of the vehicle 2 when the first image data is input. Training data including the first image data is used as vehicle image data for learning the other learning model M1. The information processing device 3 inputs the second image data to one learning model M1 to obtain the type of vehicle 2 in the waste treatment facility 1, and inputs the first image data to the other learning model M1 to determine the entry schedule. The type of vehicle 2 is obtained.

The other learning model M1 is trained to output only the type of vehicle 2 scheduled to enter among the plurality of vehicles 2 even when a plurality of vehicles 2 are shown in the first image data. good. The types of vehicles 2 scheduled to enter at this time are, for example, garbage trucks, general vehicles, and no vehicles.

(Embodiment 2)
Among the configurations according to the second embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed explanation thereof will be omitted. The display unit 4 of the second embodiment is a signal light that can turn on and blink in blue and red, for example. In the garbage disposal site 100 of the second embodiment, the information processing device 3 uses only the second image data among the first image data and the second image data when determining whether to permit the vehicle 2 to enter.

In determining whether or not to permit entry of the vehicle 2, the control unit 30 of the second embodiment specifies the type of vacant parking space in the same way as the first embodiment. The control unit 30 determines whether or not to permit the vehicle 2 to enter the waste disposal facility 1 based on the type of the specified vacant parking space.

The control unit 30 allows the garbage truck and the general vehicle to enter when the types of the specified vacant parking space are both types of vacant parking space for garbage trucks and vacant parking spaces for general vehicles. . That is, in the above case, the control unit 30 allows the vehicle 2 to enter regardless of whether the vehicle 2 is a garbage truck or a general vehicle. When permitting the garbage truck and general vehicle to enter, the control unit 30 outputs first admission permission information to the display unit 4 indicating that the garbage truck and general vehicle are permitted to enter. The display unit 4 receives the first admission permission information and lights up in blue. The display unit 4, which is lit in blue, indicates that entry of the vehicles 2, both garbage trucks and general vehicles, is permitted.

The control unit 30 allows the garbage truck to enter when the type of the specified vacant parking space is a vacant parking space for a garbage truck. The control unit 30 outputs second entry permission information to the display unit 4 indicating that the garbage truck is permitted to enter. The display unit 4 receives the second admission permission information and blinks in blue. The display unit 4 blinking in blue indicates that entry of the garbage truck is permitted.

The control unit 30 allows the general vehicle to enter when the type of the specified vacant parking space is a vacant parking space for a general vehicle. The control unit 30 outputs third admission permission information to the display unit 4 indicating that entry of general vehicles is permitted. The display unit 4 receives the third admission permission information and blinks in red. The display section 4 flashing in red indicates that general vehicles are permitted to enter.

If there is no vacant parking space, the control unit 30 outputs entry denial information to the display unit 4. The display unit 4 receives the entry denial information and lights up in red. The display unit 4 lit in red indicates that the vehicle 2 is not permitted to enter. The first admission permission information, the second admission permission information, the third admission permission information, and the admission non-permission information are included in the information on whether or not to permit admission.

The colors in which the display unit 4 lights up and blinks are not limited to blue and red. For example, the display section 4 may be a signal light that can be lit in four colors. The display unit 4 receives any one of the four types of information: first admission permission information, second admission permission information, third admission permission information, and admission non-permission information, and the information varies depending on the received information. Lights up in color. The display section 4 is not limited to a signal light. For example, the display unit 4 may be a display. Depending on the above-mentioned information received, the display displays characters such as "Entry allowed", "Garbage truck entry allowed", "General vehicles allowed", and "Entry prohibited".

FIG. 6 is a flowchart illustrating the determination process performed by the control unit 30 of the second embodiment. For example, the control unit 30 starts the following process based on an operation of an operation unit (not shown) provided in the information processing device 3 by a worker of the waste treatment facility 1.

The control unit 30 performs the processes of S21, S22, and S23. The processes in S21, S22, and S23 are similar to the processes in S11, S12, and S13, so detailed explanations will be omitted. The control unit 30 specifies the type of vacant parking space as described above (S24). The control unit 30 determines whether or not the identified vacant parking space is both a vacant parking space for garbage trucks and a vacant parking space for general vehicles, that is, both types of vacant parking spaces. It is determined whether there is space (S25).

If there are vacant parking spaces of both types (S25: YES), the control unit 30 outputs the first admission permission information to the display unit 4 (S26), and ends the determination process. If there are both types of empty parking spaces, this is a case where the types of empty parking spaces are both types of empty parking spaces.

If there are no empty parking spaces of both types (S25: NO), the control unit 30 determines whether the type of the specified empty parking space is an empty parking space for a garbage vehicle, that is, there is no empty parking space for a garbage vehicle. It is determined whether there is one (S251). If there is a vacant parking space for the garbage truck (S251: YES), the control unit 30 outputs the second admission permission information to the display unit 4 (S252), and ends the determination process. If there is a vacant parking space for a garbage vehicle, the type of vacant parking space is a vacant parking space for a garbage vehicle.

If there is no vacant parking space for a garbage vehicle (S251: NO), the control unit 30 determines whether the type of the specified vacant parking space is a vacant parking space for a general vehicle, that is, there is a vacant parking space for a general vehicle. It is determined whether or not (S253). If there is a vacant parking space for a general vehicle (S253: YES), the control unit 30 outputs the third entry permission information to the display unit 4 (S254), and ends the determination process. If there is a vacant parking space for a general vehicle, the type of vacant parking space is a vacant parking space for a general vehicle.

If there is no vacant parking space for general vehicles (S253: NO), that is, if there is no vacant parking space, the control unit 30 outputs entry denial information to the display unit 4 (S255), and ends the determination process.

As described above, the control unit 30 outputs any one of the first admission permission information, the second admission permission information, the third admission permission information, and the admission non-permission information according to the type of vacant parking space. After outputting the information, the control unit 30 may return to the process of S21 instead of ending the determination process.

For example, the control unit 30 may output entry denial information to the display unit 4 when there are no vacant parking spaces for both types. That is, when there is no available parking space for general vehicles or at least one of available parking spaces for garbage trucks, the control unit 30 does not need to permit the vehicle 2 to enter the garbage processing facility.

In this embodiment, the information processing device 3 specifies the type of empty parking space, and determines whether or not to permit entry of the vehicle 2 based on the specified type of empty parking space. Since the parking spaces 6 are provided according to the types of vehicles 2, the types of parking spaces correspond to the types of vehicles 2. The type of vehicle 2 is taken into consideration when determining whether to permit entry. As in the first embodiment, an excessive number of vehicles 2 will not enter the waste treatment facility 1, so it is possible to suppress the occurrence of contact and collision between the vehicles 2.

In determining whether or not to permit entry of the vehicle 2, the information processing device 3 does not use the first image data, so the first camera 5 does not need to be provided at the garbage disposal site 100. Therefore, the number of equipment in the garbage treatment plant 100 can be reduced.

(Embodiment 3)
FIG. 7 is a schematic diagram showing a garbage disposal plant 100 according to the third embodiment. Among the configurations according to the third embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed explanation thereof will be omitted. In the garbage disposal site 100 of the third embodiment, the information processing device 3 does not use the second image data when determining whether to permit the vehicle 2 to enter. For example, the second camera 15 is not provided in the garbage treatment facility 1.

Each parking space 6 in the third embodiment is provided with one detection unit 17 that detects the vehicle 2 parked in the parking space 6. In the waste treatment facility 1 shown in FIG. 7, the number of detection units 17 is four. The detection unit 17 is, for example, a loop coil, and is buried in the parking space 6. The detection unit 17 is not limited to a loop coil, but may be a phototube, for example. The detection unit 17 and the information processing device 3 are connected via the input/output unit 32 and can communicate with each other. Note that illustration of a connection line between the detection unit 17 and the information processing device 3 is omitted. The information processing device 3 has a wireless communication unit for wireless communication, and may communicate with the detection unit 17 via the wireless communication unit. For example, a plurality of detection units 17 may be provided in one parking space 6.

Each detection unit 17 outputs a detection result to the information processing device 3. For example, when detecting a stopped vehicle 2, the detection unit 17 outputs a signal indicating that the vehicle 2 is stopped to the information processing device 3. If the detection unit 17 does not detect the stopped vehicle 2, it outputs a signal indicating that the vehicle 2 is not stopped to the information processing device 3.

For example, each detection unit 17 is assigned an ID (Identification) number. For example, a detection unit DB (Database) in which the ID number of the detection unit 17 and the parking space 6 in which the detection unit 17 with the ID number is provided is stored in association with each other. FIG. 8 is a conceptual diagram showing an example of the contents of the detection unit DB. The detection unit DB has an ID number column of the detection unit 17 and a parking space column. In the example of FIG. 8, the ID number 1 of the detection unit 17 is associated with the first parking space 6a. At this time, the detection unit 17 whose ID number is 1 is provided in the first parking space 6a.

The control unit 30 acquires the detection result output from the detection unit 17 via the input/output unit 32. The control unit 30 can identify the detection unit 17 that outputs the detection result. For example, the detection unit 17 outputs a signal indicating its own ID number along with the detection result. Based on the acquired detection result, the control unit 30 refers to the detection unit DB to identify the parking space 6 where the vehicle 2 is parked and the vacant parking space. Hereinafter, the parking space 6 where the vehicle 2 is parked will also be referred to as the parking space 6 in use.

For example, when the control unit 30 acquires a signal indicating that the vehicle 2 is stopped as a detection result from the detection unit 17 whose ID number is 1, the control unit 30 identifies the first parking space 6a as the parking space 6 in use. . When the control unit 30 acquires a signal indicating that the vehicle 2 is not stopped as a detection result from the detection unit 17 whose ID number is 1, it specifies the first parking space 6a as an empty parking space. The detection unit 17 does not have to output a signal indicating that the vehicle 2 is not stopped. For example, when the control unit 30 does not acquire a signal indicating that the vehicle 2 is stopped from the detection unit 17 whose ID number is 1, it specifies the first parking space 6a as an empty parking space.

As described above, the control unit 30 specifies, for each of the four parking spaces 6, whether the parking space 6 is in use or an empty parking space. The control unit 30 specifies the type of vacant parking space based on the specified parking space 6 in use and the vacant parking space.

When the four parking spaces 6 are in use, the control unit 30 specifies that there are no vacant parking spaces. If the vacant parking space is at least one of the first parking space 6a, the second parking space 6b, and the third parking space 6c, and the currently used parking space 6 is the fourth parking space 6d, the control unit 30: The type of vacant parking space is identified as a vacant parking space related to a garbage truck.

When the vacant parking spaces are at least one of the first parking space 6a, the second parking space 6b, and the third parking space 6c, and the fourth parking space 6d, the control unit 30 determines that the type of the vacant parking space is related to a garbage vehicle. A vacant parking space and a vacant parking space for general vehicles are identified. If the parking spaces in use are the first parking space 6a, the second parking space 6b, and the third parking space 6c, and the vacant parking space is the fourth parking space 6d, the control unit 30 determines the type of the vacant parking space. is a vacant parking space for general vehicles.

As in the first embodiment, the control unit 30 specifies the vehicle type that can enter based on the type of the acquired vacant parking space. As in the first embodiment, the control unit 30 acquires the type of vehicle 2 scheduled to enter. The control unit 30 determines whether or not to permit the vehicle 2 to enter the waste treatment facility 1 based on the type of vehicle that can enter, which is specified based on the type of vacant parking space, and the acquired type of the vehicle 2 that is scheduled to enter. do. Specifically, similarly to the first embodiment, the control unit 30 determines whether or not the type of vehicle 2 scheduled to enter is one that allows entry. The control unit 30 outputs admission permission information or admission non-permission information to the display unit 4 according to the determination result.

FIG. 9 is a flowchart illustrating the determination process performed by the control unit 30 of the third embodiment. For example, the control unit 30 starts the following process based on an operation of an operation unit (not shown) provided in the information processing device 3 by a worker of the waste treatment facility 1.

The control unit 30 acquires detection results from each detection unit 17 (S31). As described above, the control unit 30 specifies the type of vacant parking space based on the acquired detection result, and specifies the vehicle type that can enter based on the specified type of vacant parking space (S32). The control unit 30 performs the processes of S33, S34, and S35. The processes in S33, S34, and S35 are similar to the processes in S15, S16, and S17, so detailed explanations will be omitted.

The control unit 30 determines whether the type of vehicle 2 scheduled to enter is an entry-permitted vehicle type (S36). If the type of vehicle 2 scheduled to enter is an entry-permitted vehicle type (S36: YES), the control unit 30 outputs entry permission information to the display unit 4 (S37), and ends the determination process. If the type of vehicle 2 scheduled to enter is not an entry-permitted vehicle type (S36: NO), the control unit 30 outputs entry non-permission information to the display unit 4 (S361), and ends the determination process.

For example, the control unit 30 may return to the process of S31 after outputting the admission permission information or the admission non-permission information. For example, when the learning model M1 input with the first image data outputs information indicating that no detection target has been detected, the control unit 30 may end the determination process or return to the process of S31. .

The information processing device 3 acquires the detection result from the detection unit 17 and identifies an empty parking space based on the acquired detection result. Further, first image data is acquired, and the acquired first image data is input into the learning model M1 to acquire the type of vehicle 2 scheduled to enter. In determining whether or not to permit entry, the information processing device 3 determines whether the type of vehicle 2 scheduled to enter is a vehicle type that can enter, which is specified based on the type of vacant parking space. Therefore, it can be determined whether or not to permit entry of the vehicle 2, taking into account the type of vehicle 2. The information processing device 3 can determine whether or not to permit entry without using the second image data. Since an excessive number of vehicles 2 will not enter the waste treatment facility 1, it is possible to suppress the occurrence of contact and collision between the vehicles 2.

(Embodiment 4)
FIG. 10 is a schematic diagram showing a garbage disposal plant 100 according to the fourth embodiment. Among the configurations according to the fourth embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed explanation thereof will be omitted. The waste treatment facility 1 of the fourth embodiment is provided with the second camera 15 as in the first embodiment. A detection unit 17 is provided in each parking space 6 in the fourth embodiment. Since the detection unit 17 is similar to the detection unit 17 of the third embodiment, a detailed explanation will be omitted.

In the information processing device 3 of the fourth embodiment, the control unit 30 acquires the detection result from the detection unit 17, as in the third embodiment, and specifies the parking space 6 in use based on the acquired detection result.

The control unit 30 acquires second image data from the second camera 15. Since the range imaged by the second camera 15 is fixed as described above, it is possible to specify in advance the image coordinates indicating the area where the vehicle 2 stopping at each parking space 6 in the second image data is imaged. . The image coordinates indicating the above area are specified for each parking space 6. For example, the image coordinates indicating the above area are stored in the storage unit 31 in advance.

The control unit 30 inputs the second image data to the learning model M1 to obtain the type of vehicle 2 stopping at the parking space 6 in use. Specifically, when inputting the second image data to the learning model M1, the control unit 30 causes the learning model M1 to perform the above-mentioned operations on the area related to the specified parking space 6 in use among the above-mentioned areas in the second image data. Let detection occur. For example, the control unit 30 inputs image data of the area related to the specified in-use parking space 6 in the second image data to the learning model M1. At this time, the learning model M1 outputs the type of vehicle 2 that stops at the specified parking space 6 in use.

The control unit 30 specifies the type of vacant parking space based on the acquired type of the vehicle 2 described above. Similarly to the second embodiment, the control unit 30 determines whether or not to permit the vehicle 2 to enter the waste treatment facility 1 based on the type of the specified vacant parking space. Specifically, the control unit 30 displays any one of first admission permission information, second admission permission information, third admission permission information, and admission non-permission information to the display unit 4 according to the type of vacant parking space. Output. As in the second embodiment, the display unit 4 receives the above output information and lights up or blinks in blue or red.

FIG. 11 is a flowchart illustrating the determination process performed by the control unit 30 of the fourth embodiment. For example, the control unit 30 starts the following process based on an operation of an operation unit (not shown) provided in the information processing device 3 by a worker of the waste treatment facility 1.

The control unit 30 acquires detection results from each detection unit 17 (S41). As described above, the control unit 30 identifies the parking space 6 in use based on the acquired detection result (S42). The control unit 30 acquires second image data from the second camera 15 (S43).

As described above, the control unit 30 inputs the acquired second image data to the learning model M1 (S44), causes the learning model M1 to perform the above-mentioned detection on the area related to the specified parking space 6 in use, The type of vehicle 2 that stops in the currently used parking space 6 is acquired (S45).

As described above, the control unit 30 specifies the type of vacant parking space based on the acquired type of vehicle 2 (S46). The control unit 30 determines whether there are vacant parking spaces of both types (S47). If there are vacant parking spaces of both types (S47: YES), the control unit 30 outputs the first admission permission information to the display unit 4 (S48), and ends the determination process.

If there are no vacant parking spaces for both types (S47: NO), the control unit 30 determines whether there is any vacant parking space for the garbage truck (S471). If there is a vacant parking space for the garbage truck (S471: YES), the control unit 30 outputs the second admission permission information to the display unit 4 (S472), and ends the determination process.

When there is no vacant parking space for a garbage vehicle (S471: NO), the control unit 30 determines whether there is a vacant parking space for a general vehicle (S473). If there is a vacant parking space for a general vehicle (S473: YES), the control unit 30 outputs the third entry permission information to the display unit 4 (S474), and ends the determination process. If there is no vacant parking space for general vehicles (S473: NO), that is, if there is no vacant parking space, the control unit 30 outputs entry denial information to the display unit 4 (S475), and ends the determination process.

After outputting any one of the first admission permission information, the second admission permission information, the third admission permission information, and the admission non-permission information, the control unit 30 returns to the process of S41 instead of ending the determination process. It's okay. Further, the control unit 30 may output entry denial information to the display unit 4 when there are no vacant parking spaces of both types.

The control unit 30 inputs the second image data to the learning model M1, and causes the learning model M1 to perform detection on a region related to the parking space 6 in use. The learning model M1 can easily detect the vehicle 2 in the second image data. Accuracy regarding the types of vehicles 2 in the garbage disposal facility 1 can be improved.

(Embodiment 5)
FIG. 12 is a block diagram showing a configuration example of the information processing system S according to the fifth embodiment. Among the configurations according to the fifth embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed explanation thereof will be omitted. In the information processing device 3 of the fifth embodiment, the storage unit 31 stores a first learning model (same as the learning model of the first embodiment) M1 and a second learning model M2. The control unit 30 performs the above-described determination process using the first learning model M1. The second learning model M2 will be described later.

In the information processing device 3 of the fifth embodiment, the third camera 18 is connected to the input/output section 32. The third camera 18 is, for example, a surveillance camera, and images the inside of the garbage disposal facility 1. For example, the third camera 18 is attached to the ceiling of the waste treatment facility 1. The third camera 18 is arranged in such a direction that it can image the vehicle 2, the parking space 6, and the hopper 13.

Hereinafter, the image data captured by the third camera 18 inside the garbage processing facility 1 will also be referred to as in-facility image data. The in-facility image data may be a still image of one frame or a moving image of several frames. FIG. 13 is a schematic diagram showing an example of in-facility image data. The in-facility image data in FIG. 13 shows the vehicle 2, the parking space 6, the hopper 13, and the person 7. The person 7 is, for example, a worker working at the waste disposal facility 1 or a passenger of the vehicle 2.

The control unit 30 of the fifth embodiment communicates with the third camera 18 via the input/output unit 32. The control unit 30 acquires in-facility image data from the third camera 18 . For example, the information processing device 3 may include a wireless communication unit (not shown) for performing wireless communication, and the control unit 30 and the third camera 18 may communicate with each other via the wireless communication unit.

The number of third cameras 18 may be one, or two or more. In this embodiment, an example will be described in which the third camera 18 and the second camera 15 are different cameras, but the third camera 18 and the second camera 15 may be the same camera. In this case, the in-facility image data includes an image of the vehicle 2 in the waste disposal facility 1, so the in-facility image data corresponds to the second image data.

In the second learning model M2, the relationship between the in-facility image data and the detection target area and detection target type in the in-facility image data is learned in advance. Detection targets of the second learning model M2 include the vehicle 2, the person 7, and the hopper 13. Hopper 13 corresponds to a specific device within waste treatment facility 1 . The specific device is not limited to the hopper 13, and may be any device or equipment in the waste treatment facility 1 other than the hopper 13. A plurality of specific devices may be included in the detection target.

The control unit 30 inputs the acquired in-facility image data to the second learning model M2. When the in-facility image data is input to the second learning model M2, the second learning model M2 outputs the region of the detection target in the in-facility image data and the type of the detection target. For example, the types of detection targets are a garbage truck, a general vehicle, another vehicle, a specific device, and a person 7. The area to be detected is, for example, a bounding box, and is indicated by an image coordinate system. If the detection target is not detected based on the in-facility image data, the second learning model M2 outputs information indicating that the detection target is not detected.

The training data used in learning the second learning model M2 includes in-facility image data, and area data and type data regarding the in-facility image data. The learning method of the second learning model M2 is the same as that of the learning model M1 of the first embodiment, so a detailed explanation will be omitted.

The control unit 30 inputs the acquired in-facility image data to the second learning model M2. The second learning model M2 outputs the region of the detection target in the in-facility image data and the type of the detection target. The control unit 30 acquires the region of the detection target and the type of the detection target in the output in-facility image data.

When the person 7 is detected, the control unit 30 derives the distance between the detected person 7 and the vehicle 2 or a specific device. For example, the control unit 30 determines, as the distance between the person 7 and the vehicle 2, the shortest distance between the detection target area where the type is the person 7 and the detection target area where the type is a garbage truck, general vehicle, or other vehicle. Derived based on each detection target area. Further, the control unit 30 determines the shortest distance between the detection target area where the type is the person 7 and the detection target area where the type is the specific device as the distance between the person 7 and the specific device. Derived based on. For example, the distance per pixel in the in-facility image data specified based on the resolution and dimensions of the in-facility image data is stored in advance in the storage unit 31 for deriving the distance. The control unit 30 derives the distance by multiplying the number of pixels between each detection target area by the distance per pixel.

The control unit 30 determines whether the derived distance is less than or equal to a threshold value. The threshold value is stored in advance in the storage unit 31, for example. For example, the threshold value regarding the distance between the person 7 and the specific device may be different from the threshold value regarding the distance between the person 7 and the vehicle 2. For example, distances and thresholds may be indicated by number of pixels. If the distance is less than or equal to the threshold, the control unit 30 outputs approach information to the speaker 16. The approach information is information indicating the approach between the person 7 and the vehicle 2 or a specific device.

The speaker 16 receives the output approach information and outputs a voice such as "Please move away from the vehicle or device" or a warning sound, and tells the person 7 and the vehicle 2 or Notify that a specific device is approaching. For example, the above-mentioned voice or warning sound is included in the approach information. The approach information is stored in the storage unit 31. For example, the above-mentioned voice or warning sound may be stored in advance in the speaker. The speaker 16 corresponds to an approach notification section. Note that the approach notification section is not limited to the speaker 16.

The control unit 30 may output the approach information to the facility terminal instead of or in addition to the speaker 16. For example, the facility terminal receives the output approach information and causes the speaker of the facility terminal to output the above-mentioned voice or warning sound. The facility terminal may output the above-mentioned voice or warning sound to an earphone connected to the facility terminal. Further, when the facility terminal receives the output approach information, the facility terminal may display on the display section of the facility terminal that the person 7 and the vehicle 2 or a specific device are approaching. It is possible to notify a worker in the waste disposal facility 1 that the person 7 and the vehicle 2 or a specific device are approaching. In this case, the facility terminal corresponds to the approach notification section.

FIG. 14 is a flowchart illustrating a notification process performed by the control unit 30 of the fifth embodiment to notify of the approach of the person 7 and the vehicle 2 or a specific device. For example, the notification process is performed in parallel with the determination process. For example, the control unit 30 starts the following process based on an operation of an operation unit (not shown) provided in the information processing device 3 by a worker of the waste treatment facility 1.

The control unit 30 acquires in-facility image data from the third camera 18 (S51), and inputs the acquired in-facility image data to the second learning model M2 (S52). The second learning model M2 outputs the region and type of detection target in the in-facility image data. The control unit 30 acquires the area and type of detection target in the output in-facility image data (S53).

As described above, the control unit 30 derives the distance between the person 7 and the vehicle 2 or the specific device based on the acquired area and type of detection target (S54). For example, when the person 7 is not detected, the control unit 30 ends the notification process or performs the process of S51. As described above, the control unit 30 determines whether the derived distance is less than or equal to the threshold (S55).

If the distance is less than or equal to the threshold (S55: YES), the control unit 30 outputs approach information to the speaker 16 (S56), and ends the notification process. After outputting the approach information, the control unit 30 may return to the process of S51. If the distance is not less than or equal to the threshold (S55: NO), that is, if the distance is greater than the threshold, the control unit 30 ends the notification process. The control unit 30 may return to the process of S51 instead of ending the notification process.

In this embodiment, when the distance between the person 7 and the vehicle 2 or the specific device is less than or equal to the threshold, the person 7 is notified that the person 7 and the vehicle 2 or the specific device are approaching. Therefore, it is possible to ensure an appropriate distance between the person 7 and the vehicle 2 or a specific device. By ensuring an appropriate distance, for example, it is possible to prevent the person 7 from coming into contact with the vehicle 2. Furthermore, it is possible to prevent the person 7 from getting too close to the hopper 13.

The control unit 30 may use the first learning model M1 in the notification process. For example, the detection targets of the first learning model M1 are the vehicle 2, the person 7, and a specific device. In this case, the control unit 30 acquires the second image data from the second camera, inputs the second image data to the first learning model M1, and acquires the area and type of the detection target in the second image data. The control unit 30 derives the distance between the person 7 and the vehicle 2 or a specific device based on the acquired area and type of detection target.

(Modified example)
FIG. 15 is a block diagram showing a configuration example of an information processing system S according to a modification of the fifth embodiment. Among the configurations according to the modification of the fifth embodiment, the same components as those in the fifth embodiment are denoted by the same reference numerals, and detailed explanation thereof will be omitted. In the information processing device 3 of the modification of the fifth embodiment, the input/output unit 32 is connected to an interlock device 19, which will be described later. The control section 30 communicates with the interlock device 19 via the input/output section 32. Note that the information processing device 3 may include a wireless communication section (not shown) for performing wireless communication, and the control section 30 and the interlock device 19 may communicate with each other via the wireless communication section.

In the modified example, the detection targets of the second learning model M2 include the vehicle 2, the person 7, and the dumping box 14. The dumping box 14 corresponds to a specific device within the waste treatment facility 1. The damping box 14 is imaged by a third camera 18.

In the modified example, the type of detection target output from the second learning model M2 includes the type of person 7. The types of person 7 are, for example, ordinary people and workers. The general public is, for example, a passenger such as a driver of a general vehicle. Therefore, in the modified example, the types of detection targets output from the second learning model M2 are garbage trucks, general vehicles, other vehicles, dumping boxes 14, ordinary people, and workers. For example, in the waste treatment facility 1, workers are wearing work clothes or work hats. In this case, it is possible to distinguish between ordinary people and workers by whether or not they are wearing work clothes or the above-mentioned hat.

As in the fifth embodiment, the control unit 30 inputs the in-facility image data to the second learning model M2, and acquires the region of the detection target and the type of the detection target in the in-facility image data. When an ordinary person is detected, the control unit 30 derives the distance between the detected ordinary person and the dumping box 14 based on the acquired area and type of detection target. The control unit 30 determines whether the derived distance is less than or equal to a threshold value. If the distance is less than or equal to the threshold, the control unit 30 outputs approach information to the interlock device 19. Note that the control unit 30 may output approach information to the speaker 16 and the interlock device 19 when the above-mentioned distance is less than or equal to a threshold value.

The interlock device 19 is a device that makes an emergency stop to drive the damping box 14, and is built into the damping box 14, for example. Note that the interlock device 19 may be attached to the outside of the damping box 14. The interlock device 19 includes a control section 190, an input/output section 191, and an interlock mechanism 192 for stopping the driving of the damping box 14. The input/output unit 191 is an input/output interface connected to the information processing device 3.

The control unit 190 includes one or more CPUs, or an arithmetic processing device such as an FPGA (Field Programmable Gate Array). It also has a memory such as ROM or RAM. The control unit 190 communicates with the information processing device 3 via the input/output unit 191. For example, the control unit 190 acquires approach information output from the information processing device 3 via the input/output unit 191. Furthermore, the control unit 190 controls the driving of the interlock mechanism 192.

Upon acquiring the approach information, the control unit 190 determines whether the damping box 14 is being driven. The control unit 190 drives the interlock mechanism 192 when the damping box 14 is driven. That is, the control unit 190 drives the interlock mechanism 192 when the approach information is acquired and the damping box 14 is being driven. By driving the interlock mechanism 192, the driving of the damping box 14 is stopped. That is, when the distance between an ordinary person and the driving dumping box 14 is short, the driving of the dumping box 14 is stopped. Therefore, the safety of the dumping box 14 can be improved. The interlock device 19 corresponds to a stop section.

For example, when restarting the driving of the dumping box 14, the worker operates the interlock device 19 to stop the driving of the interlock mechanism 192 (release the interlock). Note that in this embodiment, even if the distance between the worker and the dumping box 14 is short, the information processing device 3 does not output approach information. That is, even if the distance between the worker and the dumping box 14 is short, the interlock device 19 does not stop the driving of the damping box 14.

(Embodiment 6)
FIG. 16 is a block diagram showing a configuration example of the information processing system S according to the sixth embodiment. Among the configurations according to the sixth embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed explanation thereof will be omitted.

The garbage treatment plant 100 of the sixth embodiment has a plurality of garbage treatment facilities 1, for example, two garbage treatment facilities 1. Each waste treatment facility 1 receives different types of waste. For example, the first waste treatment facility 1 receives household waste. The second waste treatment facility 1 receives non-burnable waste. For example, each garbage processing facility 1 is provided with an information processing device 3. The garbage treatment plant 100 has a weighing building (not shown) for weighing the amount of garbage loaded on the vehicle 2.

Vehicle 2 goes to the weigh building before going to waste treatment facility 1. The amount of garbage loaded on the vehicle 2 is weighed at the weighing building. The worker at the weighing building asks the driver of the vehicle 2 about the type of garbage loaded on the vehicle 2, and obtains the type of garbage as an answer. The worker at the weighing building tells the driver to go to the garbage processing facility 1 according to the type of garbage they have collected. For example, if the type of garbage is household garbage, the worker at the weighing building tells the driver to go to the first garbage treatment facility 1.

In the information processing device 3 of the sixth embodiment, a garbage type information DB 310, which will be described later, is stored in the storage unit 31. The information processing device 3 includes a communication section 33 . The communication unit 33 is a communication interface that realizes communication with the in-house computer 8 in the waste disposal plant 100 via the network N. The control unit 30 communicates with the on-site computer 8 via the communication unit 33. For example, the on-site computer 8 is provided in the weighing building.

The on-site computer 8 is a dedicated or general-purpose computer. The on-site computer 8 includes a control section 80, a storage section 81, an input/output section 82, and a communication section 83. The control section 80, the storage section 81, the input/output section 82, and the communication section 83 are connected to each other.

The control unit 80 includes one or more CPUs, GPUs, or other arithmetic processing devices. It also has a memory such as ROM or RAM. The control unit 80 reads out programs stored in the storage unit 81 and executes various processes. The storage unit 81 includes a nonvolatile memory such as a hard disk or SSD, for example. The communication unit 83 is a communication interface that realizes communication with the information processing device 3 via the network N. The control unit 80 communicates with the information processing device 3 via the communication unit 83.

The input/output unit 82 is an input/output interface connected to the fourth camera 84 and the input device 85. The input device 85 is, for example, a keyboard or a touch panel, and accepts input of the type of garbage by a worker in the weighing building. The fourth camera 84 is installed in the weighing building and captures an image of the vehicle 2 where garbage is being weighed. Hereinafter, the image data captured by the fourth camera 84 of the vehicle 2 will also be referred to as measurement image data. The license plate of vehicle 2 is shown in the image data at the time of measurement.

The control unit 80 acquires image data at the time of measurement from the fourth camera 84 via the input/output unit 82 . Based on the acquired image data at the time of measurement, the control unit 80 detects the license plate number of the vehicle 2 (hereinafter referred to as the number of the vehicle 2) in the image data at the time of measurement using a known image recognition technique. Since the image recognition technique is a well-known technique, detailed explanation will be omitted.

The worker uses the input device 85 to input the type of garbage loaded on the vehicle 2. The control unit 80 obtains the input type of garbage via the input/output unit 82 . The control unit 80 stores in the storage unit 81 garbage type information in which the acquired garbage type is associated with the number of the vehicle 2 detected as described above. The control unit 80 outputs the garbage type information to the control unit 30 of the information processing device 3. Note that a worker at the weigh building may check the number of the vehicle 2 and input the number of the vehicle 2 using the input device 85.

The control unit 30 acquires the garbage type information output from the on-site computer 8 and stores the acquired garbage type information in the garbage type information DB 310. FIG. 17 is a conceptual diagram showing an example of the contents of the garbage type information DB 310. A plurality of pieces of garbage type information are stored in the garbage type information DB 310. Specifically, the garbage type information DB 310 has a number column of the vehicle 2 and a garbage type column. The garbage type information DB 310 stores a plurality of vehicle numbers 2 and a plurality of garbage types in association with each other.

The control unit 30 acquires first image data. The control unit 30 detects the number of the vehicle 2 in the first image data using a known image recognition technique based on the acquired first image data. The number of the vehicle 2 in the first image data is the number of the vehicle 2 scheduled to enter the garbage disposal facility 1. The control unit 30 refers to the garbage type information DB 310 and identifies the type of garbage loaded in the vehicle 2 scheduled to enter based on the number of the vehicle 2 scheduled to enter and the garbage type information.

The control unit 30 outputs the identified type of dirt to the speaker 16. For example, if the type of garbage is household garbage, the control unit 30 outputs to the speaker 16 audio data such as "A vehicle loaded with household garbage is scheduled to enter." For example, the audio data is stored in the storage unit 31. The speaker 16 receives the output audio data and outputs the above-mentioned audio to notify the worker of the garbage treatment facility 1 of the type of garbage loaded in the vehicle 2 scheduled to enter. Therefore, the worker at the waste treatment facility 1 does not need to confirm with the driver of the vehicle 2 the type of garbage loaded in the vehicle 2 scheduled to enter. The speaker 16 corresponds to a notification section. Note that the notification unit is not limited to the speaker 16.

The control unit 30 may output the identified type of garbage to the facility terminal instead of or in addition to the speaker 16. For example, if the type of garbage is household garbage, the control unit 30 outputs the above audio data to the facility terminal. For example, the facility terminal receives the output audio data and outputs the audio to the speaker of the facility terminal or the connected earphone. It is possible to notify the workers in the waste treatment facility 1 of the type of waste loaded in the vehicle 2 scheduled to enter. In this case, the facility terminal corresponds to the notification section.

For example, when the control unit 30 cannot identify the type of loaded garbage, for example, when the number of the detected vehicle 2 or the type of garbage associated with the number is not stored in the garbage type information DB, the control unit 30 Information indicating that species information does not exist may be output to the speaker 16 or the facility terminal. For example, the information indicating that garbage type information does not exist includes audio data such as "The type of garbage loaded in the vehicle scheduled to enter the vehicle is not registered."

Notification of the type of garbage is not limited to audio notification. For example, the control unit 30 may output the type of garbage to a display device such as a display (not shown) provided in the garbage processing facility 1. The display device displays the type of garbage that has been output, and notifies the worker of the garbage processing facility 1 of the type of garbage loaded in the vehicle 2 scheduled to enter.

For example, when the facility terminal receives the type of garbage output from the control unit 30, the facility terminal displays the received type of garbage on the display unit of the facility terminal and determines the type of garbage loaded in the vehicle 2 scheduled to enter. Workers at facility 1 may also be notified.

FIG. 18 is a flowchart illustrating a notification process performed by the control unit 30 of the sixth embodiment to notify the type of garbage loaded on the vehicle 2 scheduled to enter. For example, the notification process is performed in parallel with the determination process. Note that the determination process may be performed after the notification process. For example, the control unit 30 starts the following process based on an operation of an operation unit (not shown) provided in the information processing device 3 by a worker of the waste treatment facility 1.

The control unit 30 acquires garbage type information from the on-site computer 8 (S61), and stores the acquired garbage type information in the garbage type information DB 310. As described above, the control unit 30 acquires the first image data (S62), and detects the number of the vehicle 2 scheduled to enter based on the acquired first image data (S63). As described above, the control unit 30 refers to the garbage type information DB 310 and identifies the type of garbage loaded in the vehicle 2 scheduled to enter based on the detected number of the vehicle 2 and the garbage type information (S64). The control unit 30 outputs the identified type of garbage to the speaker 16 (S65), and ends the notification process.

As described above, the speaker 16 notifies the worker of the waste treatment facility 1 of the type of waste loaded in the vehicle 2 scheduled to enter. If the driver of the vehicle 2 makes a mistake in the garbage disposal facility 1 into which the vehicle 2 is to enter, for example, if the vehicle 2 loaded with non-combustible garbage is scheduled to enter the first garbage disposal facility 1, then the vehicle 2 enters the first garbage disposal facility 1. The worker tells the driver that he is going to the wrong waste treatment facility 1. It is possible to prevent the vehicle 2 from entering a garbage treatment facility 1 that does not correspond to the type of garbage loaded therein.

For example, the information processing device 3 may store in the storage unit 31 the types of garbage that can be accepted by the waste treatment facility 1 in which the information processing device 3 is installed. The control unit 30 determines whether the specified type of garbage is an acceptable type of garbage. If the type of garbage is not an acceptable type of garbage, the control unit 30 outputs a signal to the speaker 16 indicating that the garbage processing facility 1 is in the wrong place. The speaker 16 receives the above signal and outputs a warning sound. It is possible to notify the driver and the worker of the waste treatment facility 1 that the vehicle 2 scheduled to enter the wrong garbage treatment facility 1.

In the present embodiment, each garbage treatment facility 1 is provided with an information processing device 3, but for example, one information processing device 3 may determine whether or not to permit entry to each garbage treatment facility 1. .

(Embodiment 7)
FIG. 19 is a schematic diagram of a garbage disposal plant 100 according to the seventh embodiment. Among the configurations according to the seventh embodiment, the same components as those in the sixth embodiment are denoted by the same reference numerals, and detailed explanation thereof will be omitted. In the garbage treatment plant 100 of the seventh embodiment, one garbage treatment facility 1 receives a plurality of types of garbage, such as household garbage and non-burnable garbage. Note that the number of garbage treatment facilities 1 provided in the garbage treatment plant 100 may be one or more. The vehicle 2 goes to the weighing building before going to the waste treatment facility 1 as in the sixth embodiment. The on-site computer 8 outputs garbage type information to the information processing device 3.

The waste treatment facility 1 of the seventh embodiment is provided with a plurality of (two in FIG. 19) hoppers 13. One hopper 13 is arranged near the first parking space 6a and receives non-burnable garbage. The other hopper 13 is located near the second parking space 6b and the third parking space 6c and receives household waste. A dumping box 14 for throwing household waste into the other hopper 13 is provided near the other hopper 13 and near the fourth parking space 6d. Household waste loaded on a general vehicle is thrown into the other hopper 13 by the dumping box 14 .

The parking spaces 6 are provided according to the type of vehicle 2 and the type of garbage loaded on the vehicle 2. A garbage truck loaded with non-combustible garbage stops at the first parking space 6a. Garbage trucks loaded with household garbage stop at the second parking space 6b and the third parking space 6c. General vehicles loaded with household garbage stop at the fourth parking space 6d. Therefore, the upper limit of the number of garbage trucks loaded with non-combustible garbage that can enter the garbage processing facility 1 is one. The upper limit of the number of garbage trucks loaded with household waste that can enter the garbage processing facility 1 is two. The maximum number of general vehicles loaded with household waste that can enter the waste treatment facility 1 is one. In this embodiment, since the dumping box 14 for throwing non-burnable waste into one hopper 13 is not provided, non-burnable waste loaded on a general vehicle cannot be thrown into one hopper 13. Therefore, general vehicles loaded with non-combustible waste are not permitted to enter the waste treatment facility 1. That is, the upper limit of the number of general vehicles loaded with non-combustible waste that can enter the waste treatment facility 1 is zero. For example, the upper limit number of vehicles 2 that can enter is stored in the storage unit 31.

The control unit 30 acquires garbage type information from the on-site computer 8 and stores the acquired garbage type information in the garbage type information DB 310. Also, first image data is acquired from the first camera 5. As in the sixth embodiment, the control unit 30 detects the number of the vehicle 2 scheduled to enter based on the first image data, refers to the garbage type information DB 310, and detects the number of the vehicle 2 scheduled to enter based on the detected number of the vehicle 2 and the garbage type information. , identify the type of garbage loaded on the vehicle 2 scheduled to enter. As in the first embodiment, the control unit 30 inputs the acquired first image data into the learning model M1 to acquire the type of vehicle 2 scheduled to enter. In other words, the control unit 30 specifies the type of vehicle 2 scheduled to enter and the type of garbage loaded on the vehicle 2.

The control unit 30 acquires second image data from the second camera 15. The second image data shows the license plate of the vehicle 2 inside the waste disposal facility 1. As in the first embodiment, the control unit 30 inputs the acquired second image data to the learning model M1. The learning model M1 outputs the area of the vehicle 2 and the type of the vehicle 2 in the second image data. The control unit 30 acquires the area of the vehicle 2 and the type of the vehicle 2 in the output second image data.

When the vehicle 2 is detected, the control unit 30 uses a known image recognition technique to detect the number of the detected vehicle 2 based on the acquired area of the vehicle 2 and the second image data. The control unit 30 refers to the garbage type information DB 310 and identifies the type of garbage loaded on the vehicle 2 based on the detected number of the vehicle 2 and the garbage type information. In other words, the control unit 30 specifies the type of vehicle 2 in the waste treatment facility 1 and the type of garbage loaded on the vehicle 2.

The control unit 30 specifies the type of vacant parking space based on the type of vehicle 2 in the waste treatment facility 1 and the type of garbage loaded on the vehicle 2. In this embodiment, the types of vacant parking spaces include vacant parking spaces for garbage trucks loaded with household waste, vacant parking spaces for garbage trucks loaded with non-combustible garbage, and vacant parking spaces for general vehicles loaded with household garbage. This is a combination of three empty parking spaces. For example, if the detected type of vehicle 2 is only a general vehicle and the type of garbage loaded on the vehicle 2 is household garbage, the control unit 30 determines that the type of empty parking space is loaded with household garbage. A vacant parking space for a garbage truck and a vacant parking space for a garbage truck loaded with non-combustible garbage are identified.

In this embodiment, when a general vehicle is loaded with household waste, it enters the waste treatment facility 1 and stops at the fourth parking space 6d. Furthermore, if a general vehicle is loaded with non-combustible garbage, it will not enter the waste treatment facility 1 (general vehicles loaded with non-combustible garbage will not be allowed to enter). Therefore, the general vehicles in the waste treatment facility 1 are considered to be general vehicles loaded with household waste. Therefore, for example, when the detected type of vehicle 2 is only a general vehicle, the control unit 30 does not detect the license plate number of the vehicle 2 and detects that the type of vacant parking space is a vacant parking space related to a garbage truck loaded with household garbage. It may be specified as a space and an empty parking space related to a garbage truck loaded with non-combustible garbage.

The control unit 30 determines whether or not to permit entry of the vehicle 2 based on the type of vehicle 2 scheduled to enter, the type of garbage loaded on the vehicle 2, and the type of vacant parking space. Specifically, the control unit 30 determines whether the type of vehicle 2 scheduled to enter and the type of garbage in the vehicle 2 correspond to the type of vacant parking space.

When the types of vacant parking spaces include vacant parking spaces for garbage trucks loaded with household waste, and when the garbage truck loaded with household garbage is scheduled to enter, the control unit 30 permits entry. When the type of vacant parking space does not include a vacant parking space for a garbage truck loaded with household waste, and the garbage truck loaded with household garbage is scheduled to enter, the control unit 30 does not permit entry.

When the types of vacant parking spaces include vacant parking spaces for garbage trucks loaded with non-combustible garbage, and when the garbage truck loaded with non-combustible garbage is scheduled to enter, the control unit 30 permits entry. When the type of vacant parking space does not include a vacant parking space for a garbage truck loaded with non-combustible garbage and the garbage truck loaded with non-combustible garbage is scheduled to enter, the control unit 30 does not permit entry.

If the type of vacant parking space includes a vacant parking space for a general vehicle loaded with household waste, and if the general vehicle loaded with household garbage is scheduled to enter, the control unit 30 permits entry. If the type of vacant parking space does not include a vacant parking space for a general vehicle loaded with household waste, and a general vehicle loaded with household garbage is scheduled to enter, the control unit 30 does not permit entry. When a general vehicle loaded with non-combustible garbage is scheduled to enter, for example, if the driver of the general vehicle enters the wrong waste treatment facility 1, the control unit 30 does not permit entry regardless of the type of vacant parking space.

The control unit 30 outputs admission permission information or admission non-permission information to the display unit 4 according to the determination result of whether or not to permit admission. The control unit 30 may output the type of garbage loaded on the vehicle 2 scheduled to enter to the speaker 16 as in the sixth embodiment.

FIG. 20 is a flowchart illustrating the determination process performed by the control unit 30 of the seventh embodiment. For example, the control unit 30 starts the following process based on an operation of an operation unit (not shown) provided in the information processing device 3 by a worker of the waste treatment facility 1.

The control unit 30 performs the processes of S71 and S72. Since the processing in S71 and S72 is similar to the processing in S61 and S62, detailed explanation will be omitted. As described above, the control unit 30 detects the number of the vehicle 2 scheduled to enter based on the first image data, and detects the garbage loaded in the vehicle 2 scheduled to enter based on the detected number of the vehicle 2 and garbage type information. The type is specified (S73). The control unit 30 performs the processes of S74 and S75. Since the processing in S74 and S75 is similar to the processing in S16 and S17, detailed explanation will be omitted. The vehicle 2 scheduled to enter and the type of garbage loaded on the vehicle 2 are identified.

The control unit 30 performs the processes of S76, S77, and S78. The processes in S76, S77, and S78 are similar to the processes in S11, S12, and S13, so a detailed explanation will be omitted. As described above, the control unit 30 detects the number of the vehicle 2 in the garbage treatment facility 1 based on the second image data, and detects the number of the vehicle 2 in the garbage treatment facility 1 based on the detected number of the vehicle 2 and garbage type information. The type of garbage No. 2 is specified (S79). The control unit 30 specifies the type of vacant parking space based on the type of vehicle 2 in the waste treatment facility 1 and the type of garbage in the vehicle 2 (S80).

The control unit 30 determines whether the type of vehicle 2 scheduled to enter and the type of garbage in the vehicle 2 correspond to the type of vacant parking space (S81). If the type of vehicle 2 scheduled to enter and the type of garbage of the vehicle 2 corresponds to the type of vacant parking space (S81: YES), entry permission information is output (S82) and the determination process ends. . If the type of vehicle 2 scheduled to enter and the type of garbage of the vehicle 2 do not correspond to the type of vacant parking space (S81: NO), entry denial information is output (S811) and the determination process ends. do. After the control unit 30 outputs the admission permission information or the admission non-permission information, the control unit 30 may return to the process of S71.

Even if the parking space 6 is provided according to the type of vehicle 2 and the type of garbage, the information processing device 3 can determine the type of vehicle 2 scheduled to enter and the type of garbage loaded on the vehicle 2. It can be determined whether or not to permit entry of the vehicle 2 by taking this into consideration. Note that a plurality of parking spaces 6 may be provided where garbage trucks loaded with non-combustible garbage stop, and parking spaces 6 where general vehicles loaded with household garbage are parked.

For example, the waste treatment facility 1 may be provided with a parking space 6 where general vehicles loaded with non-combustible waste stop. That is, general vehicles loaded with non-combustible waste may be allowed to enter the waste treatment facility 1. For example, in this case, a dumping box 14 for throwing non-combustible waste into the hopper 13 is provided near the hopper 13 that receives non-combustible waste.

For example, the waste treatment facility 1 may be provided with a container for storing household waste and a container for storing non-burnable waste. For example, each container is arranged near the fourth parking space 6d. In this case, if there is a vacant parking space for a general vehicle, entry of the general vehicle may be permitted regardless of the type of garbage loaded on the general vehicle. For example, suppose that the garbage processing facility 1 is configured such that a general vehicle loaded with household garbage and a general vehicle loaded with non-combustible garbage enter the garbage treatment facility 1 and stop at the fourth parking space 6d. At this time, when the type of vehicle 2 detected based on the second image data is a general vehicle, the information processing device 3 instructs the vehicle 2 to detect the number of the vehicle 2 and identify the type of garbage loaded. You don't have to do it.

The embodiments disclosed herein are illustrative in all respects and should not be considered restrictive. The technical features described in each embodiment can be combined with each other, and the scope of the present invention is intended to include all changes within the scope of the claims and the range of equivalents to the scope of the claims. be done.

M1 learning model (first learning model)
M2 Second learning model P Program S Information processing system 100 Garbage treatment plant 1 Garbage treatment facility 13 Hopper (specific device)
14 Damping box 15 Second camera 16 Speaker 17 Detection unit 2 Vehicle 3 Information processing device 30 Control unit 4 Display unit 5 First camera 6 Parking space 7 Person

Claims (4)

In a garbage treatment facility where multiple types of vehicles including garbage trucks enter, first image data is obtained in which the vehicle entering the garbage treatment facility is imaged;
obtaining garbage type information in which the number of the vehicle and the type of garbage loaded on the vehicle are associated;
identifying the type of garbage loaded on the vehicle entering the garbage processing facility based on the number detected based on the acquired first image data and the garbage type information;
A program that causes a computer to execute a process of outputting the identified type of garbage to a notification unit that notifies the type of garbage. In a garbage treatment facility where multiple types of vehicles including garbage trucks enter, second image data is obtained in which the vehicle in the garbage treatment facility is imaged;
The acquired second image data is input to a learning model that is trained to output the type of detection target including the vehicle, person, and specific equipment in the waste treatment facility when the second image data is input. input and obtain the type of detection target,
Deriving a distance between the person and the vehicle or the device based on the acquired type of detection target,
When the derived distance is less than or equal to a threshold, information indicating the approach of the person and the vehicle or the device is transmitted to an approach notification unit that notifies the person of the approach of the person and the vehicle or the device, and a drive of the device. A program that causes a computer to perform a process of outputting to at least one of the stop units to be stopped. In a waste processing facility where multiple types of vehicles including garbage trucks enter, first image data in which the vehicle entering the waste processing facility is imaged, and second image data in which the vehicle in the waste processing facility is imaged. obtain vehicle image data including at least one of the data;
inputting the acquired vehicle image data to a learning model trained to output a vehicle type when vehicle image data is input to acquire the vehicle type;
Based on the relationship between the acquired type of vehicle and a parking space provided in the garbage treatment facility according to the type of vehicle, it is determined whether or not to permit the vehicle to enter the garbage treatment facility. death,
obtaining garbage type information in which the number of the vehicle and the type of garbage loaded on the vehicle are associated;
acquiring the first image data;
identifying the type of garbage loaded on the vehicle entering the garbage processing facility based on the number detected based on the acquired first image data and the garbage type information;
A program that causes a computer to execute a process of outputting the identified type of garbage to a notification unit that notifies the type of garbage. In a waste processing facility where multiple types of vehicles including garbage trucks enter, first image data in which the vehicle entering the waste processing facility is imaged, and second image data in which the vehicle in the waste processing facility is imaged. obtain vehicle image data including at least one of the data;
inputting the acquired vehicle image data to a learning model trained to output a vehicle type when vehicle image data is input to acquire the vehicle type;
Based on the relationship between the acquired type of vehicle and a parking space provided in the garbage treatment facility according to the type of vehicle, it is determined whether or not to permit the vehicle to enter the garbage treatment facility. death,
acquiring the second image data;
The acquired second image data is trained to output the type of detection target including the vehicle, person, and specific equipment in the garbage processing facility when the second image data is input. to obtain the type of detection target,
Deriving a distance between the person and the vehicle or the device based on the acquired type of detection target,
When the derived distance is less than or equal to a threshold, information indicating the approach of the person and the vehicle or the device is transmitted to an approach notification unit that notifies the person of the approach of the person and the vehicle or the device, and a drive of the device. A program that causes a computer to perform a process of outputting to at least one of the stop units to be stopped.

Images