JP2022042328A - Object recognition controller and method for controlling object recognition - Google Patents

Abstract

To properly recognize a person.SOLUTION: A controller 20 as an object recognition controller includes: a picture data acquisition unit 21 for acquiring picture data taken by a camera 11 for imaging around a vehicle; a recognition processing unit 22 for determining that there is a person when a score which shows a possibility that there is a person in the picture data acquired by the picture data acquisition unit 21 is at least a threshold value; a person distribution information acquisition unit 23 for acquiring person distribution information around the vehicle; and a presentation processing unit 24 for presenting information of the person recognized by the recognition processing unit 22 to the driver of the vehicle. The recognition processing unit 22 changes the threshold value of the score on the basis of the person distribution information acquired by the person distribution information acquisition unit 23.SELECTED DRAWING: Figure 1

Description

The present invention relates to a person recognition control device and a person recognition control method.

In the recognition process using the person recognition dictionary, the person recognition in the captured image is determined to be a person when the score indicating the person-likeness is equal to or higher than a predetermined threshold value. There is known a technique for determining that an object is a pedestrian based on a high score indicating pedestrian-likeness using a recognition dictionary for captured images (see, for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2016-015029

When there are many people around the vehicle, the driver of the vehicle drives while paying attention to the person, so even if the recognition rate of the person by the recognition process is low, the safety hindrance is small. On the other hand, if there are few people around the vehicle, the driver of the vehicle may not pay attention to the person, so if the recognition process cannot recognize the person, safety may be impaired. ..

The present invention has been made in view of the above, and an object of the present invention is to appropriately recognize a person.

In order to solve the above-mentioned problems and achieve the object, the person recognition control device according to the present invention has a video data acquisition unit that acquires video data taken by a shooting unit that shoots the periphery of the vehicle, and the video data acquisition unit. When the score indicating the personality of the person is equal to or higher than the threshold value for the video data acquired by the unit, the recognition processing unit that recognizes the person as the person and the person distribution information acquisition unit that acquires the person distribution information around the vehicle. The recognition processing unit includes a presentation processing unit that presents the information of the person recognized by the recognition processing unit to the driver of the vehicle, and the recognition processing unit is based on the person distribution information acquired by the person distribution information acquisition unit. To change the threshold.

The person recognition control method according to the present invention shows a person-likeness with respect to a video data acquisition step of acquiring video data taken by a shooting unit that shoots the periphery of a vehicle and a video data acquired in the video data acquisition step. When the score is equal to or higher than the threshold value, the recognition processing step of recognizing the person as the person, the person distribution information acquisition step of acquiring the person distribution information around the vehicle, and the information of the person recognized in the recognition processing step of the vehicle are obtained. In the recognition processing step, including the presentation processing step presented to the driver, the threshold value is changed based on the person distribution information acquired in the person distribution information acquisition step.

According to the present invention, there is an effect that a person can be appropriately recognized.

FIG. 1 is a block diagram showing a configuration example of a person recognition device having a person recognition control device according to the first embodiment. FIG. 2 is a flowchart showing a processing flow in the person recognition control device according to the first embodiment. FIG. 3 is a block diagram showing a configuration example of a person recognition device having the person recognition control device according to the second embodiment. FIG. 4 is a flowchart showing a processing flow in the person recognition control device according to the second embodiment.

Hereinafter, embodiments of the person recognition control device and the person recognition control method according to the present invention will be described in detail with reference to the accompanying drawings. The present invention is not limited to the following embodiments.

[First Embodiment]
<People recognition device>
FIG. 1 is a block diagram showing a configuration example of a person recognition device 10 having a control unit 20 which is a person recognition control device according to the first embodiment. The person recognition device 10 recognizes a person by performing recognition processing using a person recognition dictionary on the video data. The person recognition device 10 notifies the driver of information about the recognized person.

The person recognition device 10 includes a camera (shooting unit) 11, a recognition dictionary storage unit 12, a display unit 13, and a control unit (person recognition control device) 20. The person recognition device 10 may be implemented as a function of, for example, a device having a safe driving support function, a navigation device, a drive recorder, etc., which are installed in advance in the vehicle.

The camera 11 is a camera that photographs the periphery of the vehicle. The camera 11 is mounted on the vehicle and photographs the traveling direction of the vehicle. The camera 11 is composed of a far-infrared camera, a visible light camera, or a far-infrared camera and a visible light camera. The camera 11 is arranged, for example, at a position where the front of the vehicle, which is the traveling direction of the vehicle, can be photographed. The camera 11 constantly captures images from the start of the engine to the stop of the engine, that is, while the vehicle is in operation. The camera 11 outputs the captured video data to the video data acquisition unit 21 of the control unit 20. The video data is, for example, a moving image composed of an image of 30 frames per second. In this embodiment, the camera 11 is arranged in front of the vehicle.

The recognition dictionary storage unit 12 stores dictionary data for recognizing various objects from video data. The recognition dictionary storage unit 12 stores, for example, a person recognition dictionary capable of machine learning various images taken by a person and collating that the object included in the image data is a person. The recognition dictionary storage unit 12 is, for example, a storage device such as a ROM (Read Only Memory), a semiconductor memory element such as a flash memory, or an external storage device via a network.

The display unit 13 is a device that displays various types of information, and as an example, is a display device unique to the person recognition device 10, or a display device shared with other systems including a drive recorder and a navigation system. The display unit 13 is a display including, for example, a liquid crystal display (LCD: Liquid Crystal Display) or an organic EL (Electro-luminescence) display. The display unit 13 displays a video based on the video signal output from the presentation processing unit 24 of the control unit 20. In the present embodiment, the display unit 13 is arranged at a position that can be visually recognized by the driver of the vehicle while driving.

<People recognition control device>
The control unit 20 is, for example, an arithmetic processing unit (control device) configured by a CPU (Central Processing Unit) or the like. The control unit 20 loads the stored program into the memory and executes an instruction included in the program. The control unit 20 includes an internal memory (not shown), and the internal memory is used for temporary storage of data in the control unit 20 and the like. Therefore, the control unit 20 causes the person recognition device 10 to execute the person recognition method. Further, the control unit 20 is a computer that operates the program according to the present invention. The control unit 20 includes a video data acquisition unit 21, a recognition processing unit 22, a person distribution information acquisition unit 23, and a presentation processing unit 24, which are connected to the bus 20X.

The video data acquisition unit 21 acquires video data taken by the camera 11. More specifically, the video data acquisition unit 21 acquires video data output by the camera 11 and captured in front of the vehicle in the traveling direction.

The recognition processing unit 22 recognizes a person from the video data acquired by the video data acquisition unit 21. The recognition processing unit 22 performs pattern matching on the video data using the person recognition dictionary stored in the recognition dictionary storage unit 12 to perform person recognition processing for recognizing a person. A known method can be used in the person recognition process. The recognition processing unit 22 may constantly perform processing while the person recognition device 10 is activated. The recognition processing unit 22 outputs the person recognition result to the person distribution information acquisition unit 23.

The recognition processing unit 22 performs pattern matching on the video data acquired by the video data acquisition unit 21, and recognizes the video data as a person when the score indicating the person-likeness is equal to or higher than a predetermined threshold value. Let A be the threshold value (specified value) of the normal score. The threshold value of the score is changed according to the person distribution information acquired by the person distribution information acquisition unit 23. Let B (B <A) be the threshold of the changed score.

The recognition processing unit 22 recognizes a person by changing the threshold value of the score based on the person distribution information acquired by the person distribution information acquisition unit 23. In the present embodiment, the recognition processing unit 22 lowers the threshold value of the score when the person distribution information acquired by the person distribution information acquisition unit 23 indicates that the number of people distributed around the vehicle is small. More specifically, when the person distribution is less than the distribution threshold value, for example, two people, the recognition processing unit 22 recognizes the person as a person using a threshold value B lower than the threshold value A of the normal score. As a result, when the person distribution is smaller than the distribution threshold value, in other words, when there are few people around the vehicle, it is easy to be recognized as a person. When the person distribution is equal to or larger than the distribution threshold value, the recognition processing unit 22 recognizes the person as a person by using the threshold value A of the normal score.

For example, the maximum score indicating personality is 1.0. As an example, the threshold value A is 0.9 and the threshold value B is 0.7. For example, the recognition processing unit 22 normally determines that the detected object is a person when the score is 0.9 or more. For example, the recognition processing unit 22 determines that the detected object is a person when the person recognition result shows that the person distribution in the image is less than two people and the score is 0.7 or more. For example, when the recognition processing unit 22 performs person recognition on the condition that the score is 0.7 or more, the recognition processing unit 22 uses a normal score threshold value when there are two or more people in the image. Return to the recognition process.

The threshold value of the score may be changed stepwise or linearly in proportion to the distribution of people.

The person distribution information acquisition unit 23 acquires person distribution information indicating the distribution of people around the vehicle. In the present embodiment, the person distribution information acquisition unit 23 acquires the person distribution information around the vehicle based on the person recognition result of the recognition processing unit 22.

The person distribution information is information showing the distribution of people existing around the vehicle. The person distribution information includes information on the number of people existing around the vehicle. The person distribution information may include information on the position of a person existing in the vicinity of the vehicle.

The person existing in the vicinity of the vehicle may be any person recognized as a person from the video data. The person existing around the vehicle is a person existing in front of the vehicle in the traveling direction taken by the camera 11. For example, a person recognized from the video data captured in front of the vehicle in the traveling direction may be a person existing in front of the vehicle in the traveling direction. The person existing in the vicinity of the vehicle may be limited to the recognized person who is located near the vehicle. For example, if the size of the recognized person is equal to or larger than a predetermined size, or if the recognized position in the video data is within a predetermined range, the person may be located near the vehicle.

In the present embodiment, the recognition of a person is described as an example of the specific object recognized by the recognition processing unit 22, but the bicycle on which the person is riding may be recognized as an example of the specific object.

The presentation processing unit 24 presents the information of the person recognized by the recognition processing unit 22 to the driver of the vehicle. By presenting the information of the person recognized by the presentation processing unit 24, the driver of the vehicle is alerted to the person. The presentation processing unit 24 presents the recognized person information to the driver by using the display by the display unit 13 or the voice output from the voice output unit (not shown). Therefore, the presentation processing unit 24 functions as a display control unit when the information of the specific object is presented to the driver by the display by the display unit 13. The presentation processing unit 24 generates an image that presents information to the driver of the vehicle. The presentation processing unit 24 generates, for example, presentation video data for displaying video data by superimposing a frame line on a range of recognized persons. The presentation processing unit 24 may generate presentation video data for displaying characters or icons that call attention to a person, for example. The presentation processing unit 24 may generate audio such as a warning sound to be output together with the video. The presentation processing unit 24 outputs a video signal for displaying the generated video to the display unit 13 to display the video. The presentation processing unit 24 may output an audio signal for outputting the audio generated together with the video to a speaker (not shown).

<Processing in the control unit>
Next, the flow of processing in the control unit 20 will be described with reference to FIG. FIG. 2 is a flowchart showing a processing flow in the control unit 20 which is the person recognition control device according to the first embodiment. When the person recognition device 10 is activated, the processing of the flowchart shown in FIG. 2 is disclosed. The start of the process of FIG. 2 is started under arbitrary conditions. For example, when the vehicle becomes available, such as when the engine of the vehicle equipped with the person recognition device 10 starts, or when the operation of the person recognition device 10 is started by the user's operation. Further, the process of step S101 may be executed on condition that the vehicle is running.

With the start of the process of FIG. 2, the control unit 20 starts the shooting and the person recognition process (step S101). More specifically, the control unit 20 starts shooting by the camera 11. The control unit 20 acquires the video data output by the camera 11 by the video data acquisition unit 21. The control unit 20 recognizes a person around the vehicle from the video data acquired by the video data acquisition unit 21 by the recognition processing unit 22. The threshold value of the score of the person recognition process in step S101 is the threshold value A of the score at the normal time. For example, the control unit 20 determines that the detected object is a person when the score is 0.9 or more by the recognition processing unit 22. The control unit 20 proceeds to step S102.

The process in step S101 is started, and the control unit 20 determines whether or not the person distribution is less than a predetermined threshold value (step S102). More specifically, the control unit 20 acquires the person distribution information around the vehicle by the person distribution information acquisition unit 23 based on the person recognition result of the recognition processing unit 22. Based on the acquired person distribution information, the control unit 20 proceeds to step S104 when the person distribution around the vehicle is less than the distribution threshold value (Yes in step S102). For example, the control unit 20 proceeds to step S104 when the person distribution information has less than two people distributed around the vehicle (Yes in step S102). Based on the acquired person distribution information, the control unit 20 proceeds to step S103 when the person distribution around the vehicle is equal to or greater than the distribution threshold value (No in step S102). For example, the control unit 20 proceeds to step S103 when the person distribution information has two or more people distributed around the vehicle (No in step S102). The process of step S102 may proceed to step S104 when the state in which the distribution of people around the vehicle is less than the distribution threshold value continues for a predetermined period or longer, for example, 60 seconds.

When the distribution of people around the vehicle is larger than the distribution threshold value (No in step S102), the control unit 20 executes the person recognition process by the recognition processing unit 22 using the threshold value A of the normal score (step S104). .. More specifically, the recognition processing unit 22 performs pattern matching on the video data, and determines that the detected object is a person when the score indicating the person-likeness is 0.9 or more. The control unit 20 proceeds to step S105.

When the distribution of people around the vehicle is less than the distribution threshold value (Yes in step S102), the control unit 20 performs the person recognition process by the recognition processing unit 22 using the threshold value B lower than the threshold value A of the normal score. Execute (step S104). More specifically, the recognition processing unit 22 performs pattern matching on the video data, and if the score indicating the character-likeness is 0.7 or more, the recognition processing unit 22 determines that the detected object is a person. As a result, when the distribution of people in the image is small, the detected object is more likely to be recognized as a person than in normal times. The control unit 20 proceeds to step S105.

The control unit 20 determines whether or not a person has been recognized based on the person recognition result of the recognition processing unit 22 (step S105). When the control unit 20 determines that the person has been recognized (Yes in step S105), the control unit 20 proceeds to step S106. If the control unit 20 does not determine that the person has been recognized (No in step S105), the control unit 20 proceeds to step S107.

The control unit 20 displays a frame line in the range of the person recognized by the presentation processing unit 24 (step S106). More specifically, the control unit 20 causes the presentation processing unit 24 to superimpose a frame line surrounding a person around the vehicle on the video data and display it on the display unit 13. The control unit 20 proceeds to step S107.

The control unit 20 determines whether or not to end the shooting and the person recognition process (step S107). The control unit 20 determines that the shooting and the person recognition process are finished, for example, when the operation for ending the person recognition process is detected, or when the vehicle is stopped and the engine is turned off. When the control unit 20 determines that the shooting and the person recognition process are completed (Yes in step S107), the control unit 20 ends the process. If the control unit 20 does not determine that the shooting and the person recognition process are finished (No in step S107), the control unit 20 re-executes the process of step S102.

<Effect>
As described above, in the present embodiment, the threshold value of the score in the person recognition process is changed based on the person distribution information around the vehicle. According to the present embodiment, it is possible to appropriately recognize a person according to the person distribution information.

In the present embodiment, when the person distribution information indicates that the number of people distributed around the vehicle is small, the threshold value of the score in the person recognition process is lowered. According to the present embodiment, when there are few people around the vehicle and there is a possibility that the driver's attention to the person may be distracted, it is possible to facilitate recognition of the person in the person recognition process. In this way, according to the present embodiment, safer driving can be supported.

The present embodiment presents information about the recognized person to the driver. The present embodiment makes it easier to recognize a person in the person recognition process when there is a possibility that the driver's attention to the person may be distracted. As a result, according to the present embodiment, when there is a possibility that the driver's attention to the person may be distracted, it is possible to increase the opportunity to call the driver of the vehicle to pay attention to the person.

Further, in the present embodiment, when it is shown that the number of people distributed around the vehicle is small, in order to lower the threshold value of the score in the person recognition process, it is like a person who is partially hidden behind an object. Even if it is not recognized as a person in a normal score, it is recognized as a person. Therefore, the driver of the vehicle can more appropriately know the existence of the person.

Further, in the present embodiment, when it is shown that the number of people distributed around the vehicle is small, the threshold value of the score in the person recognition process is lowered, so that the erroneous recognition of an object that is not a person as a person increases. do. However, even if it is a misrecognition, it is presented as the existence of a person, so that the driver of the vehicle can be made to drive with more emphasis on safety.

[Second Embodiment]
The person recognition device 10A according to the present embodiment will be described with reference to FIGS. 3 and 4. FIG. 3 is a block diagram showing a configuration example of a person recognition device 10A having a control unit 20A, which is a person recognition control device according to the second embodiment. FIG. 4 is a flowchart showing a processing flow in the control unit 20A, which is the person recognition control device according to the second embodiment. The basic configuration of the person recognition device 10A is the same as that of the person recognition device 10 of the first embodiment. In the following description, the same components as those of the person recognition device 10 are designated by the same reference numerals or corresponding reference numerals, and detailed description thereof will be omitted. The person recognition device 10A recognizes that it has a GNSS (Global Navigation Satellite System) receiving unit 14A and a map information storage unit 15A, and that the control unit 20A has a position information acquisition unit 24A and a map information acquisition unit 25A. The processing in the processing unit 22A and the person distribution information acquisition unit 23A is different from the first embodiment.

The GNSS receiving unit 14A receives a GNSS signal from a GNSS satellite (not shown). The GNSS receiving unit 14A outputs the signal of the received radio wave to the position information acquisition unit 24A of the control unit 20A.

The map information storage unit 15A includes information on roads including general roads, motorways and expressways, information on private houses and facilities existing around the roads, information on sidewalks, and information on cycling roads. Memorize map information. The map information storage unit 15A is shared with other systems including a navigation system, for example.

The position information acquisition unit 24A calculates the current position information of the vehicle by a known method based on the GNSS signal received by the GNSS reception unit 14A.

The map information acquisition unit 25A acquires the map information stored in the map information storage unit 15A. In the present embodiment, the map information acquisition unit 25A acquires map information around the vehicle.

The person distribution information acquisition unit 23A acquires the person distribution information around the vehicle based on the current position information indicating the current position of the vehicle acquired by the position information acquisition unit 24A and the map information acquired by the map information acquisition unit 25A. do. The person distribution information acquisition unit 23A may further acquire the person distribution information around the vehicle based on the time information indicating the current time and the environment information indicating the environment around the vehicle.

Environmental information is information indicating the environment around the vehicle that affects the distribution of people. The environmental information includes, for example, information indicating at least one of the weather, the season, the day of the week, and whether or not an event is held around the current position of the vehicle. Environmental information can be acquired from an external device via a communication unit (not shown).

For example, when the current position of the vehicle is a general road where the entry of people is not restricted, the person distribution information acquisition unit 23A acquires the person distribution information indicating that the place has a large number of people. For example, when the current position of the vehicle is a highway where the entry of people is restricted, the person distribution information acquisition unit 23A acquires the person distribution information indicating that the place has a small number of people. A place with a large distribution of people is a place where there is a high possibility that a person will exist. A place with a small distribution of people is a place where it is unlikely that a person will exist.

For example, when the current position of the vehicle is near a private house or a facility, the person distribution information acquisition unit 23A acquires the person distribution information indicating that the place has a large number of people. For example, when the current position of the vehicle is not near a private house or a facility, the person distribution information acquisition unit 23A acquires the person distribution information indicating that the place has a small person distribution.

The person distribution information acquisition unit 23A acquires person distribution information indicating that the place has a large number of people distribution, for example, when the current position of the vehicle is a road having a sidewalk or a road running in parallel with or intersecting a cycling road. do. The person distribution information acquisition unit 23A acquires the person distribution information indicating that the place has a small person distribution, for example, when the current position of the vehicle is not a road having a sidewalk or a road running in parallel with or intersecting a cycling road. do.

For example, when the current time is daytime, the person distribution information acquisition unit 23A acquires the person distribution information indicating that the person distribution is large or small. For example, when the current time is nighttime, the person distribution information acquisition unit 23A acquires the person distribution information indicating that the person distribution is small or large. As a result, for example, in office districts, shopping districts, tourist spots with no private houses in the vicinity, and around schools, person distribution information indicating that the distribution of people is large during the day and the distribution of people is small at night is acquired. ..

The person distribution information acquisition unit 23A acquires, for example, the person distribution information indicating that the person distribution is large when the weather at the current position of the vehicle is fine. The person distribution information acquisition unit 23A acquires, for example, the person distribution information indicating that the person distribution is small when the weather at the current position of the vehicle is rainy or stormy.

The person distribution information acquisition unit 23A acquires, for example, the person distribution information indicating that the person distribution is large or small depending on the season of the current position of the vehicle. As a result, for example, for a beach, the person distribution information indicating that the person distribution is large in summer and the person distribution is small except in summer is acquired.

For example, when the day of the week is a holiday, the person distribution information acquisition unit 23A acquires the person distribution information indicating that the person distribution is large or small. For example, when the day of the week is a weekday, the person distribution information acquisition unit 23A acquires the person distribution information indicating that the person distribution is small or large. As a result, for example, in the office district, the person distribution information indicating that the person distribution is large on weekdays and the person distribution is small on holidays is acquired.

The person distribution information acquisition unit 23A acquires, for example, the person distribution information indicating that the person distribution is large when the event is held around the current position of the vehicle. The person distribution information acquisition unit 23A acquires, for example, the person distribution information indicating that the person distribution is small when the event is not held around the current position of the vehicle.

The person distribution information acquisition unit 23A may acquire the person distribution information by combining a plurality of the above conditions.

The recognition processing unit 22A scores when it indicates that the current position of the vehicle is a place where the number of people distributed around the vehicle is small, based on the person distribution information acquired by the person distribution information acquisition unit 23A. Lower the threshold of. The recognition processing unit 22A has a small number of people whose current position, current time, and environment of the vehicle are distributed around the vehicle based on the person distribution information acquired by the person distribution information acquisition unit 23A. The score threshold may be lowered to indicate at least one of location, time, or environment.

Next, the flow of processing in the control unit 20A will be described with reference to FIG. The processing of step S111, step S113 to step S117 shown in FIG. 4 is the same as that of step S101, step S103 to step S107 of the flowchart of FIG.

While the process in step S111 is started, the control unit 20A determines whether or not the person distribution is in a place, time, or environment (step S112). More specifically, the control unit 20A acquires the person distribution information around the vehicle based on the time information and the environmental information in addition to the current position information and the map information of the vehicle by the person distribution information acquisition unit 23A. When the control unit 20A indicates that the number of people distributed around the vehicle is small, the time, or the environment based on the person distribution information acquired by the person distribution information acquisition unit 23A (Yes in step S112). ), Proceed to step S114. When the control unit 20A indicates that the location, time, or environment has a large number of people distributed around the vehicle based on the person distribution information acquired by the person distribution information acquisition unit 23A (No in step S112). ), Proceed to step S113.

As described above, in the present embodiment, when the current position of the vehicle indicates that the number of people distributed around the vehicle is small, the threshold value of the score of the person recognition process is lowered. According to the present embodiment, when there are few people around the vehicle and there is a possibility that the driver's attention to the person may be distracted, it is possible to make it easier to recognize the person. According to this embodiment, safer driving can be supported.

In the present embodiment, when it is shown that the number of people distributed around the vehicle is small, the time, or the environment, the threshold value of the score can be lowered. According to this embodiment, safe driving can be supported more appropriately.

In the present embodiment, the person distribution information is acquired based on at least one of the weather, the season, the day of the week, and the information indicating whether or not the event is held around the current position of the vehicle. According to this embodiment, safe driving can be supported more appropriately.

Further, in the present embodiment, when it is shown that the number of people distributed around the vehicle is small, the time, or the environment, a part of the shadow of the object is hidden in order to lower the threshold value of the score in the person recognition process. Even if is not recognized as a person by a normal score like a hidden person, it is recognized as a person. Therefore, the driver of the vehicle can more appropriately know the existence of the person.

Further, in the present embodiment, when indicating that the number of people distributed around the vehicle is small, the time, or the environment, the threshold value of the score in the person recognition process is lowered, so that an object that is not a person is regarded as a person. Misrecognition as a judgment increases. However, even if it is a misrecognition, it is presented as the existence of a person, so that the driver of the vehicle can be made to drive with more emphasis on safety.

Although the person recognition device 10 according to the present invention has been described so far, it may be implemented in various different forms other than the above-described embodiment.

Each component of the illustrated person recognition device 10 is functionally conceptual and does not necessarily have to be physically configured as shown in the figure. That is, the specific form of each device is not limited to the one shown in the figure, and all or part of the device is functionally or physically dispersed or integrated in an arbitrary unit according to the processing load and usage status of each device. You may.

The configuration of the person recognition device 10 is realized, for example, by a program loaded in a memory as software. In the above embodiment, it has been described as a functional block realized by cooperation of these hardware or software. That is, these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof.

The above-mentioned components include those that can be easily assumed by those skilled in the art and those that are substantially the same. Further, the above configurations can be combined as appropriate. Further, various omissions, substitutions or changes of the configuration can be made without departing from the gist of the present invention.

<Modification example>
In the above, the case of recognizing a person has been described, but it can also be applied to the case of recognizing a vehicle or a two-wheeled vehicle, for example.

In the above, the device has been described as a device for assisting the driver of the vehicle, but it may be used for the person recognition process in the autonomous driving vehicle. The person recognition device 10 includes an automatic driving control unit that outputs a control signal for controlling the vehicle to decelerate or stop when a person is recognized by the recognition processing unit 22. The person recognition device 10 may not include the display unit 13 and the presentation processing unit 24.

10 Person recognition device 11 Camera (shooting unit)
12 Recognition dictionary storage unit 13 Display unit 20 Control unit (person recognition control device)
21 Video data acquisition unit 22 Recognition processing unit 23 Person distribution information acquisition unit 24 Presentation processing unit

Claims (5)

The video data acquisition unit that acquires the video data taken by the shooting unit that shoots the surroundings of the vehicle, and the video data acquisition unit.
When the score indicating the personality of the person is equal to or higher than the threshold value for the video data acquired by the video data acquisition unit, the recognition processing unit recognizes the person as the person.
A person distribution information acquisition unit that acquires person distribution information around the vehicle, and a person distribution information acquisition unit.
A presentation processing unit that presents the information of the person recognized by the recognition processing unit to the driver of the vehicle, and a presentation processing unit.
Equipped with
The recognition processing unit changes the threshold value based on the person distribution information acquired by the person distribution information acquisition unit.
Person recognition control device. The person distribution information acquisition unit acquires the person distribution information based on the person recognition result of the recognition processing unit.
When the person distribution information acquired by the person distribution information acquisition unit indicates that the number of people distributed around the vehicle is small, the recognition processing unit lowers the threshold value.
The person recognition control device according to claim 1. The person distribution information acquisition unit acquires the person distribution information in the vicinity of the vehicle based on the current position information and the map information indicating the current position of the vehicle.
The recognition processing unit indicates that the current position of the vehicle is a place where the number of people distributed around the vehicle is small, based on the person distribution information acquired by the person distribution information acquisition unit. If the threshold is lowered,
The person recognition control device according to claim 1. The person distribution information acquisition unit further acquires the person distribution information around the vehicle based on the time information indicating the current time and the environment information indicating the environment around the vehicle.
The recognition processing unit is a person whose current position, current time, and environment of the vehicle are distributed around the vehicle based on the person distribution information acquired by the person distribution information acquisition unit. If it indicates that the number of people is at least one of a small place, time, and environment, lower the threshold.
The environmental information includes at least one of the weather, season, day of the week and whether or not an event is held around the current position of the vehicle.
The person recognition control device according to claim 3. A video data acquisition step to acquire video data taken by the shooting unit that shoots the surroundings of the vehicle,
When the score indicating humanity is equal to or higher than the threshold value for the video data acquired in the video data acquisition step, the recognition processing step of recognizing the person as a person and the recognition processing step.
The person distribution information acquisition step for acquiring the person distribution information around the vehicle, and
The presentation processing step of presenting the information of the person recognized in the recognition processing step to the driver of the vehicle, and the presentation processing step.
Including
In the recognition processing step, the threshold value is changed based on the person distribution information acquired in the person distribution information acquisition step.
Person recognition control method.

Images