In-vehicle detection device and in-vehicle detection method

Abstract

The invention relates to a device for detecting a passenger in a vehicle cabin, comprising: a first data acquisition unit (11) for acquiring first data acquired by a first sensor (1); a second data acquisition unit (12) for acquiring second data acquired by a second sensor (2) disposed above a seat in the vehicle cabin; a front seat passenger position estimation unit (1411) for estimating a position of a passenger based on the first data acquired by the first data acquisition unit (11); a detection range determination unit (15) for determining a detection range of the second sensor (2) based on the position of the front seat passenger estimated by the front seat passenger position estimation unit (1411); and a rear seat passenger detection unit (142) for detecting a rear seat passenger based on the second data acquired by the second data acquisition unit (12) and the detection range of the second sensor (2) determined by the detection range determination unit (15).

Description

TECHNICAL FIELD

[0001] The present disclosure relates to an in-cabin detection device and an in-cabin detection method. BACKGROUND

[0002] A technology is known that detects a passenger (hereinafter referred to as "rear seat passenger") present on a rear seat (hereinafter referred to as "rear seat") based on an electric wave sensor provided in a cabin.

[0003] For example, a passenger state detection system that detects a rear seat passenger by providing an electric wave sensor in a front seat (hereinafter referred to as "front seat") and transmitting an electric wave from the provided position to the rear seat direction is disclosed in Patent Literature 1.

[0004] PRIOR ART DOCUMENTS

[0005] PATENT LITERATURE

[0006] Patent Literature 1

[0007] Japanese Patent Application Publication No. 2018-202921 SUMMARY

[0008] PROBLEMS TO BE SOLVED BY THE INVENTION

[0009] As the passenger state detection system disclosed in Patent Literature 1, in the case where an electric wave sensor is provided in a front seat, it is possible that a rear seat passenger cannot be detected due to vibration of the front seat or a seat pocket or the like provided in the front seat.

[0010] Therefore, a method can be considered that detects a rear seat passenger by providing an electric wave sensor above a seat and transmitting an electric wave from the provided position to the rear seat direction.

[0011] However, in the case where an electric wave sensor is provided above a seat, depending on the seating state of a passenger (hereinafter referred to as "front seat passenger") present in a front seat, it is possible that the front seat passenger is erroneously detected as a rear seat passenger. The seating state of the front seat passenger refers to the posture or seating position or the like of the front seat passenger.

[0012] The present disclosure is proposed in order to solve the above problems, and aims to detect a rear seat passenger while taking into account the seating state of a front seat passenger.

[0013] MEANS OF SOLVING THE PROBLEMS

[0014] The in-vehicle detection device according to the present disclosure includes a first data acquisition section that acquires first data acquired by a first sensor, a second data acquisition section that acquires second data acquired by a second sensor disposed above a seat in a vehicle cabin, a front-seat passenger position estimation section that estimates a position of a front-seat passenger based on the first data acquired by the first data acquisition section, a detection range decision section that decides a detection range of the second sensor based on the position of the front-seat passenger estimated by the front-seat passenger position estimation section, and a rear-seat passenger detection section that detects a rear-seat passenger based on the second data acquired by the second data acquisition section and the detection range of the second sensor decided by the detection range decision section.

[0015] Effects of Invention

[0016] According to the present disclosure, the in-vehicle detection device can detect a rear-seat passenger while taking into account the seating state of a front-seat passenger. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1A is a block diagram showing a configuration example of the in-vehicle detection device according to Embodiment 1.

[0018] Figure 1B is a block diagram showing another configuration example of the in-vehicle detection device according to Embodiment 1.

[0019] Figure 2A is a diagram showing a placement example of a camera in a vehicle in Embodiment 1, Figure 2B is a diagram showing a placement example of a radar in a vehicle in Embodiment 1.

[0020] Figure 3A is a diagram showing one example of an image of a photographable range of the camera in Embodiment 1, Figure 3B is a diagram showing one example of an image of a detectable range of the radar in Embodiment 1.

[0021] Figure 4A , Figure 4B and Figure 4C is a diagram showing one example of a positional relationship of a front seat, a rear seat, and a divided region in Embodiment 1.

[0022] Figure 5A , Figure 5B , Figure 5C and Figure 5D is a diagram showing one example of an image of a detection range decided by the detection range decision section in Embodiment 1.

[0023] Figure 6is a diagram showing an example of a state in the vehicle cabin when the detection range determination unit determines that there is no detection range of the radar in Embodiment 1.

[0024] Figure 7 is a flowchart for explaining the operation of the in-vehicle cabin detection device according to Embodiment 1.

[0025] Figure 8A and Figure 8B is a diagram showing an example of a hardware structure of the in-vehicle cabin detection device according to Embodiment 1.

[0026] Figure 9 is a diagram showing an example of an image of the detection range when the detection range is determined by changing the direction in which the radar transmits millimeter waves in Embodiment 1.

[0027] Figure 10A and Figure 10B is a diagram showing an example of a radar installation position when the radar is installed at a position other than above the rear seat and an example of the detection range of the radar.

[0028] Figure 11 is a diagram showing an example of a division method of the detectable range of the detection range determination unit in Embodiment 1. DETAILED DESCRIPTION

[0029] Hereinafter, in order to explain the present disclosure in more detail, a mode for carrying out the present disclosure will be explained with reference to the accompanying drawings.

[0030] Embodiment 1.

[0031] Figure 1A is a block diagram showing an example of a structure of the in-vehicle cabin detection device 10 according to Embodiment 1. The in-vehicle cabin detection device 10 is connected to the first sensor 1 and the second sensor 2. The first sensor 1, the second sensor 2, and the in-vehicle cabin detection device 10 are mounted on a vehicle 30 (see FIG. 2).

[0032] The first sensor 1 and the second sensor 2 are sensors capable of detecting a passenger in the vehicle, for example, a camera or a radar. The radar uses a wave band of a microwave band or a millimeter wave or the like in a high frequency band lower than an optical frequency band.

[0033] The in-vehicle cabin detection device 10 detects a passenger present in the vehicle cabin based on data output from the first sensor 1 and the second sensor 2. The in-vehicle cabin detection device 10 can accurately detect a rear seat passenger by the combination of the first data and the second data.

[0034] The in-cabin detection device 10 performs detection of the rear-seat passenger on the basis of the second data acquired by the second sensor 2, with consideration of the position of the front-seat passenger presumed on the basis of the first data acquired by the first sensor 1, when detecting the rear-seat passenger.

[0035] The passenger detection performed by the in-cabin detection device 10 can be performed while the vehicle 30 is running, or can be performed after the vehicle 30 is stopped. Further, the timing at which the in-cabin detection device 10 performs the passenger detection can be the instant at which the vehicle 30 is stopped, or can be after a certain time has elapsed from the instant at which the vehicle 30 is stopped. Here, the stop of the vehicle 30 can be a temporary stop (i.e., parking), or can be a sustained stop (i.e., parking). In addition, the in-cabin detection device 10 can acquire information indicating whether or not the vehicle 30 is stopped, such as information of a shift lever or a parking brake of the vehicle 30, from the vehicle 30.

[0036] The in-cabin detection device 10 includes a first data acquisition section 11, a second data acquisition section 12, a seat information acquisition section 13, a passenger detection section 14, a detection range determination section 15, and a sensor control section 16.

[0037] The passenger detection section 14 includes a front-seat passenger detection section 141 and a rear-seat passenger detection section 142. The front-seat passenger detection section 141 includes a front-seat passenger position presumption section 1411.

[0038] The first data acquisition section 11 acquires the first data acquired by the first sensor 1. The first data acquisition section 11 outputs the acquired first data to the front-seat passenger detection section 141 of the passenger detection section 14.

[0039] The second data acquisition section 12 acquires the second data acquired by the second sensor 2. The second data acquisition section 12 outputs the acquired second data to the rear-seat passenger detection section 142 of the passenger detection section 14.

[0040] The seat information acquisition section 13 acquires information related to seats in the cabin (hereinafter referred to as "seat information"). The seat information is, for example, information related to a front-rear slide position of a seat, an angle of a seatback, or a height of a seat surface. The seat information acquisition section 13 acquires the seat information, for example, from a seat control section (omitted from illustration) provided in the vehicle 30. The seat control section controls the front-rear slide position of the seat, the angle of the seatback, or the height of the seat surface, for example, by driving a slide motor (omitted from illustration) or a tilt motor (omitted from illustration).

[0041] The seat information acquisition section 13 outputs the acquired seat information to the front-seat passenger detection section 141 of the passenger detection section 14.

[0042] The passenger detection unit 14 detects a passenger present in the vehicle 30 on the basis of the first data acquired by the first data acquisition unit 11 or the second data acquired by the second data acquisition unit 12.

[0043] The front seat passenger detection unit 141 of the passenger detection unit 14 detects a front seat passenger of the vehicle 30 on the basis of the first data acquired by the first data acquisition unit 11.

[0044] When the front seat passenger detection unit 141 detects a front seat passenger, the front seat passenger position estimation unit 1411 estimates a position of the front seat passenger on the basis of the first data acquired by the first data acquisition unit 11 and the seat information acquired by the seat information acquisition unit 13. In Embodiment 1, as one example, the position of the front seat passenger is a head position of the front seat passenger.

[0045] The front seat passenger position estimation unit 1411 outputs the estimated information on the position of the front seat passenger to the detection range decision unit 15.

[0046] In addition, the front seat passenger detection unit 141 outputs information on a case where no front seat passenger is detected to the detection range decision unit 15 in a case where no front seat passenger is detected.

[0047] The rear seat passenger detection unit 142 of the passenger detection unit 14 detects a rear seat passenger of the vehicle 30 on the basis of the second data acquired by the second data acquisition unit 12 and the detection range of the second sensor 2 decided by the detection range decision unit 15.

[0048] Specifically, the rear seat passenger detection unit 142 detects a rear seat passenger on the basis of the second data acquired in the detection range of the second sensor 2.

[0049] The passenger detection by the passenger detection unit 14 can be detection performed by analyzing the acquired data, or can be detection performed by comparing the acquired data with data prepared in advance or a threshold value.

[0050] The detection range decision unit 15 decides the detection range of the second sensor 2 on the basis of the position of the front seat passenger estimated by the front seat passenger position estimation unit 1411. The detection range of the second sensor 2 decided by the detection range decision unit 15 will be described later in detail.

[0051] The detection range decision section 15 determines that the rear seat passenger cannot be detected on the basis of the second data acquired by the second sensor 2, and decides that there is no detection range of the second sensor 2, in a case where the position of the front seat passenger estimated by the front seat passenger position estimation section 1411 is located within the rear seat passenger non-detectable region in the vehicle cabin. In Embodiment 1, the "rear seat passenger non-detectable region" is a region in which it is assumed that the rear seat passenger cannot be correctly detected by the front seat passenger in a case where the front seat passenger is detected in the rear seat passenger non-detectable region. The rear seat passenger non-detectable region is appropriately set. Details regarding the rear seat passenger non-detectable region are described later.

[0052] The detection range decision section 15 outputs information regarding the decided detection range of the second sensor 2 to the rear seat passenger detection section 142.

[0053] When it is decided that there is no detection range of the second sensor 2, the detection range decision section 15 outputs information that the detection range of the second sensor 2 is decided to the rear seat passenger detection section 142 and the sensor control section 16.

[0054] The sensor control section 16 turns off the power supply of the second sensor 2 in a case where the detection range decision section 15 decides that there is no detection range of the second sensor 2. Specifically, for example, the sensor control section 16 transmits a control signal for turning off the power supply of the second sensor 2 to the second sensor 2.

[0055] Hereinafter, an example in which the first sensor 1 is a camera and the second sensor 2 is a radar that transmits millimeter waves will be described. In addition, the embodiments shown below are examples, but the present application is not limited to these embodiments.

[0056] Figure 1B is a block diagram showing a configuration example of the in-vehicle cabin detection device 10 according to Embodiment 1, and is an example in which a camera 1A is used as the first sensor 1 and a radar 2A is used as the second sensor 2. In this example, the first data acquisition section 11 is referred to as an image acquisition section 11A that acquires image data photographed by the camera 1A, and the second data acquisition section 12 is referred to as a distance acquisition section 12A that acquires distance data measured by the radar 2A.

[0057] Fig. 2 is a diagram showing a setting example of the camera 1A and the radar 2A in the vehicle 30. Figure 2A is a diagram showing a setting example of the camera 1A, Figure 2B is a diagram showing a setting example of the radar 2A.

[0058] Fig. 3 is a diagram showing one example of an image of a range (hereinafter referred to as "photographable range") la that can be photographed by the camera 1A and one example of an image of a range (hereinafter referred to as "detectable range") 2a that can detect an object in the vehicle cabin by the radar 2A.Figure 3A is a diagram showing one example of an image illustrating a photographable range la of the camera 1A, Figure 3B is a diagram showing one example of an image illustrating a detectable range 2a of the radar 2A.

[0059] In the illustrated example of FIGS. 2 and 3, the front seat passenger 33 and the rear seat passenger 34 are seated in the front seat 31 and the rear seat 32, respectively. In addition, in the illustrated example of FIGS. 2 and 3, the front seat passenger 33 is a passenger on the driver's seat and the front passenger's seat, and the rear seat passenger 34 is a passenger seated behind the front passenger's seat. Furthermore, in the illustrated example shown in FIGS. 2 and 3, the front seat passenger 33 and the rear seat passenger 34 are both adults.

[0060] The camera 1A is provided on an instrument panel or a dashboard, or the like, in the front of the cabin, and photographs at least the front seat 31. The camera 1A can be shared with, for example, a camera possessed by a so-called "Driver Monitoring System (DMS)" mounted on the vehicle 30 in order to monitor the state of the driver in the vehicle.

[0061] The radar 2A is provided above the rear seat 32. Above the rear seat 32 means a position higher than the rear seat 32. In addition, in Embodiment 1, above the seat provided in the cabin means above the headrest of the seat in a state in which the headrest of the seat is positioned at the highest position.

[0062] The radar 2A transmits millimeter waves toward the rear seat 32, and receives a reflected wave formed by the millimeter waves being reflected by a moving object. The radar 2A measures the distance from the position at which the radar 2A is provided to the moving object that reflects the millimeter waves using the result of the transmission and reception of the millimeter waves. In addition, by using the radar 2A, the in-cabin detection device 10 is able to detect a passenger from, for example, chest movement caused by breathing or a heartbeat.

[0063] There is a dead angle such as a portion hidden by the front seat 31 in the photographable range la of the camera 1A, and the in-cabin detection device 10 is unable to detect a passenger present in the dead angle from the image (hereinafter referred to as an "in-cabin image") photographed by the camera 1A alone. In addition, even if the number of settings of the camera 1A is increased to eliminate the dead angle that is not reflected in the in-cabin image, the in-cabin detection device 10 is unable to detect a passenger whose entire body is covered by, for example, a blanket.

[0064] In order to detect a passenger present in a dead angle that does not appear in the in-cabin image, or a passenger whose entire body is covered by, for example, a blanket, a method of detecting a passenger by a sensor (a thin film switch) provided under a seat can be considered, for example. However, this method is unable to determine whether or not there is a difference in the size of a passenger such as a child who is light in weight, an infant placed in a child safety seat, or a pet.

[0065] In recent years, pets or infants left behind in a vehicle cabin have become a social problem, and passenger detection that takes into account differences in body size is desired.

[0066] Therefore, research is being conducted on a vehicle cabin detection device that mounts an electric wave sensor that can accurately detect passengers of various body sizes, passengers in various postures, or infants seated in child safety seats, and detects passengers based on electric waves transmitted by the electric wave sensor. The electric waves transmitted by the electric wave sensor can respond to passengers that do not appear in a vehicle interior image. Furthermore, passengers covered with a blanket can also be detected because the electric waves penetrate the blanket. Passenger detection using the electric wave sensor can also determine differences in body size of detected passengers.

[0067] In Embodiment 1, the vehicle cabin detection device 10 uses an electric wave sensor, specifically, a radar 2A, to detect a rear seat passenger 34 that is difficult to detect. The vehicle cabin detection device 10 combines the camera 1A and the radar 2A to seek to improve the accuracy of passenger detection. Furthermore, the vehicle cabin detection device 10 detects passengers that take into account differences in body size of passengers by millimeter waves transmitted by the radar 2A. Hereinafter, passenger detection that takes into account differences in body size of passengers by millimeter waves transmitted by the radar 2A will be described by citing specific examples.

[0068] The vehicle cabin detection device 10 includes an image acquisition section 11A, a distance acquisition section 12A, a seat information acquisition section 13, a passenger detection section 14, a detection range determination section 15, and a sensor control section 16.

[0069] The passenger detection section 14 includes a front seat passenger detection section 141 and a rear seat passenger detection section 142. The front seat passenger detection section 141 includes a front seat passenger position estimation section 1411.

[0070] The image acquisition section 11A acquires image data captured by the camera 1A inside the vehicle cabin from the camera 1A. The image acquisition section 11A outputs the image data acquired from the camera 1A to the front seat passenger detection section 141 of the passenger detection section 14.

[0071] The distance acquisition section 12A acquires distance data measured by the radar 2A to each object inside the vehicle cabin from the radar 2A. The distance acquisition section 12A outputs the distance data acquired from the radar 2A to the rear seat passenger detection section 142 of the passenger detection section 14.

[0072] The seat information acquisition section 13 acquires seat information.

[0073] The seat information acquisition section 13 outputs the acquired seat information to the front seat passenger detection section 141 of the passenger detection section 14.

[0074] The passenger detection unit 14 detects a passenger present in the vehicle 30 on the basis of image data acquired by the image acquisition unit 11A or distance data acquired by the distance acquisition unit 12A.

[0075] The front seat passenger detection unit 141 of the passenger detection unit 14 detects the front seat passenger 33 on the basis of image data acquired by the image acquisition unit 11A.

[0076] Specifically, for example, the front seat passenger detection unit 141 can detect the front seat passenger 33 by analyzing an in-cabin image based on the image data using a known image recognition technique. In addition, the front seat passenger detection unit 141 can also determine, for example, whether the front seat passenger 33 is dozing off, the drowsiness of the front seat passenger 33, or whether the front seat passenger 33 is looking aside, from the open eye degree or the face orientation of the front seat passenger 33, and the like. The front seat passenger detection unit 141 can also feed back the determination result of dozing off or drowsiness to a control device for maintaining a comfortable indoor environment, such as a replay control device (omitted from illustration) that replays music in the cabin or an air conditioning control device (omitted from illustration) that controls air conditioning, in the case where the front seat passenger 33 is a passenger seated on the front passenger seat.

[0077] When the front seat passenger detection unit 141 detects the front seat passenger 33, the front seat passenger position estimation unit 1411 estimates the position of the front seat passenger 33 on the basis of image data acquired by the image acquisition unit 11A and seat information acquired by the seat information acquisition unit 13. Note that, as described above, in Embodiment 1, the position of the front seat passenger 33 refers to the head position of the front seat passenger 33. Here, the front seat passenger position estimation unit 1411, for example, sets the center of the face of the front seat passenger 33 on the in-cabin image based on the image data as the head position of the front seat passenger 33.

[0078] Specifically, for example, the front seat passenger position estimation unit 1411 first detects the face of the front seat passenger 33 on the basis of the image data using a known image recognition technique. Then, the front seat passenger position estimation unit 1411 determines the size of the detected face of the front seat passenger 33. The front seat passenger position estimation unit 1411 can determine the size of the face of the front seat passenger 33 from the position of the face on the coordinates of the in-cabin image based on the image data. The front seat passenger position estimation unit 1411 determines the distance from the camera 1A to the center of the face of the front seat passenger 33 from the determined size of the face of the front seat passenger 33. For example, in the case where an adult having a standard face size is seated in the standard position of the front seat 31 in advance, the size of the face photographed on the in-cabin image and the distance from the camera 1A to the center of the face are stored as the size of the face and the distance from the camera 1A to the center of the face of the face at the reference position as a reference. The standard position of the front seat 31 refers to a seating position when an adult having a standard face size is seated on a seat having a standard sliding position, a standard seat surface height, and a standard backrest inclination.

[0079] The front-seat passenger position estimation section 1411 can determine the distance between the camera 1A and the center of the face of the front-seat passenger 33 by comparing the size of the face serving as a reference with the size of the face of the front-seat passenger 33 determined. For example, in the case where the size of the face of the front-seat passenger 33 is larger than the size of the face serving as a reference, the center of the face of the front-seat passenger 33 is correspondingly closer to the camera 1A. In other words, the distance between the camera 1A and the center of the face serving as a reference is shorter than the distance between the camera 1A and the center of the face of the front-seat passenger 33.

[0080] The front-seat passenger position estimation section 1411 determines the distance between the camera 1A and the face of the front-seat passenger 33 based on the seat information when determining the distance between the camera 1A and the center of the face of the front-seat passenger 33, taking into account the sliding position of the seat, the seat surface height, and how much the inclination of the seat deviates from the position, the seat surface height, and the inclination of the seat of the standard seat, respectively.

[0081] For example, assume that the sliding position of the seat of the current front seat 31 is ahead of the sliding position of the standard seat. Also, assume that the size of the face of the front-seat passenger 33 is extremely small compared with the size of the standard face. In this case, the face of the front-seat passenger 33 appearing in the cabin interior image can be approximately the same size as the standard face. When the front-seat passenger position estimation section 1411 determines the distance between the camera 1A and the center of the face of the front-seat passenger 33 based only on the cabin interior image, it determines the distance between the camera 1A and the center of the face of the front-seat passenger 33 assuming that the front-seat passenger 33 is seated at the standard position. Therefore, the front-seat passenger position estimation section 1411 takes the seat information into account when estimating the position of the front-seat passenger 33. The front-seat passenger position estimation section 1411 can know that the front-seat passenger 33 is seated at a position ahead of the standard position by taking the seat information into account. Then, the front-seat passenger position estimation section 1411 can more accurately determine the distance between the camera 1A and the center of the face of the front-seat passenger 33.

[0082] Since the setting position and the field of view angle of the camera 1A are known in advance, the front-seat passenger position estimation section 1411 can estimate the position of the front-seat passenger 33 if it can estimate the distance between the camera 1A and the center of the face of the front-seat passenger 33, in other words, the center position of the face of the front-seat passenger 33 in the cabin.

[0083] The front-seat passenger position estimation section 1411 outputs the estimated information about the position of the front-seat passenger 33 to the detection range decision section 15.

[0084] In addition, the front-seat passenger detection section 141 outputs information that the front-seat passenger 33 has not been detected to the detection range decision section 15 in the case where the front-seat passenger 33 has not been detected.

[0085] The rear seat passenger detection section 142 of the passenger detection section 14 detects the rear seat passenger 34 of the vehicle 30 based on the distance data acquired by the distance acquisition section 12A and the detection range of the radar 2A decided by the detection range decision section 15.

[0086] Specifically, for example, the rear seat passenger detection section 142 detects whether or not the rear seat passenger 34 of the vehicle 30 is present based on the distance data acquired in the detection range of the radar 2A.

[0087] In addition, the rear seat passenger detection section 142 does not perform detection of the rear seat passenger 34 in a case where the detection range decision section 15 decides that there is no detection range of the radar 2A.

[0088] The detection range decision section 15 decides the detection range of the radar 2A based on the position of the front seat passenger 33 estimated by the front seat passenger position estimation section 1411.

[0089] Here, the detection range of the radar 2A decided by the detection range decision section 15 will be described.

[0090] The detection range decision section 15 divides the detectable range 2a of the radar 2A into a plurality of regions (hereinafter referred to as "divided regions"), and decides the divided region of the region for detecting the rear seat passenger 34 as the detection range of the radar 2A in the detectable range 2a of the radar 2A. In addition, the detectable range 2a of the radar 2A is a range in which the radar 2A can acquire distance data in the vehicle cabin, in other words, a range in which the radar 2A can transmit millimeter waves and receive reflected waves of the millimeter waves reflected by a moving object.

[0091] Before describing the detection range of the radar 2A decided by the detection range decision section 15, first, several examples will be described using FIG. 4.

[0092] Figure 4A A state in the vehicle cabin is shown, in which the front seat passenger 33 and the rear seat passenger 34 are seated in the standard position of the front seat 31 and the standard position of the rear seat 32, respectively. The front seat passenger 33 and the rear seat passenger 34 are both adults.

[0093] The detection range determination unit 15 divides the detectable range 2a of the radar 2A into, for example, a first region 41 close to the installation position of the radar 2A, a second region 42 farther from the installation position than the first region 41, a third region 43 farther from the installation position than the second region 42, a fourth region 44 farther from the installation position than the third region 43, a fifth region 45 farther from the installation position than the fourth region 44, and a sixth region 46 farther from the installation position than the fifth region 45. The first region 41, second region 42, third region 43, fourth region 44, fifth region 45, and sixth region 46 are the divided regions. In the example shown, the first region 41 includes the headrest of the rear seat 32. The second region 42 and the third region 43 include the backrest of the rear seat 32. The region including the backrest is divided into a headrest-side region and a seat-side region of the rear seat 32. The second region 42 is designated as the headrest-side region, and the third region 43 as the seat-side region. The fourth region 44 includes the seat surface of the rear seat 32. The fifth area 45 and the sixth area 46 are the areas under the feet of the rear seat 32.

[0094] Here, in order to detect rear-seat passengers 34 of various body types or in various postures based on the distance data measured by radar 2A, it is desirable to set the detection range of radar 2A to be as large as possible. However, in order to avoid misdetecting front-seat passenger 33 as rear-seat passenger 34, the detection range of radar 2A must be a range that front-seat passenger 33 cannot enter.

[0095] For example, in Figure 4A In the case of the state inside the carriage shown, when the fourth area 44, the fifth area 45 and the sixth area 46 are included in the detection range of the radar 2A, the rear passenger detection unit 142 misdetects the front passenger 33 as the rear passenger 34 based on the distance data classified as the fourth area 44, the fifth area 45 and the sixth area 46.

[0096] On the other hand, for example, such as Figure 4B As shown, the carriage is in a position where the front passenger 33 has significantly shifted the sliding position of the front seat 31 towards the rear seat 32 to take their seat. The rear passenger 34 is seated in the rear seat 32. Both the front passenger 33 and the rear passenger 34 are adults.

[0097] exist Figure 4B In the situation shown inside the carriage, when not only the fourth area 44, the fifth area 45, and the sixth area 46, but also the third area 43 is included in the detection range of radar 2A, the rear passenger detection unit 142, based on the distance data classified in the third area 43, the fourth area 44, the fifth area 45, and the sixth area 46, mistakenly detects the front passenger 33 as the rear passenger 34. Figure 4BIn this case, although the rear seat passenger 34 is present in the rear seat 32, the rear seat passenger detection section 142 erroneously detects the front seat passenger 33 as the rear seat passenger 34 based on the distance data classified in the third region 43, the fourth region 44, the fifth region 45, and the sixth region 46 even in a case where the rear seat passenger 34 is not present in the rear seat 32.

[0098] Suppose that the state in the vehicle cabin as shown in Figure 4B is obtained, the detection range of the radar 2A is fixedly decided in a manner that the third region 43 to the sixth region 46 are not included in the detection range. In this way, the rear seat passenger detection section 142 has a possibility of erroneously detecting the rear seat passenger 34. For example, as shown in Figure 4C , the state in the vehicle cabin is a state where a baby 35 sitting on a child seat is present as the rear seat passenger 34 in the rear seat 32. There is no passenger other than the baby 35.

[0099] In the case of the state in the vehicle cabin as shown in Figure 4C , when the third region 43 or the fourth region 44 is not included in the detection range of the radar 2A, the rear seat passenger detection section 142 erroneously detects that the rear seat passenger 34 is not present.

[0100] Thus, when the detection of the rear seat passenger 34 is performed by the radar 2A, depending on the presence or absence of the front seat passenger 33, the position of the front seat passenger 33 when the front seat passenger 33 is present, the posture of the front seat passenger 33, or the position of the rear seat passenger 34 when the rear seat passenger 34 is present, the range in which the rear seat passenger 34 can be correctly detected changes in the detectable range 2a.

[0101] The detection range of the radar 2A must be set to be as large as possible in the detectable range 2a (the first region 41 to the sixth region 46), and to include a region in which the rear seat passenger 34 is not erroneously detected.

[0102] Therefore, the detection range decision section 15 dynamically decides the detection range of the radar 2A based on the position of the front seat passenger 33.

[0103] Fig. 5 is a diagram for explaining one example of the detection range of the radar 2A decided by the detection range decision section 15.

[0104] In Fig. 5, the detection range of the radar 2A decided by the detection range decision section 15 is indicated as a detection range 51.

[0105] For example, in the case where the state in the vehicle cabin is the state as shown in Figure 4A , the detection range decision section 15 decides the first region 41, the second region 42, and the third region 43 as the detection range 51 of the radar 2A in order not to erroneously detect the front seat passenger 33 as the rear seat passenger 34 (see Figure 5A ).

[0106] The backseat passenger detection section 142 is set to be able to detect the backseat passenger 34 in a case where the distance data is classified into the first region 41, the second region 42, or the third region 43, that is, in a case where the millimeter wave transmitted by the radar 2A is reflected in the first region 41, the second region 42, or the third region 43. The backseat passenger detection section 142 determines the size of the detected backseat passenger 34 according to which region the distance data is classified into, and can determine whether the backseat passenger 34 is an adult, the infant 35, or a pet according to the determination result. Specifically, for example, in a case where the distance data is classified into the first region 41, the backseat passenger detection section 142 determines that the detected backseat passenger 34 is an adult. Further, for example, in a case where the distance data is classified into the third region 43, the backseat passenger detection section 142 determines that the detected backseat passenger 34 is the infant 35. In addition, in Embodiment 1, an adult is a passenger of a size that is able to get out of the vehicle 30 by own strength even if forgotten on the vehicle 30, and the infant 35 is a passenger of a size that is difficult to get out of the vehicle 30 by own strength if forgotten on the vehicle 30.

[0107] For example, in a case where the state in the vehicle cabin is a state as shown in FIG. 6, the detection range decision section 15 decides the first region 41 and the second region 42 as the detection range 51 of the radar 2A (refer to FIG. 7) in order not to falsely detect the front seat passenger 33 as the backseat passenger 34. Figure 4B Figure 5B For example, in a case where the state in the vehicle cabin is a state as shown in FIG. 6, the detection range decision section 15 decides the first region 41 and the second region 42 as the detection range 51 of the radar 2A (refer to FIG. 7) in order not to falsely detect the front seat passenger 33 as the backseat passenger 34.

[0108] The backseat passenger detection section 142 is set to be able to detect the backseat passenger 34 in a case where the distance data is classified into the first region 41 and the second region 42, that is, in a case where the millimeter wave transmitted by the radar 2A is reflected in the first region 41 or the second region 42.

[0109] For example, in a case where the state in the vehicle cabin is a state as shown in FIG. 6, the detection range decision section 15 decides the first region 41 and the second region 42 as the detection range 51 of the radar 2A (refer to FIG. 7) in order not to falsely detect the front seat passenger 33 as the backseat passenger 34. Figure 4C Figure 5C For example, in a case where the state in the vehicle cabin is a state as shown in FIG. 6, the detection range decision section 15 decides the first region 41 and the second region 42 as the detection range 51 of the radar 2A (refer to FIG. 7) in order not to falsely detect the front seat passenger 33 as the backseat passenger 34.

[0110] ​​In a case where the rear seat 32 has the infant 35 sitting on the child safety seat, the rear seat passenger detection section 142 can detect the infant 35 even if the first region 41, the second region 42, the third region 43, and the fourth region 44 are taken as the detection range 51 of the radar 2A. However, the detection range determination section 15 sets the detection range 51 of the radar 2A to the largest range possible in a case where the front seat passenger 33 is not present. That is, the detection range determination section 15 includes the fifth region 45 and the sixth region 46 in the detection range 51 of the radar 2A. Thus, the rear seat passenger detection section 142 can more reliably detect the rear seat passenger 34 than in a case where the fifth region 45 and the sixth region 46 are not included in the detection range 51 of the radar 2A. Specifically, for example, in a case where the infant 35 is playing at the feet, or in a case where a pet (omitted from the drawing) is present at the feet, the rear seat passenger detection section 142 can also detect the infant 35 or the pet (see Figure 5D ).

[0111] That is, the detection range determination section 15 does not erroneously detect the front seat passenger 33 as the rear seat passenger 34 in a case where the front seat passenger 33 is not detected, and thus sets the detectable range 2a of the radar 2A to the detection range of the radar 2A in order to expand the detection range of the radar 2A as much as possible. In addition, the detection range determination section 15 acquires information that the front seat passenger 33 is not detected from the front seat passenger detection section 141.

[0112] The detection range determination section 15 can also determine that there is no detection range of the radar 2A.

[0113] Figure 6 is a drawing that shows one example of the state inside the vehicle cabin in a case where the detection range determination section 15 determines that there is no detection range of the radar 2A in Embodiment 1.

[0114] In Figure 6 , the state inside the vehicle cabin is set to a state in which the front seat passenger 33 greatly inclines the seatback of the front seat 31 toward the rear seat 32. The front seat passenger 33 is an adult. The rear seat 32 does not have the rear seat passenger 34.

[0115] For example, in a case where the state inside the vehicle cabin is the state shown in Figure 6 , the position of the front seat passenger 33 greatly intrudes into the detectable range 2a of the radar 2A (the first region 41 to the sixth region 46). The rear seat passenger detection section 142 is highly likely to erroneously detect the front seat passenger 33 as the rear seat passenger 34.

[0116] In a case where the position of the front seat passenger 33 greatly intrudes into the detectable range 2a of the radar 2A, the detection range determination section 15 determines that it is not possible to detect the rear seat passenger 34 based on the distance data, and determines that there is no detection range of the radar 2A.

[0117] The detection range determination section 15 determines whether the position of the front seat passenger 33 greatly intrudes into the detectable range 2a of the radar 2A, depending on whether the position of the front seat passenger 33 is located within the rear seat passenger non-detectable region in the vehicle cabin. The rear seat passenger non-detectable region can be appropriately set. The rear seat passenger non-detectable region can be, for example, the region of the second region 42 in the detectable range 2a of the radar 2A, or a region determined in advance in the vehicle cabin.

[0118] Thus, the detection range determination section 15 dynamically determines the detection range of the radar 2A based on the position of the front seat passenger 33 estimated by the front seat passenger position estimation section 1411.

[0119] The detection range determination section 15 outputs information about the determined detection range of the radar 2A to the rear seat passenger detection section 142.

[0120] When it is determined that there is no detection range of the radar 2A, the detection range determination section 15 outputs information indicating that it is determined that there is no detection range of the radar 2A to the rear seat passenger detection section 142 and the sensor control section 16.

[0121] The sensor control section 16 turns off the power of the radar 2A when it is determined by the detection range determination section 15 that there is no detection range of the radar 2A. Specifically, for example, the sensor control section 16 transmits a control signal for turning off the power of the radar 2A to the radar 2A.

[0122] An operation of the in-vehicle cabin detection device 10 according to Embodiment 1 will be described.

[0123] Figure 7 is a flowchart for describing an operation of the in-vehicle cabin detection device 10 according to Embodiment 1.

[0124] The image acquisition section 11A acquires image data captured by the camera 1A in the vehicle cabin from the camera 1A (step ST701). The image acquisition section 11A outputs the image data acquired from the camera 1A to the front seat passenger detection section 141 of the passenger detection section 14.

[0125] The seat information acquisition section 13 acquires seat information (step ST702).

[0126] The seat information acquisition section 13 outputs the acquired seat information to the front seat passenger detection section 141.

[0127] The front seat passenger detection section 141 detects the front seat passenger 33 of the vehicle 30 based on the image data acquired by the image acquisition section 11A in step ST701 (step ST703).

[0128] The front-seat passenger detection section 141 outputs information that the front-seat passenger 33 has not been detected to the detection range decision section 15 in a case where the front-seat passenger 33 has not been detected. Then, the operation of the in-vehicle detection device 10 skips step ST704 and proceeds to step ST705.

[0129] When the front-seat passenger detection section 141 detects the front-seat passenger 33 in step ST703, the front-seat passenger position estimation section 1411 estimates the position of the front-seat passenger 33 on the basis of the image data acquired by the image acquisition section 11A in step ST701 and the seat information acquired by the seat information acquisition section 13 in step ST702 (step ST704).

[0130] The front-seat passenger position estimation section 1411 outputs the estimated information about the position of the front-seat passenger 33 to the detection range decision section 15.

[0131] The detection range decision section 15 decides the detection range of the radar 2A on the basis of the position of the front-seat passenger 33 estimated by the front-seat passenger position estimation section 1411 in step ST704 (step ST705).

[0132] The detection range decision section 15 outputs information about the decided detection range of the radar 2A to the rear-seat passenger detection section 142.

[0133] When it is decided that there is no detection range of the radar 2A, the detection range decision section 15 outputs information that it is decided that there is no detection range of the radar 2 to the rear-seat passenger detection section 142 and the sensor control section 16.

[0134] The distance acquisition section 12A acquires distance data measured by the radar 2A to the distance to each object in the vehicle cabin from the radar 2A (step ST706). The distance acquisition section 12A outputs the distance data acquired from the radar 2A to the rear-seat passenger detection section 142.

[0135] The rear-seat passenger detection section 142 detects the rear-seat passenger 34 of the vehicle 30 on the basis of the distance data acquired by the distance acquisition section 12A in step ST706 and the detection range of the radar 2A decided by the detection range decision section 15 in step ST705 (step ST707).

[0136] In addition, the rear-seat passenger detection section 142 does not perform detection of the rear-seat passenger 34 in a case where it is decided by the detection range decision section 15 that there is no detection range of the radar 2A. In this case, the sensor control section 16 turns off the power supply of the radar 2A.

[0137] The operation of the in-vehicle detection device 10 shown in the flowchart of Figure 7 The operation of the in-vehicle detection device 10 shown in the flowchart of

[0138] Further, in the flowchart of Figure 7 The in-cabin detection device 10 can act in the order of the step ST701 and the step ST702, or in the order of the step ST702 and the step ST701.

[0139] Thus, the in-cabin detection device 10 determines the detection range of the radar 2A based on the position of the front-seat passenger 33 estimated from the image data. Then, the in-cabin detection device 10 detects the rear-seat passenger 34 based on the distance data acquired from the radar 2A and the determined detection range of the radar 2A.

[0140] The in-cabin detection device 10 dynamically determines the detection range of the radar 2A according to the position of the front-seat passenger 33, and thus can detect the rear-seat passenger 34 while taking the seating state of the front-seat passenger 33 into account.

[0141] Information related to the passenger detected by the in-cabin detection device 10 (hereinafter referred to as "detected passenger information") is output by an output section (omitted from illustration) included in the in-cabin detection device 10, and is used for various functions. The detected passenger information includes information on whether or not a passenger is detected, information on whether the detected passenger is the front-seat passenger 33 or the rear-seat passenger 34 when a passenger is detected, or information on whether the detected passenger is an adult or not (an infant 35 or a pet), and the like. Further, as described above, the in-cabin detection device 10 can perform passenger detection taking the difference in body size into account, according to which of the divided areas the distance data in the detection range of the radar 2A is classified.

[0142] The output section of the in-cabin detection device 10 outputs the detected passenger information to an external device (omitted from illustration). The external device can be at least one of a speaker or a display mounted on the vehicle 30, can be a passenger seated in the vehicle 30 or a mobile terminal or the like held by the owner of the vehicle 30, or can be a hazard signal lamp or a horn (car horn) or the like mounted on the vehicle 30.

[0143] For example, the detection passenger information is used for a vehicle theft prevention function. As a specific example, assume that in a state where the vehicle 30 is parked, the control section (omitted from illustration) of the vehicle 30 detects that a large impact is applied to the vehicle 30 or that the opening and closing of the door is performed in a locked state. In this case, the in-cabin detection device 10 activates the camera 1A and the radar 2A, acquires image data and distance data from the camera 1A and the radar 2A, respectively, and performs detection of a person present in the cabin. Then, the in-cabin detection device 10 outputs the detection passenger information to a mobile terminal held by the owner of the vehicle 30. Thereby, the owner of the vehicle 30 can detect, for example, that a person has entered the vehicle 30. Further, for example, the in-cabin detection device 10 outputs the image data acquired from the camera 1A to a security system together with the detection passenger information. The in-cabin detection device 10 is connected to the security system via a network.

[0144] In the security system, the situation in the cabin is grasped on the basis of the detection passenger information and the camera 1A, and the vehicle 30 is remotely controlled, so that theft of the vehicle 30 can be prevented. The control of the vehicle 30 performed by the security system remotely is, for example, control to make the engine not start.

[0145] Further, for example, in a case where the in-cabin detection device 10 performs detection of a passenger in a state where the vehicle 30 is parked, the detection passenger information is used for a baby forgetting detection function. The in-cabin detection device 10 outputs the detection passenger information to a baby forgetting detection device (omitted from illustration). The in-cabin detection device 10 is connected to the baby forgetting detection device via a network.

[0146] The baby forgetting detection device determines that baby forgetting is likely to have occurred in a case where, for example, the detection passenger information indicating that the baby 35 is detected is output from the in-cabin detection device 10. At this time, the baby forgetting detection device determines that it is not baby forgetting if it can be determined from the detection passenger information that an adult other than the baby 35 is detected, and, for example, determines that the vehicle owner (adult) intentionally rests or waits in the cabin, so that the forgetting alarm can be turned off. In addition, the in-cabin detection device 10 can output the image data acquired from the camera 1A to the baby forgetting detection device together with the detection passenger information, and the baby forgetting detection device can determine whether an adult other than the baby 35 is present on the basis of the image data.

[0147] Further, for example, in a case where the in-cabin detection device 10 performs detection of a passenger in a state where the vehicle 30 is running, the detection passenger information is used for a seat belt reminder function.

[0148] Figure 8A 、 Figure 8B FIG. 1 is a diagram showing one example of a hardware structure of the in-cabin detection device 10 according to Embodiment 1.

[0149] In Embodiment 1, the functions of the image acquisition unit 11A, the distance acquisition unit 12A, the seat information acquisition unit 13, the passenger detection unit 14, the detection range determination unit 15, and the sensor control unit 16 are implemented by the processing circuitry 801. That is, the in-vehicle compartment detection device 10 includes the processing circuitry 801 for performing control to detect a passenger in the vehicle 30.

[0150] The processing circuitry 801 can be a dedicated hardware as shown in FIG. 8A, or a CPU (Central Processing Unit) 805 that executes a program stored in a memory 806 as shown in FIG. 8B. Figure 8A Figure 8B The processing circuitry 801 can be a dedicated hardware as shown in FIG. 8A, or a CPU (Central Processing Unit) 805 that executes a program stored in a memory 806 as shown in FIG. 8B.

[0151] In the case where the processing circuitry 801 is a dedicated hardware, the processing circuitry 801 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof.

[0152] ​In a case where the processing circuit 801 is the CPU 805, the functions of the image acquisition section 11A, the distance acquisition section 12A, the seat information acquisition section 13, the passenger detection section 14, the detection range decision section 15, and the sensor control section 16 are implemented by software, firmware, or a combination of software and firmware. That is, the image acquisition section 11A, the distance acquisition section 12A, the seat information acquisition section 13, the passenger detection section 14, the detection range decision section 15, and the sensor control section 16 are implemented by the CPU 805, a system LSI (Large-Scale Integration), or the like processing circuit that executes a program stored in the HDD (Hard Disk Drive) 802, the memory 806, or the like. Further, the program stored in the HDD 802, the memory 806, or the like can also be said to be a program that causes a computer to execute the steps or methods of the image acquisition section 11A, the distance acquisition section 12A, the seat information acquisition section 13, the passenger detection section 14, the detection range decision section 15, and the sensor control section 16. Here, the memory 806, for example, corresponds to a nonvolatile or volatile semiconductor memory such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable ROM), an EEPROM (Electrically Erasable Programmable Read-Only Memory), or the like, or a magnetic disk, a floppy disk, an optical disk, a compact disc, a mini disc, a DVD (Digital Versatile Disc), or the like.

[0153] In addition, as for the functions of the image acquisition section 11A, the distance acquisition section 12A, the seat information acquisition section 13, the passenger detection section 14, the detection range decision section 15, and the sensor control section 16, a part thereof can be implemented by dedicated hardware and a part thereof can be implemented by software or firmware. For example, as for the image acquisition section 11A, the distance acquisition section 12A, and the seat information acquisition section 13, the functions thereof can be implemented by the processing circuit 801 as dedicated hardware, and as for the passenger detection section 14, the detection range decision section 15, and the sensor control section 16, the functions thereof can be implemented by the processing circuit 801 reading and executing a program stored in the memory 806.

[0154] Further, the in-vehicle detection device 10 has an input interface device 803 and an output interface device 804 for wired or wireless communication with the camera 1A, the radar 2A, or the like device, or an external device.

[0155] In Embodiment 1 above, the detection range determining section 15 divides the detectable range 2a in which the radar 2A can measure distance data into a plurality of divided areas (the first area 41, the second area 42, the third area 43, the fourth area 44, the fifth area 45, and the sixth area 46) for determining whether the acquired distance data can be used for detection of the rear seat passenger 34, and dynamically determines the detection range of the radar 2A by changing the divided areas. However, this is only one example. The detection range determining section 15 can dynamically determine the detection range of the radar 2A by changing the direction in which the radar 2A transmits millimeter waves. Furthermore, the detection range determining section 15 can determine the detection range of the radar 2A by a combination of the divided areas and the direction in which the radar 2A transmits millimeter waves.

[0156] For example, in Embodiment 1 above, in the case where the state in the vehicle cabin is the state illustrated in FIG. 4, the detection range determining section 15 describes one example in which the first area 41 and the second area 42 among the detectable range 2a (the first area 41 to the sixth area 46) of the radar 2A are determined as the detection range of the radar 2A (refer to FIG. 4). Figure 5B For example, in the case where the state in the vehicle cabin is the state illustrated in FIG. 4, the detection range determining section 15 can change the direction in which the radar 2A transmits millimeter waves to be more toward the rear seat 32 side, and determine the divided areas that do not include the position of the front seat passenger 33 as the detection range of the radar 2A on this basis.

[0157] Figure 9 is a diagram for describing one example of the image of the detection range in the case where, when the state in the vehicle cabin is the state illustrated in FIG. 4, the detection range determining section 15 determines the divided areas that do not include the position of the front seat passenger 33 as the detection range of the radar 2A on the basis of changing the direction in which the radar 2A transmits millimeter waves to be more toward the rear seat 32 side.

[0158] Figure 9 In, the detection range determining section 15 causes millimeter waves transmitted from the radar 2A toward the approximate center of the detectable range 2a to be transmitted toward the headrest of the rear seat 32, and sets the first area 41, the second area 42, the third area 43, the fourth area 44, and the fifth area 45 as the detection range of the radar 2A on this basis.

[0159] In addition, the detection range determining section 15 can change the direction in which the radar 2A transmits millimeter waves via the sensor control section 16.

[0160] Furthermore, in Embodiment 1 above, the radar 2A is provided above the rear seat 32 (refer to FIG. 2, FIG. 4 to Figure 6 ), but this is only one example. The radar 2A can be provided above a seat provided in the vehicle cabin. If the radar 2A is provided above a seat provided in the vehicle cabin, it can be provided on a pillar, for example.

[0161] In addition, the radar 2A is preferably disposed above the seat provided in the vehicle cabin and more to the front seat 31 side than the position above the rear seat 32. This is because, for example, when a metal plate is embedded in the backrest portion of the rear seat 32, the millimeter wave transmitted by the radar 2A can not be able to penetrate the rear seat 32 due to the metal plate. If the millimeter wave does not penetrate, in the in-cabin detection device 10, the rear seat passenger 34 of the rear seat 32 cannot be detected.

[0162] Here, FIG. 10 is a diagram for explaining one example of the setting position of the radar 2A when the radar 2A is disposed at a place other than above the rear seat 32 and one example of the detection range 1001 of the radar 2A decided by the detection range decision portion 15.

[0163] In FIG. 10, the radar 2A is disposed above the front seat 31. In addition, the radar 2A is disposed at a position above the front seat 31 when the front seat 31 is in the standard position.

[0164] Figure 10A With Figure 4A Also, the state in the vehicle cabin is shown, in which the front seat passenger 33 and the rear seat passenger 34 are seated in the front seat 31 and the rear seat 32, respectively, in the standard positions.

[0165] Figure 10B With Figure 4B Also, the state in the vehicle cabin is shown, in which the front seat passenger 33 is seated with the sliding position of the front seat 31 largely shifted to the rear seat 32 side, and the rear seat passenger 34 is seated in the rear seat 32.

[0166] In the case where the state in the vehicle cabin is Figure 10A the state shown in FIG. 10, one example of the detection range of the radar 2A decided by the detection range decision portion 15 is explained.

[0167] The detection range decision portion 15 divides the detectable range 2a of the radar 2A into a plurality of regions. Figure 10AIn this case, the detection range determination section 15 divides the detectable range 2a into, for example, a first region 101 to which millimeter waves from the radar 2A are transmitted to the rearmost side with respect to the traveling direction of the vehicle 30, a second region 102 that is more forward than the first region 101 with respect to the traveling direction of the vehicle 30 (hereinafter, referred to as "front side of the vehicle 30"), a third region 103 that is more forward than the second region 102 with respect to the front side of the vehicle 30, a fourth region 104 that is more forward than the third region 103 with respect to the front side of the vehicle 30, a fifth region 105 that is more forward than the fourth region 104 with respect to the front side of the vehicle 30, a sixth region 106 that is more forward than the fifth region 105 with respect to the front side of the vehicle 30, a seventh region 107 that is more forward than the sixth region 106 with respect to the front side of the vehicle 30, and an eighth region 108 that is more forward than the seventh region 107 with respect to the front side of the vehicle 30. The first region 101 to the eighth region 108 are each a divided region. In the illustrated example, the first region 101 is a region that includes the headrest of the rear seat 32. The second region 102 and the third region 103 are regions that include the backrest of the rear seat 32, the region that includes the backrest is divided into a region on the headrest side and a region on the seat surface side of the rear seat 32, the second region 102 is set as the region on the headrest side, and the third region 103 is set as the region on the seat surface side. The fourth region 104 is a region that includes the seat surface of the rear seat 32. The fifth region 105 is a region that includes the feet of the rear seat 32. The sixth region 106 and the seventh region 107 are regions that include the front seat 31, and the eighth region 108 is a region that includes the feet of the front seat 31.

[0168] Thus, the detection range determination 15 can also divide the detectable range 2a along the direction in which millimeter waves are transmitted from the radar 2A.

[0169] In addition, the detection range determination section 15 can divide the detectable range 2a into divided regions by combining a method of dividing according to the distance from the installed position of the radar 2A (for example, refer to the first region 41 to the sixth region 46 of FIG. 4) and a method of dividing as shown in Figure 10A

[0170] For example, in the case of the state in the vehicle cabin as shown in Figure 10A When the fourth region 104 to the eighth region 108 are included in the detection range of the radar 2A, the rear seat passenger detection section 142 erroneously detects the front seat passenger 33 as the rear seat passenger 34 based on the distance data classified in the fourth region 104 to the eighth region 108.

[0171] Therefore, the detection range determination section 15 determines the first region 101, the second region 102, and the third region 103 as the detection range 1001 of the radar 2A to prevent the rear seat passenger detection section 142 from erroneously detecting the front seat passenger 33 as the rear seat passenger 34.

[0172] On the other hand, the state in the vehicle cabin is as shown in​Figure 10B In the case where the state inside the vehicle cabin is the state shown in FIG. 10, not only the fourth region 104 to the eighth region 108, but also the third region 103 is included in the detection range of the radar 2A, and the rear seat passenger detection section 142 erroneously detects the front seat passenger 33 as the rear seat passenger 34 based on the distance data classified in the third region 103 to the eighth region 108. In Figure 10B , the rear seat 32 is provided with the rear seat passenger 34, but even in the case where the rear seat 32 is not provided with the rear seat passenger 34, the rear seat passenger detection section 142 detects the front seat passenger 33 as the rear seat passenger 34.

[0173] Therefore, the detection range decision section 15 decides the first region 101 and the second region 102 as the detection range 1001 of the radar 2A in order to prevent the front seat passenger 33 from being erroneously detected as the rear seat passenger 34.

[0174] In addition, the detection range decision section 15 can divide the detectable range 2a in accordance with the width of the range from which the millimeter wave is transmitted from the radar 2A, as shown in Figure 10A .

[0175] Thus, for example, as shown in Figure 6 , in the case where the state inside the vehicle cabin is the state in which the front seat passenger 33 greatly inclines the seat back of the front seat 31 to the side of the rear seat 32, the detection range decision section 15 can divide the detectable range 2a in accordance with the width of the range from which the millimeter wave is transmitted, and decide the divided region along the seat back of the rear seat 32 as the detection range (refer to Figure 11 ). Thus, the rear seat passenger detection section 142 can detect the rear seat passenger 34 without erroneously detecting the front seat passenger 33 as the rear seat passenger 34 in the case where the rear seat passenger 34 is present. However, in this case, the rear seat passenger detection section 142 cannot perform the passenger detection that discriminates the body size difference. Specifically, the rear seat passenger detection section 142 cannot discriminate whether the rear seat passenger 34 is an adult or a child or an infant in the case where the rear seat passenger 34 is present.

[0176] Further, in the above-described embodiment 1, the front seat passenger position estimation section 1411 estimates the position of the front seat passenger 33 based on the image data acquired by the image acquisition section 11A and the seat information acquired by the seat information acquisition section 13, but this is only one example. The front seat passenger position estimation section 1411 can also estimate the position of the front seat passenger 33 from the image data without using the seat information. In this case, the inside vehicle cabin detection device 10 can be configured to include the seat information acquisition section 13.

[0177] Further, in the above-described embodiment 1, the inside vehicle cabin detection device 10 includes the sensor control section 16, but the inside vehicle cabin detection device 10 is not necessarily required to include the sensor control section 16. However, the inside vehicle cabin detection device 10 can prevent the unnecessary operation of the radar 2A by including the sensor control section 16.

[0178] Further, in the above embodiment 1, the position of the front seat passenger 33 refers to a head position of the front seat passenger 33, but this is only an example. The position of the front seat passenger 33 can be, for example, a shoulder position of the front seat passenger 33, or a wrist position of the front seat passenger 33.

[0179] Further, in the above embodiment 1, the front seat passenger detection section 141 detects the front seat passenger 33 based on the image data acquired by the image acquisition section 11A, but this is only an example. For example, the rear seat passenger detection section 142 can also detect the front seat passenger 33 while detecting the rear seat passenger 34 based on the distance data acquired by the distance acquisition section 12A. For example, when the radar 2A is provided at the position shown in Fig. 10, the rear seat passenger detection section 142 can detect the front seat passenger 33 based on the distance data.

[0180] Further, in the above embodiment 1, the first sensor 1 is the camera 1A, but the first sensor 1 can also be a radar.

[0181] Further, in the above embodiment 1, the in-vehicle detection device 10 is mounted as an in-vehicle device on the vehicle 30, and the image acquisition section 11A, the distance acquisition section 12A, the seat information acquisition section 13, the passenger detection section 14, the detection range decision section 15, and the sensor control section 16 are provided in the in-vehicle detection device 10. This is not limiting, and part of the image acquisition section 11A, the distance acquisition section 12A, the seat information acquisition section 13, the passenger detection section 14, the detection range decision section 15, and the sensor control section 16 can be mounted in an in-vehicle device of the vehicle, and the other part is provided in a server connected to the in-vehicle device via a network, and the in-vehicle detection system is constituted by the in-vehicle device and the server.

[0182] As described above, the in-vehicle detection device 10 according to the embodiment 1 includes: a first data acquisition section 11 that acquires first data acquired by a first sensor 1; a second data acquisition section 12 that acquires second data acquired by a second sensor 2 provided above a seat in a vehicle cabin; a front seat passenger position estimation section 141 that estimates a position of a front seat passenger 33 based on the first data acquired by the first data acquisition section 11; a detection range decision section 15 that decides a detection range of the second sensor 2 based on the position of the front seat passenger 33 estimated by the front seat passenger position estimation section 141; and a rear seat passenger detection section 142 that detects a rear seat passenger 34 based on the second data acquired by the second data acquisition section 12 and the detection range of the second sensor 2 decided by the detection range decision section 15. Therefore, the in-vehicle detection device 10 can detect the rear seat passenger 34 while taking into account the seating state of the front seat passenger 33.

[0183] Further, in the in-vehicle compartment detection device 10 according to Embodiment 1, the detection range decision section 15 decides that there is no detection range of the second sensor 2 in a case where the position of the front seat passenger 33 estimated by the front seat passenger position estimation section 1411 is located in the rear seat passenger non-detectable region in the in-vehicle compartment, and in a case where the detection range decision section 15 decides that there is no detection range of the second sensor 2, the in-vehicle compartment detection device 10 is configured to include the sensor control section 16 for turning off the power supply of the second sensor 2. Thus, the in-vehicle compartment detection device 10 can prevent unnecessary operation of the radar 2A.

[0184] In addition, the present disclosure can be modified in the range of the disclosure to change any structural element of the embodiments, or omit any structural element of the embodiments.

[0185] Industrial Applicability

[0186] The in-vehicle compartment detection device according to the present disclosure detects a passenger present in a rear seat in consideration of a seating state of a passenger present in a front seat, and is thus suitable for an in-vehicle compartment detection device for detecting a passenger in a vehicle compartment.

[0187] Explanation of Reference Numerals

[0188] 1 first sensor, 1A camera, 1a photographable range, 2 second sensor, 2A radar, 2a detectable range, 51, 1001 detection range, 10 in-vehicle compartment detection device, 11 first data acquisition section, 11A image acquisition section, 12 second data acquisition section, 12A distance acquisition section, 13 seat information acquisition section, 14 passenger detection section, 141 front seat passenger detection section, 142 rear seat passenger detection section, 1411 front seat passenger position estimation section, 15 detection range decision section, 16 sensor control section, 31 front seat, 32 rear seat, 33 front seat passenger, 34 rear seat passenger, 35 infant, 41, 101 first region, 42, 102 second region, 43, 103 third region, 44, 104 fourth region, 45, 105 fifth region, 106 sixth region, 107 seventh region, 108 eighth region, 801 processing circuit, 802 HDD, 803 input interface device, 804 output interface device, 805 CPU, 806 memory.

Claims

1. A detection device for use inside a train carriage, characterized in that, include: A first data acquisition unit acquires first data acquired by a first sensor; The second data acquisition unit acquires second data obtained by a second sensor installed above the seats in the carriage. The front passenger position estimation unit estimates the position of the front passenger based on the first data acquired by the first data acquisition unit. A detection range determination unit determines the detection range of the second sensor based on the position of the front-seat passenger estimated by the front-seat passenger position estimation unit; and The rear passenger detection unit detects rear passengers based on the second data acquired by the second data acquisition unit and the detection range of the second sensor determined by the detection range determination unit.

2. The in-car detection device as described in claim 1, characterized in that, It includes a seat information acquisition unit that acquires seat information related to the seats in the carriage. The front passenger position estimation unit estimates the position of the front passenger based on the first data obtained by the first data acquisition unit and the seat information obtained by the seat information acquisition unit.

3. The in-car detection device as described in claim 1, characterized in that, If the detection range determination unit determines that the detection range of the second sensor is not present when the position of the front passenger, as estimated by the front passenger position estimation unit, is located within the undetectable area of ​​the rear passenger in the carriage, then the detection range determination unit determines that the detection range of the second sensor is not present. The in-car detection device includes a sensor control unit, which disconnects the power supply to the second sensor when the detection range determination unit determines that there is no detection range for the second sensor.

4. The in-car detection device as described in any one of claims 1 to 3, characterized in that, The first sensor is a camera. The first data is image data obtained by the camera from taking pictures of the interior of the carriage.

5. The in-car detection device as described in any one of claims 1 to 3, characterized in that, The second sensor is radar. The second data is distance data obtained by the radar measuring the distance to the moving object inside the carriage.

6. The in-car detection device as described in claim 5, characterized in that, The detection range determination unit classifies the area where the radar can measure the distance data into multiple segmented areas for determining whether the acquired distance data can be used for passenger detection, and determines the detection range by changing the segmented areas that constitute the detection range or the angle at which the radar transmits radio waves.

7. A method for detecting inside a train carriage, characterized in that, include: The first data acquisition unit acquires the first data acquired by the first sensor; The second data acquisition unit acquires the second data obtained by the second sensor installed above the seats in the carriage; The step of the front passenger position estimation unit estimating the position of the front passenger based on the first data acquired by the first data acquisition unit; The step of the detection range determination unit determining the detection range of the second sensor based on the position of the front passenger estimated by the front passenger position estimation unit; as well as The step of the rear passenger detection unit detecting rear passengers based on the second data acquired by the second data acquisition unit and the detection range of the second sensor determined by the detection range determination unit.

Images