WO2021210041A1

WO2021210041A1 - Face detection device and face detection method

Info

Publication number: WO2021210041A1
Application number: PCT/JP2020/016277
Authority: WO
Inventors: 和樹國廣; 太郎熊谷
Original assignee: 三菱電機株式会社
Priority date: 2020-04-13
Filing date: 2020-04-13
Publication date: 2021-10-21

Abstract

A face detection device (10) is provided with: a vehicle interior image obtaining unit (11) that obtains a vehicle interior image, which is a photographed image of the interior of a vehicle; a face detection unit (12) that searches the vehicle interior image for a person's face and specifies a face area in the vehicle interior image, the face area being an area in which the person's face has been detected; and a search area setting unit (13) that sets, on the basis of the position of the face area, a search area in the vehicle interior image, the search area being an area in which the person's face is to be searched for. The face detection unit (12) searches the search area of the vehicle interior image for the person's face, with respect to the vehicle interior image obtained after the search area has been set.

Description

Face detection device and face detection method

This disclosure relates to a technique for detecting the face of a vehicle occupant.

An image of the inside of the vehicle (hereinafter referred to as "in-vehicle image") is analyzed to recognize the occupant's face, and the recognition result is used to authenticate the occupant and monitor the condition (for example, detection of the driver's doze or inattentiveness). Various in-vehicle devices have been proposed. For example, Patent Document 1 below discloses an authentication device capable of recognizing the faces of a plurality of occupants from an in-vehicle image taken by one wide-angle camera.

International Publication No. 2018/116373

When the in-vehicle device recognizes the occupant's face from the in-vehicle image, it is first necessary to search for the occupant's face from the in-vehicle image. When the vehicle interior image is taken by a wide-angle camera, the size of the vehicle interior image is large. Therefore, when the entire vehicle interior image is analyzed to search for the occupant's face, the processing load becomes large. This problem is particularly noticeable in applications that require repeated searches for the occupant's face, such as occupant condition monitoring.

In Patent Document 1, an area to be analyzed (for example, an area corresponding to the periphery of the headrest of each seat) is preset in the vehicle interior image, and if the face is searched only within the area, the face can be searched. The processing load required for searching can be suppressed. However, the position of the occupant's face shown in the vehicle interior image is not constant depending on the physique and posture of the occupant, the seated position, and the like, so the position of the region is not always appropriate.

The present disclosure has been made in order to solve the above problems, and it is possible to appropriately set the area for searching the occupant's face from the image taken inside the vehicle and reduce the processing load required for the face search. An object of the present invention is to provide a face detection device.

The face detection device according to the present disclosure includes an in-vehicle image acquisition unit that acquires an in-vehicle image that is an image of the inside of the vehicle, and a region in which a person's face is detected in the in-vehicle image by searching for a person's face from the in-vehicle image. The face detection unit includes a face detection unit that identifies a certain face area and a search area setting unit that sets a search area that is an area for searching a person's face in an in-vehicle image based on the position of the face area. For the in-vehicle image acquired after the search area is set, the face of a person is searched in the search area of the in-vehicle image.

According to the face detection device according to the present disclosure, the search area is set based on the position of the face area specified earlier, so that the position of the search area depends on the physique and posture of the occupant, the seated position, and the like. It will be in the proper position. Further, since the face detection unit searches for a face in the search area for the in-vehicle image acquired after setting the search area, the processing load required for the face search is reduced.

The purpose, features, aspects, and advantages of the present disclosure will be made clearer by the following detailed description and accompanying drawings.

It is a block diagram which shows the structure of the face detection apparatus which concerns on Embodiment 1. FIG. It is a figure which shows the example of the in-vehicle image taken by the in-vehicle photographing device. It is a figure which shows the example of the face region specified by the face detection part. It is a figure which shows the example of the search area set by the search area setting part. It is a flowchart which shows the operation of the face detection apparatus which concerns on Embodiment 1. FIG. It is a figure which shows the hardware configuration example of a face detection apparatus. It is a figure which shows the hardware configuration example of a face detection apparatus. It is a flowchart which shows the operation of the face detection apparatus which concerns on Embodiment 2. It is a flowchart which shows the operation of the face detection apparatus which concerns on Embodiment 3. It is a flowchart which shows the operation of the face detection apparatus which concerns on Embodiment 4. FIG.

<Embodiment 1>
FIG. 1 is a block diagram showing the configuration of the face detection device 10 according to the first embodiment. In the present embodiment, it is assumed that the face detection device 10 is mounted on the vehicle. However, the face detection device 10 does not have to be permanently installed in the vehicle, and may be realized on a portable device that can be brought into the vehicle, such as a mobile phone, a smartphone, or a PND (Portable Navigation Device). Further, some or all of the functions of the face detection device 10 may be realized on a server installed outside the vehicle.

As shown in FIG. 1, the face detection device 10 is connected to an in-vehicle photographing device 20 included in a vehicle equipped with the face detection device 10. The in-vehicle photographing device 20 is a camera that photographs the inside of the vehicle. The in-vehicle photographing device 20 is preferably installed at a position where a person (occupant) seated at least in a plurality of seats including the driver's seat can be photographed at the same time, such as in the vicinity of the center panel of the vehicle or the rear view mirror. FIG. 2 shows an example of an in-vehicle image taken when the in-vehicle photographing device 20 is installed near the rear view mirror. The in-vehicle image of FIG. 2 shows five people, that is, the driver's seat occupant P1, the passenger seat occupant P2, the right rear seat occupant P3, the left rear seat occupant P4, and the rear seat center occupant P5. ..

Further, as shown in FIG. 1, the face detection device 10 includes an in-vehicle image acquisition unit 11, a face detection unit 12, and a search area setting unit 13.

The vehicle interior image acquisition unit 11 acquires the vehicle interior image taken by the vehicle interior photographing device 20. The face detection unit 12 searches for a person's face from the vehicle interior image by analyzing the vehicle interior image acquired by the vehicle interior image acquisition unit 11, and obtains a "face area" which is an area in which the person's face is detected in the vehicle interior image. Identify. Hereinafter, the process of searching for the face of a person (occupant) from the image inside the vehicle may be simply referred to as "face search".

The method in which the face detection unit 12 identifies the face area may be any method. In the present embodiment, the face detection unit 12 specifies a rectangle surrounding the detected face image, specifically, a region in the rectangle consisting of four sides in contact with the contour line of the face as a face region. And. For example, when the occupants P1 to P5 are shown in the vehicle interior image as shown in FIG. 2, the face detection unit 12 identifies the face areas F1 to F5 of the occupants P1 to P5 as shown in FIG. Further, the face detection unit 12 acquires the coordinate values (coordinate values in the vehicle interior image) of each vertex of the face region, and outputs those coordinate values as "face area information" indicating the position and size of the face region. .. In the example of FIG. 3, the face detection unit 12 outputs the face area information of each of the face areas F1 to F5. The face area information output by the face detection unit 12 is output from the face detection device 10 and input to the search area setting unit 13.

The search area setting unit 13 targets an area in which the face detection unit 12 searches for a person's face in an in-vehicle image, that is, a face search target, based on the position of the face area indicated by the face area information input from the face detection unit 12. Set the "search area" which is the area to be. The search region is a region wider than the face region, which includes the face region detected by the face detection unit 12. The search area may be set by any method, and here, it is assumed that a rectangular area slightly larger than the face area is set as the search area. For example, when the face areas F1 to F5 of the occupants P1 to P5 are specified as shown in FIG. 3, the search area setting unit 13 searches for each of the face areas F1 to F5 as shown in FIG. Areas S1 to S5 are set.

When the search area is set by the search area setting unit 13, the face detection unit 12 performs a face search in the search area of the vehicle interior image for the vehicle interior image acquired thereafter. Therefore, if the search area is set small, the processing load required for face search can be reduced. However, if the search area is too small, there is a high possibility that the face will deviate from the search area when the person moves, so that the face detection accuracy will decrease. Therefore, the size of the search area is preferably as small as possible within the range in which the required face detection accuracy can be ensured.

The size of the search area may be a fixed value, but in the present embodiment, the search area setting unit 13 sets the size of the search area according to the size of the face area. For example, in the example of FIG. 4, the face areas F1 and F2 of the front seat occupants P1 and P2 shown in the foreground are large, and the face areas F3 to F5 of the rear seat occupants P3 to P5 shown in the rear are small. Large search areas S1 and S2 are set for F2, and small search areas S3 to S5 are set for the small face areas F3 to F5.

As described above, in the face detection device 10 according to the first embodiment, the search area setting unit 13 sets the search area based on the position of the face area specified by the face detection unit 12. Therefore, the position of the search area is set to an appropriate position according to the physique and posture of the occupant, the seated position, and the like. Further, since the face detection unit 12 performs a face search in the search area for the in-vehicle image acquired after the search area is set, the processing load on the face search is reduced.

As described above, the face area information (information on the position and size of the face area in the vehicle interior image) output by the face detection unit 12 is the output of the face detection device 10. The face area information output from the face detection device 10 is provided to any application that requires the information. Such applications include, for example, an occupant authentication device that authenticates the occupant's face, monitors the degree of arousal and the direction of the line of sight from the occupant's face, and warns against, for example, the driver's dozing or inattentiveness, or occupant's physical condition change. A occupant monitoring device that emits a signal, an automatic driving device that retracts the vehicle to a safe place when the driver becomes inoperable due to a physical condition, etc. can be considered.

FIG. 5 is a flowchart showing the operation of the face detection device 10 according to the first embodiment. Hereinafter, the operation of the face detection device 10 will be described with reference to FIG. In the following, for the sake of simplification of the explanation, it is assumed that only one person's face is detected from the in-vehicle image.

When the face detection device 10 is activated, the vehicle interior image acquisition unit 11 acquires the vehicle interior image from the vehicle interior photographing device 20 (step ST101), and the face detection unit 12 searches for a person's face from the vehicle interior image (step ST102). .. If the face of the person is not detected at this time (NO in step ST103), the process returns to step ST101, and steps ST101 and ST102 are repeatedly executed. On the other hand, if a person's face is detected from the vehicle interior image (YES in step ST103), the face detection unit 12 identifies the face area corresponding to the detected face, and indicates the position and size of the face area. Information is output from the face detection device 10 (step ST104). Further, the face area information is also output to the search area setting unit 13, and the search area setting unit 13 sets the search area including the face area based on the position and size of the face area (step ST105).

When the search area is set in step ST105, the in-vehicle image acquisition unit 11 acquires a new in-vehicle image from the in-vehicle image capturing device 20 (step ST106), and the face detection unit 12 causes the face detection unit 12 to acquire a person in the search area of the in-vehicle image. Search for the face of (step ST107). Since the search area is a part of the vehicle interior image, the face search processing load in step ST107 is smaller than the face search processing load in step ST102.

If a person's face is detected in step ST107 (YES in step ST108), the face detection unit 12 identifies a face area corresponding to the detected face, and provides face area information indicating the position and size of the face area. Output from the face detection device 10 (step ST109). However, if the face of a person is not detected in step ST107 (NO in step ST108), the search area setting unit 13 cancels the setting of the search area (step ST110), and the process returns to step ST101.

As described above, in the face detection device 10 according to the first embodiment, after the search area is set in step ST105, as long as the face of a person continues to be detected in step ST107, the processing load of step ST107 is small. Face detection is performed. Since the search area is set to include the position of the face (face area) detected in step ST102, basically, if the person does not move the face significantly, the face of the person continues to be detected in step ST107. , It can be expected that the processing load of the face detection device 10 will be greatly reduced.

The case where the face of the person is not detected in step ST107 is not only when the person moves and the position of the face deviates from the search area, but also when the position of the face is in the search area but the face is a hand or hair. It may be covered with. Hereinafter, the fact that the face of the person is not detected in step ST107, that is, the loss of sight of the face of the person once detected by the face detection unit 12, is referred to as "lost" of the face.

In the above description, it is assumed that only one person's face is detected from the in-vehicle image in step ST102, but as in the examples of FIGS. 2 to 4, a plurality of people's faces may be detected. In that case, a plurality of search areas are set in step ST105, and steps ST106 to ST110 are executed for each of the plurality of search areas. Further, the step ST110 may be executed for all of the plurality of search areas when the face is lost in any of the plurality of search areas.

Further, the process for first setting the search area (process of steps ST101 to ST105) may be executed by using the detection of a specific state of the vehicle as a trigger. The specific state of the vehicle referred to here is, for example, a state in which the door is locked, a state in which the vehicle door is closed, a state in which the seat belt is fastened, a state in which the vehicle has started running, and a state in which the gear is set in the drive. And so on.

6 and 7 are diagrams showing an example of the hardware configuration of the face detection device 10, respectively. Each function of the component of the face detection device 10 shown in FIG. 1 is realized by, for example, the processing circuit 50 shown in FIG. That is, the face detection device 10 acquires an in-vehicle image which is an image of the inside of the vehicle, searches for a person's face from the in-vehicle image, and identifies a face area which is an area in which the person's face is detected in the in-vehicle image. , A search area that searches for a person's face in the vehicle interior image is set based on the position of the face region, and the vehicle interior image acquired after the search area is set is within the vehicle interior image search area. The processing circuit 50 for searching the face of a person is provided. The processing circuit 50 may be dedicated hardware, or a processor (Central Processing Unit (CPU)) that executes a program stored in a memory, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, and the like. It may be configured by using a DSP (also called a Digital Signal Processor)).

When the processing circuit 50 is dedicated hardware, the processing circuit 50 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable). GateArray), or a combination of these, etc. The functions of the components of the face detection device 10 may be realized by individual processing circuits, or these functions may be collectively realized by one processing circuit.

FIG. 7 shows an example of the hardware configuration of the face detection device 10 when the processing circuit 50 is configured by using the processor 51 that executes the program. In this case, the function of the component of the face detection device 10 is realized by software (software, firmware, or a combination of software and firmware). The software or the like is described as a program and stored in the memory 52. The processor 51 realizes the functions of each part by reading and executing the program stored in the memory 52. That is, when executed by the processor 51, the face detection device 10 acquires a vehicle interior image which is an image of the inside of the vehicle, searches for a person's face from the vehicle interior image, and displays the person's face in the vehicle interior image. The process of identifying the face area, which is the detected area, the process of setting the search area, which is the area for searching the face of a person in the vehicle interior image based on the position of the face area, and the process of setting the search area, which is acquired after the search area is set. The vehicle interior image is provided with a memory 52 for storing a process of searching for a person's face in the search area of the vehicle interior image and a program to be executed as a result. In other words, it can be said that this program causes the computer to execute the procedure and method of operation of the components of the face detection device 10.

Here, the memory 52 is a non-volatile or non-volatile memory such as a RAM (RandomAccessMemory), a ROM (ReadOnlyMemory), a flash memory, an EPROM (ErasableProgrammableReadOnlyMemory), and an EEPROM (ElectricallyErasableProgrammableReadOnlyMemory). Volatile semiconductor memory, HDD (Hard Disk Drive), magnetic disk, flexible disk, optical disk, compact disk, mini disk, DVD (Digital Versatile Disc) and its drive device, etc., or any storage medium used in the future. You may.

The configuration in which the functions of the components of the face detection device 10 are realized by either hardware or software has been described above. However, the present invention is not limited to this, and a configuration in which a part of the components of the face detection device 10 is realized by dedicated hardware and another part of the components is realized by software or the like may be used. For example, for some components, the function is realized by the processing circuit 50 as dedicated hardware, and for some other components, the processing circuit 50 as the processor 51 is stored in the memory 52. It is possible to realize the function by reading and executing it.

As described above, the face detection device 10 can realize each of the above-mentioned functions by hardware, software, or a combination thereof.

<Embodiment 2>
In the first embodiment, the position of the search area set by the search area setting unit 13 is fixed unless the setting of the search area is canceled, but in the second embodiment, the search area setting unit 13 is the face detection unit. The position of the search area is changed by following the change in the position of the face area specified by 12. The configuration of the face detection device 10 in this embodiment is the same as that in FIG.

FIG. 8 is a flowchart showing the operation of the face detection device 10 according to the second embodiment. The flow of FIG. 8 is obtained by adding step ST111 after step ST109 to the flow shown in FIG. In step ST111, the search area setting unit 13 adjusts the position of the search area so as to follow the position of the face area specified by the face detection unit 12 in step ST107. Since other processes are the same as those described in the first embodiment, the description here will be omitted.

According to the face detection device 10 according to the second embodiment, the position of the search area moves following the movement of the face of the person, so that the position of the face of the person deviates from the search area and the face is lost. Is suppressed. That is, in step ST107 of FIG. 8, the period during which the face of the person continues to be detected can be lengthened, and the processing load can be further reduced. Further, since the position of the face of the person is less likely to deviate from the search area, the size of the search area can be reduced to reduce the processing load.

Further, in the present embodiment, an example is shown in which the search area setting unit 13 changes the position of the search area by following the change in the position of the face area, but further, the search area setting unit 13 is made to follow the change in the size of the face area. The size of the search area may be changed.

<Embodiment 3>
In the first and second embodiments, when the face is lost and NO is determined in step ST108, step ST110 is immediately executed and the setting of the search area is released. In the third embodiment, even if the face is lost, the search area setting unit 13 does not immediately cancel the setting of the search area, and the position of the search area is finally detected by the person's face for a certain period of time. Fix it at the time position (the position just before the face is lost). The configuration of the face detection device 10 in this embodiment is also the same as in FIG.

FIG. 9 is a flowchart showing the operation of the face detection device 10 according to the third embodiment. The flow of FIG. 9 is obtained by adding the following steps ST112 to ST116 to the flow shown in FIG. Since the other steps are the same as those described in the first and second embodiments, only steps ST112 to ST116 will be described here.

Step ST112 is executed when the face detection unit 12 loses the face and is determined to be NO in step ST108. In step ST112, the search area setting unit 13 fixes the position of the search area to the position when the face of the person was last detected in step ST102 or ST107.

In step ST113, the vehicle interior image acquisition unit 11 acquires a new vehicle interior image from the vehicle interior photographing device 20. In step ST114, the face detection unit 12 searches for the face of a person in the search area of the vehicle interior image acquired in step ST113. If a person's face is detected at this time (YES in step ST115), the face detection unit 12 determines that the same face as the lost face has been detected, and proceeds to step ST109. Hereinafter, the face search in step ST114 is referred to as "face search after lost".

On the other hand, if the face of a person is not detected in step ST114 (NO in step ST115), the search area setting unit 13 determines whether a certain period of time has passed without detecting the face (the face detection unit 12 keeps the face lost and constant). Whether or not the time has passed) is confirmed (step ST116). At this time, if a certain time has not elapsed (NO in step ST116), the process returns to step ST113. However, if a certain period of time has elapsed (YES in step ST116), the search area setting unit 13 cancels the setting of the search area (step ST110), and returns to step ST101.

For example, if a person moves and the position of the face deviates from the search area, or if the face is covered with hands or hair and the face is lost, if the person returns to the original posture, it will be before the loss occurs. It is highly possible that the face can be detected at almost the same position. Therefore, even if the face is lost, if the search area is fixed for a certain period of time, there is a high possibility that the face of the same person will be detected again in the search area. Therefore, according to the present embodiment, the frequency with which the setting of the search area is canceled can be reduced. That is, since the frequency of performing the face search in step ST102, which has a high processing load, can be reduced, it is possible to contribute to the reduction of the processing load.

In the above description, it is assumed that only one person's face is detected from the in-vehicle image in step ST102, but as in the examples of FIGS. 2 to 4, when the faces of a plurality of people are detected. There is also. In that case, a plurality of search areas are set in step ST105, and steps ST106 to ST116 are executed for each of the plurality of search areas. Further, step ST110 may be executed for all of the plurality of search areas when a certain period of time elapses while the face is lost in any of the plurality of search areas.

Further, when a plurality of search areas are set by the search area setting unit 13 and the face detection unit 12 loses a face in the search area set at a position corresponding to the driver's seat among the plurality of search areas, the face detection unit 12 It is desirable that the face search after loss (step ST114) in the search area corresponding to the driver's seat is executed with a higher priority than the face search in the other search areas. For example, it is conceivable to execute the face search after the lost in the search area corresponding to the driver's seat with a higher frequency (shorter cycle) than the face search in the other search areas.

<Embodiment 4>
As can be seen from the examples of FIGS. 2 to 4, the faces of the occupants in the front seats and the faces of the occupants in the rear seats tend to be close to each other in the in-vehicle image. Therefore, it is possible that two or more faces are included in one search area. The face detection unit 12 can distinguish the face area corresponding to each face by tracking the position of each face area, whereby two or more faces in one search area are included. However, it is prevented that the face detection unit 12 confuses those faces.

However, when the face is lost in the third embodiment, the tracking of the position of the face region is interrupted, so that in the face search after the loss (step ST114), the face of a person other than the lost face is erroneous. May be detected. For example, in the example of FIG. 4, when the face detection unit 12 loses the face of the occupant P1 in the driver's seat and the face search after the loss is performed, the occupant P3 in the right rear seat moves and the face of the occupant P3 becomes the search area. If the vehicle enters S1, the face of the occupant P3 in the right rear seat may be erroneously detected as the face of the occupant P1 in the driver's seat.

Therefore, in the fourth embodiment, the face detection unit 12 compares the lost face with the size of the face detected in the face search after the lost, so that the face detected in the face search after the lost is the lost face. Determine if it is the same as. Specifically, when the face of a person is newly detected from the search area after the face of the person is no longer detected from the search area, the face detection unit 12 does not detect the face of the person before the face of the person is no longer detected (face). If the difference between the size of the face area (before lost) and the size of the face area corresponding to the face of the newly detected person is equal to or greater than a predetermined threshold, the detection result of the face of the newly detected person Is invalid. The configuration of the face detection device 10 in this embodiment is also the same as in FIG.

FIG. 10 is a flowchart showing the operation of the face detection device 10 according to the fourth embodiment. The flow of FIG. 10 is obtained by adding the following steps ST117 and ST118 to the flow shown in FIG. Since the other steps are the same as those described in the first to third embodiments, only steps ST117 and ST118 will be described here.

Step ST117 is executed when a person's face is detected in the face search after lost (step ST114) and a YES is determined in step ST115. In step ST117, the face detection unit 12 determines the size of the face area before the face is lost (the size of the face area last specified in step ST103 or ST107) and the face newly detected in step ST114. By comparing with the size of the corresponding face area, it is determined whether or not the difference between the two is equal to or greater than a predetermined threshold value. If the difference between the two is smaller than the threshold value (NO in step ST117), the face detection unit 12 determines that the same face as the lost face could be detected, and proceeds to step ST109.

Step ST118 is executed when the difference from the size of the face area corresponding to the face newly detected in step ST114 is equal to or greater than the threshold value (when YES is determined in step ST117). In step ST118, the face detection unit 12 determines that a face different from the lost face was erroneously detected in step ST114, and invalidates the face detection result in step ST114, that is, the face is detected in step ST114. It is considered that it did not exist. After step ST118, the process proceeds to step ST116.

According to the fourth embodiment, the face detection unit 12 and the lost face are based on the difference between the size of the face area specified before the face is lost and the size of the face area newly specified after the face is lost. It is possible to prevent false detection of different faces.

It is possible to freely combine each embodiment, and to appropriately modify or omit each embodiment.

It is understood that the above description is an example in all aspects, and innumerable variations that are not illustrated can be assumed.

10 face detection device, 11 in-vehicle image acquisition unit, 12 face detection unit, 13 search area setting unit, 20 in-vehicle photography device, 50 processing circuit, 51 processor, 52 memory, P1 to P5 occupants, F1 to F5 face area, S1 to S5 Search area.

Claims

An in-vehicle image acquisition unit that acquires an in-vehicle image that is an image of the inside of the vehicle,
A face detection unit that searches for a person's face from the in-vehicle image and identifies a face area that is an area in which the person's face is detected in the in-vehicle image.
A search area setting unit that sets a search area that is an area for searching a person's face in the vehicle interior image based on the position of the face area, and a search area setting unit.
With
The face detection unit searches for a person's face in the search area of the vehicle interior image with respect to the vehicle interior image acquired after the search area is set.
Face detector.
The search area setting unit sets the size of the search area according to the size of the face area.
The face detection device according to claim 1.
The search area setting unit changes the position of the search area in accordance with a change in the position of the face area.
The face detection device according to claim 1.
When the face of the person is no longer detected from the search area, the search area setting unit fixes the position of the search area to the position when the face of the person was last detected.
The face detection device according to claim 3.
When a certain period of time elapses while the person's face is not detected from the search area, the search area setting unit cancels the setting of the search area, and the face detection unit searches for the person's face from the entire vehicle interior image. ,
The face detection device according to claim 4.
When a person's face is newly detected from the search area after the person's face is no longer detected from the search area, the face detection unit has the size of the face area before the person's face is no longer detected. If the difference from the size of the face area corresponding to the newly detected face of the person is equal to or greater than a predetermined threshold value, the detection result of the newly detected face of the person is invalidated.
The face detection device according to claim 1.
When the faces of a plurality of persons are detected from the in-vehicle image, the face detection unit identifies a plurality of the face areas corresponding to each of the faces of the plurality of persons, and the search area setting unit performs the plurality of said. A plurality of the search areas corresponding to each of the face areas are set, and a plurality of the search areas are set.
The face detection unit searches for a person's face in each of the search areas for the in-vehicle image acquired after the plurality of search areas are set.
The face detection device according to claim 1.
When the face of a person is no longer detected from the search area corresponding to the driver's seat of the vehicle, the face detection unit has a higher priority than the other search areas and the person in the search area corresponding to the driver's seat. Explore the face of
The face detection device according to claim 7.
The search area setting unit sets the search area by using the detection of a specific state of the vehicle as a trigger.
The face detection device according to claim 1.
The specific state of the vehicle may be one or more of a locked door, a closed door, a seatbelt fastened, a start running, and a gear set on the drive. The face detection device according to claim 9, which includes.
The in-vehicle image acquisition unit of the face detection device acquires the in-vehicle image, which is an image of the inside of the vehicle.
The face detection unit of the face detection device searches for a person's face from the in-vehicle image, identifies a face area in which the person's face is detected in the in-vehicle image, and identifies the face area.
The search area setting unit of the face detection device sets a search area, which is an area for searching for a person's face in the in-vehicle image, based on the position of the face area.
The face detection unit searches for a person's face in the search area of the vehicle interior image with respect to the vehicle interior image acquired after the search area is set.
Face detection method.