JP7314085B2 - Image processing device, computer program, and anomaly estimation system - Google Patents

Description

The present invention relates to an image processing apparatus that processes an image in which a driver of a mobile object is captured. The present invention also relates to a computer program executed by a processing section of the image processing apparatus. The present invention also relates to an abnormality estimation system for estimating whether the posture of the driver is abnormal based on the image.

Patent Literature 1 discloses a technique for detecting an abnormality in the posture of a driver of a vehicle, which is an example of a moving body. The posture abnormality is estimated based on an image of the driver captured by an imaging device mounted on the vehicle.

JP 2019-105872 A

SUMMARY OF THE INVENTION An object of the present invention is to increase the likelihood of estimating an abnormal posture of a driver of a mobile object based on an image acquired by an imaging device.

One aspect for achieving the above object is an image processing device,
a reception unit that receives image information corresponding to an image in which the driver of the moving object is captured;
a processing unit that estimates whether the posture of the driver is abnormal based on the image information;
and
The processing unit, based on the image information,
identifying a plurality of feature points on the driver's upper body reflected in the image;
The posture is estimated to be abnormal when the amount of decrease in the distance in the direction corresponding to the vertical direction of the moving object between the plurality of feature points reflected in the image exceeds a threshold.

One aspect for achieving the above object is a computer program executable by a processing unit of an image processing apparatus,
By being executed, the image processing device
receiving image information corresponding to an image in which a driver of a mobile object is captured;
Based on the image information,
specifying a plurality of feature points on the upper body of the driver reflected in the image;
If the amount of decrease in the distance in the direction corresponding to the vertical direction of the moving body between the plurality of feature points reflected in the image is below a threshold, it is estimated that the driver's posture is abnormal.

One aspect for achieving the above object is an anomaly estimation system,
an imaging device that outputs image information corresponding to an image in which a driver of a mobile object is captured;
an image processing device that estimates whether the posture of the driver is abnormal based on the image information;
a control device that controls the operation of a controlled device mounted on the moving body based on the estimation result that the posture is abnormal by the image processing device;
and
The image processing device, based on the image information,
identifying a plurality of feature points on the driver's upper body reflected in the image;
The posture is estimated to be abnormal when the amount of decrease in the distance in the direction corresponding to the vertical direction of the moving object between the plurality of feature points captured in the image is below a threshold.

For example, there are abnormal postures, such as the “prostrate posture” defined by the Ministry of Land, Infrastructure, Transport and Tourism, in which the face is less likely to appear in the image acquired by the imaging device. In this case, it may be difficult to estimate the abnormal posture based on the direction and position of the face. However, in the configuration according to each of the above aspects, by selecting a plurality of feature points from the driver's upper body at which the distance in the vertical direction of the moving object decreases as the vehicle deviates from the normal posture, the direction and position of the face affect the estimation of the abnormal posture. As a result, it is possible to increase the likelihood of estimating the abnormal posture of the driver based on the image acquired by the imaging device.

1 illustrates a functional configuration of an abnormality estimation system according to one embodiment; 1 illustrates a vehicle in which the abnormality estimation system of FIG. 1 can be installed. 2 illustrates the flow of processing executed by the image processing apparatus of FIG. 1; 2 illustrates an image that may be acquired by the imaging device of FIG. 1; The image in FIG. 4 illustrates a state in which a skeleton model is applied. It is a figure for demonstrating a "push-down posture." 4 shows an example of the flow of posture estimation processing in FIG. 3 ; FIG. 11 is a diagram for explaining details of posture estimation processing; FIG. 11 is a diagram for explaining details of posture estimation processing; It illustrates a method of identifying the rotation angle of the head in the pitch direction of the vehicle. It is a figure for demonstrating the "sideways posture" to the left. It is a figure for demonstrating the "lateral fall posture" to the right.

Exemplary embodiments are described in detail below with reference to the accompanying drawings. FIG. 1 illustrates the functional configuration of an abnormality estimation system 10 according to one embodiment. The abnormality estimation system 10 is a system that estimates whether the posture of the driver 30 is abnormal based on the image of the vehicle 20 illustrated in FIG. 2 in which the driver 30 is reflected. Vehicle 20 is an example of a mobile object.

Posture anomalies are estimated to detect anomalies of driver 30 . The term "driver's abnormality" used in this specification means a sudden change in physical condition that is difficult to predict in advance.

In the accompanying drawings, arrow F represents the forward direction as seen from driver 30 . Arrow B represents the rearward direction viewed from driver 30 . An arrow L indicates the left direction as seen from the driver 30 . An arrow R indicates the right direction as seen from the driver 30 . An arrow U represents an upward direction as seen from the driver 30 . An arrow D represents the downward direction viewed from the driver 30 .

As illustrated in FIG. 1 , an abnormality estimation system 10 includes an imaging device 11 . The imaging device 11 is arranged at an appropriate location on the vehicle 20 . A driver 30 seated on a seat 22 arranged in a cabin 21 of a vehicle 20 illustrated in FIG. 2 is imaged by the imaging device 11 .

As illustrated in FIG. 1, the imaging device 11 is configured to output image information I corresponding to the captured image. The image information I may be in the form of analog data or in the form of digital data.

The abnormality estimation system 10 includes an image processing device 12 . The image processing device 12 includes a reception section 121 and a processing section 122 . The reception unit 121 is configured to receive the image information I from the imaging device 11 . If the image information I is in the form of analog data, the reception unit 121 can include an appropriate conversion circuit including an A/D converter. The processing unit 122 processes image information I in the form of digital data.

The processing unit 122 is configured to execute processing for estimating whether the posture of the driver 30 is abnormal based on the image information I. Details of the processing will be described later.

The image processing device 12 has an output unit 123 . The processing unit 122 is configured to output the control information C through the output unit 123 when it is estimated that the posture of the driver 30 is abnormal. The control information C may be in the form of digital data or in the form of analog data. If the control information C is in the form of analog data, the output section 123 may include suitable conversion circuitry including a D/A converter.

The abnormality estimation system 10 includes a control device 13 . The control device 13 is mounted on the vehicle 20 . The control device 13 is configured to control the operation of the controlled device 40 mounted on the vehicle 20 based on the control information C output from the image processing device 12 .

Specifically, the control device 13 is configured to enable driving assistance for the vehicle 20 when it is estimated that the posture of the driver 30 is abnormal. The term “driving assistance” as used herein means control processing that at least partially performs at least one of driving operation (steering, acceleration, deceleration, etc.), monitoring of the driving environment, and driving operation backup. In other words, it includes everything from partial driving assistance such as the collision damage mitigation brake function and lane keeping assist function to fully automated driving operations.

For example, based on the control information C output from the image processing device 12, the control device 13 causes the controlled device 40 to decelerate and stop the vehicle 20 on the shoulder of the road. Examples of the controlled device 40 include a device that configures the driving system of the vehicle 20, a lamp, and a device that notifies passengers and pedestrians of the vehicle 20 and other vehicles that the driving support operation is effective.

Next, the processing for estimating the posture of the driver 30 executed by the processing unit 122 of the image processing device 12 will be described in detail with reference to FIGS. 3 to 5. FIG. FIG. 3 exemplifies the flow of the processing.

The processing unit 122 receives the image information I from the imaging device 11 through the receiving unit 121 (STEP 1). FIG. 4 exemplifies an image IM in which the driver 30 is captured by the imaging device 11 . Image information I corresponds to image IM.

Subsequently, the processing unit 122 performs processing for applying the skeleton model (STEP 2 in FIG. 3). As used herein, the term “processing to apply a skeleton model” means detecting a plurality of feature points defined in the skeleton model in the driver captured in the image acquired by the imaging device, and connecting the plurality of feature points with a plurality of skeleton lines defined in the skeleton model.

FIG. 5 shows an example in which the skeletal model M is applied to the driver 30 appearing in the image IM acquired by the imaging device 11 . In this example, the skeleton model M includes a head feature point H, a neck feature point NK, a left shoulder feature point LS, and a right shoulder feature point RS.

The head feature point H is a point corresponding to the center of the head of the model human body. The neck feature point NK is a point corresponding to the neck of the model human body. The left shoulder feature point LS is a point corresponding to the left shoulder of the model human body. The right shoulder feature point RS is a point corresponding to the right shoulder of the model human body. The head minutiae H and the neck minutiae NK are connected by a skeleton line. The neck feature point NK is connected to each of the left shoulder feature point LS and the right shoulder feature point RS by skeleton lines.

The processing unit 122 detects points corresponding to each of the head feature point H, the neck feature point NK, the left shoulder feature point LS, and the right shoulder feature point RS in the driver 30 captured in the image IM, and connects the plurality of detected points with the skeleton lines described above. Algorithms for performing this process are well known and will not be described in detail.

As illustrated in FIG. 1 , the image processing device 12 includes a storage section 124 . The processing unit 122 stores the position of each feature point in the image IM in the storage unit 124 as representing the posture of the driver 30 in the normal state. Specifically, the position of the pixel in the image IM corresponding to each feature point is saved in the storage unit 124 .

Subsequently, the processing unit 122 executes posture estimation processing for estimating whether the posture of the driver 30 is abnormal (STEP 3 in FIG. 3). The details of the orientation estimation process will be described later.

As a result of the posture estimation processing, when it is estimated that the posture of the driver 30 is not abnormal (NO in STEP4), the processing returns to STEP1, and the next image information I is accepted. The cycle in which the reception of the image information I is repeated can correspond to the frame rate of the imaging device 11 .

When it is determined that the posture of driver 30 is abnormal as a result of posture estimation processing (YES in STEP4), processing unit 122 outputs control information C from output unit 123 (STEP5). Control information C is transmitted to the control device 13 . The control information C may cause the controlled device 40 to perform the same action, or may cause the controlled device 40 to perform different actions depending on the type of the estimated abnormal posture.

Next, details of posture estimation processing (STEP 3 in FIG. 3) executed by the processing unit 122 will be described with reference to FIGS. 6 to 9. FIG. In this example, it is estimated whether the driver 30 is in a "bent down posture".

The “bent-down posture” is one of multiple types of “posture collapse patterns” defined by the Ministry of Land, Infrastructure, Transport and Tourism. A "bent down posture" is defined as a state in which the driver continues to fall forward with his/her face positioned near the steering wheel.

FIG. 6 exemplifies the "prostrate position". The “bent down posture” is defined as a state in which the forward movement amount ΔDF of the vehicle 20 from the normal state exceeds the threshold, the downward movement amount ΔDD of the vehicle 20 exceeds the threshold, and the downward pitch angle θPD of the driver's face exceeds the threshold. A threshold value of ΔDF is, for example, 200 mm. A threshold value of ΔDD is, for example, 180 mm. A threshold value of θPD is, for example, 30°.

FIG. 7 shows an example of the flow of posture estimation processing. FIG. 8 shows an enlarged portion of the image IM illustrated in FIGS. 4 and 5 in which the upper half of the body of the driver 30 is reflected. In this example, the process of applying the skeletal model M can also detect nose feature points NS. The nose feature point NS is a point corresponding to the nose of the model human body.

First, the processing unit 122 identifies the distance d in the direction corresponding to the vertical direction of the vehicle 20 between the nose feature point NS and the left shoulder feature point LS of the driver 30 captured in the image IM acquired by the imaging device 11 (STEP 11 in FIG. 7). The nose feature point NS and the left shoulder feature point LS are examples of multiple feature points on the upper body. The distance d is specified by the number of pixels. The specified value of the distance d is stored in the storage unit 124 .

Subsequently, the processing unit 122 compares the value of the distance d specified in STEP 11 with the value of the distance d stored in the storage unit 124, and determines whether the amount of decrease exceeds the threshold value dth (STEP 12).

FIG. 9 illustrates the change over time of the distance d between the nose feature point NS and the left shoulder feature point LS when the driver 30 takes the "bent down posture". As the driver 30 falls forward, the distance d in the vertical direction of the vehicle 20 between the nose feature point NS and the left shoulder feature point LS rapidly decreases. In the illustrated example, the distance d starts to decrease at time t1, and the amount of decrease exceeds the threshold dth at time t2.

Therefore, when it is determined that the amount of decrease in the vertical distance d of the vehicle 20 between the nose feature point NS and the left shoulder feature point LS exceeds the threshold value dth (YES in STEP 12), the processing unit 122 estimates that the driver 30 is in the "downward posture" (STEP 13). The processing is reflected in the YES determination in STEP4 of FIG. The threshold dth can be statistically determined as the amount of decrease in the distance d that can occur when the general driver 30 assumes a “bent posture”.

On the other hand, if it is determined that the amount of decrease in distance d in the vertical direction of vehicle 20 between nose feature point NS and left shoulder feature point LS does not exceed threshold value dth (NO in STEP 12), processing unit 122 estimates that driver 30 is not in the "bent down posture" (STEP 14). The processing is reflected in the determination of NO in STEP4 of FIG.

In view of the definition of the “bent down posture” shown in FIG. 6, it is preferable that the orientation and position of the face of the driver 30 are specified. However, when the driver 30 actually takes the "bent down posture", the face is less likely to appear in the image IM, and it may be difficult to identify the direction and position of the face. In the configuration described above, by selecting a plurality of characteristic points from the upper body of the driver 30 at which the distance in the vertical direction of the vehicle 20 decreases when the "bent down posture" is taken, the influence of the direction and position of the face on the estimation of the "bent down posture" is suppressed. As a result, it is possible to increase the likelihood of estimating the posture abnormality of the driver 30 based on the image IM acquired by the imaging device 11 .

As described above, depending on the direction and position of the driver's 30 face, it may be difficult to specify the nose feature point NS. Therefore, as exemplified in FIG. 7, the processing unit 122 can determine whether or not the nose feature point NS cannot be specified (STEP 15).

If the nose feature point NS has been identified (NO in STEP15), the process proceeds to STEP12, and as described above, it is determined whether the amount of decrease in the distance d in the vertical direction of the vehicle 20 between the nose feature point NS and the left shoulder feature point LS exceeds the threshold value dth.

If the nose feature point NS cannot be specified (YES in STEP 15), as illustrated in FIG. 8, the processing unit 122 specifies the distance d′ in the vertical direction of the vehicle 20 between the head feature point H and the left shoulder feature point LS (STEP 16). The head feature point H and the left shoulder feature point LS are examples of a plurality of feature points on the upper body. The distance d' is specified by the number of pixels. The specified value of distance d′ is stored in storage unit 124 .

In this case, the processing unit 122 compares the value of the distance d′ specified in STEP 16 with the value of the distance d′ stored in the storage unit 124, and determines whether the amount of decrease exceeds the threshold dth′ (STEP 12). The threshold dth can be statistically determined as the amount of decrease in the distance d' that can occur when the general driver 30 takes a "bent down posture".

If it is determined that the amount of decrease in the distance d' in the vertical direction of the vehicle 20 between the head feature point H and the left shoulder feature point LS exceeds the threshold value dth' (YES in STEP 12), the processing unit 122 estimates that the driver 30 is in a "bent down posture" (STEP 13). The processing is reflected in the YES determination in STEP4 of FIG.

On the other hand, if it is determined that the amount of decrease in the distance d' in the vertical direction of the vehicle 20 between the head feature point H and the left shoulder feature point LS does not exceed the threshold dth' (NO in STEP 12), the processing unit 122 estimates that the driver 30 is not in the "bent down posture" (STEP 14). The processing is reflected in the determination of NO in STEP4 of FIG.

According to such a configuration, it is possible to further suppress the influence of the direction and position of the face on the estimation of the "bent down posture".

As illustrated in FIG. 8 , a rectangular frame FM for estimating the region where the head 31 of the driver 30 is located in the image IM can be set by applying the skeleton model M to the driver 30 captured in the image IM acquired by the imaging device 11. In this case, the processing unit 122 can detect the rotation angle of the head 31 of the driver 30 reflected in the image IM in the pitch direction of the vehicle 20 .

As used herein, the term “pitch direction” refers to the direction of rotation about an axis extending in the left-right direction of vehicle 20 . Regarding the "pitch direction", the counterclockwise direction viewed from the left side of the driver 30 is defined as the "lower pitch direction", and the clockwise direction viewed from the left side of the driver 30 is defined as the "upper pitch direction".

FIG. 10 is a diagram for explaining a method of assuming the center coordinates CT of the head 31 using the set frame FM. In the figure, the state of the head 31 viewed from the left is schematically shown. In this example, the width of the head 31 in the vertical direction of the vehicle 20 and the width of the head 31 in the longitudinal direction of the vehicle 20 are considered to be the same as the width W of the frame FM shown in FIG. Under this premise, the central coordinate CT of the head 31 is assumed to be a point (W/2) away from the head feature point H to the rear of the vehicle 20 . Processing unit 122 stores the number of pixels corresponding to the distance (W/2) in storage unit 124 .

At this time, the angle θP0 formed by the straight line connecting the center coordinate CT and the nose feature point NS with respect to the straight line connecting the center coordinate CT and the head feature point H can be specified by obtaining the arctangent of (W/2) and the distance dHN between the head feature point H and the nose feature point NS in the lateral direction of the vehicle 20. This angle θP0 is stored in storage unit 124 as an initial angle when driver 30 is in a normal posture.

When the driver 30 assumes the “bent down posture”, the head 31 rotates in the pitch direction of the vehicle 20 . At this time, the angle θP1 formed by the straight line connecting the central coordinate CT and the nose feature point NS with respect to the straight line connecting the central coordinate CT and the head feature point H (the longitudinal direction of the vehicle 20) changes from the initial angle θP0. The amount of change (θP1−θP0) can be regarded as the rotation angle θP of the head 31 in the pitch direction of the vehicle 20. FIG. Therefore, when the absolute value of the amount of change (θP1−θP0) exceeds the threshold, it can be estimated that the driver 30 is in a “bent down posture”. The threshold can be determined as appropriate based on definitions by the Ministry of Land, Infrastructure, Transport and Tourism.

When the processing unit 122 estimates that the driver 30 is in the “bent down posture” (STEP 13 in FIG. 7), the processing unit 122 may perform processing to identify the rotation angle θP of the head 31 in the downward pitch direction of the vehicle 20 based on the above principle. Specifically, as exemplified in FIG. 8, the distance dHN between the head feature point H and the nose feature point NS of the driver 30 captured in the acquired image IM in the vertical direction of the vehicle 20 is specified. The distance dHN is specified as the number of pixels. Subsequently, the processing unit 122 determines the angle θP1 by obtaining the arctangent of the distance (W/2) and the distance dHN stored in the storage unit 124 .

Furthermore, the processing unit 122 determines the rotation angle θP of the head 31 in the downward pitch direction of the vehicle 20 by taking the difference value between the specified angle θP1 and the initial angle θP0. Processing unit 122 determines whether the absolute value of the identified rotation angle θP exceeds the threshold value stored in storage unit 124 .

When the absolute value of the specified rotation angle θP exceeds the threshold, the processing unit 122 confirms the estimation that the driver 30 is in the "bent down posture".

On the other hand, if the absolute value of the identified rotation angle θP does not exceed the threshold even though the distance d in the direction corresponding to the vertical direction of the vehicle 20 between the nose feature point NS and the left shoulder feature point LS is estimated to be taken by the driver 30, the processing unit 122 cancels the estimation that the driver 30 is in the "downward posture." For example, such a condition may be satisfied when the driver 30 leans his upper body forward and places his chin on the steering wheel while looking straight ahead. This position does not fall under the category of "push-down position".

Therefore, according to the configuration as described above, it is possible to further increase the probability of estimating the posture abnormality of the driver 30 based on the image IM acquired by the imaging device 11 .

The processing unit 122 having each function described so far can be implemented by a general-purpose microprocessor operating in cooperation with a general-purpose memory. A CPU, MPU, and GPU can be exemplified as a general-purpose microprocessor. Examples of general-purpose memory include ROM and RAM. In this case, the ROM may store a computer program for executing the above-described process. ROM is an example of a storage medium that stores computer programs. The processor designates at least part of the computer program stored in the ROM, develops it on the RAM, and cooperates with the RAM to execute the above-described processing. The above computer program may be pre-installed in a general-purpose memory, or may be downloaded from an external server via a communication network and installed in a general-purpose memory. In this case, the external server is an example of a storage medium storing computer programs.

The processing unit 122 may be implemented by a dedicated integrated circuit such as a microcontroller, ASIC, FPGA, etc., capable of executing the above computer programs. In this case, the computer program is pre-installed in a storage element included in the dedicated integrated circuit. The storage element is an example of a storage medium storing a computer program. Processing unit 122 may also be implemented by a combination of a general-purpose microprocessor and a dedicated integrated circuit.

Storage unit 124 can be realized by a semiconductor memory or a hard disk device. The storage unit 124 may be implemented by the general-purpose memory or storage element described above.

The above embodiments are merely examples for facilitating understanding of the present invention. The configurations according to the above embodiments can be modified and improved as appropriate without departing from the scope of the present invention.

In the above embodiment, the nose feature point NS or the head feature point H and the left shoulder feature point LS are used to estimate whether the driver 30 is in the "bent down posture". However, an appropriate combination can be selected from those included in the upper body of the driver 30 as long as the feature points are such that the distance in the vertical direction of the vehicle 20 decreases as the vehicle 20 deviates from the normal posture. Examples of such combinations include nose and neck minutiae, left ear and left shoulder minutiae, and the like.

By appropriately selecting a plurality of feature points from the upper body of the driver 30 where the distance in the vertical direction of the vehicle 20 decreases as the vehicle 20 deviates from the normal posture, the processing unit 122 of the image processing device 12 can also estimate whether the driver 30 is in a "sideways posture."

The "sideways posture" is one of multiple types of "posture collapse patterns" defined by the Ministry of Land, Infrastructure, Transport and Tourism. A "sideways posture" is defined as a state in which the driver's upper body is tilted leftward or rightward and the driver's face is also tilted in the same direction.

FIG. 11 exemplifies a "sideways posture" to the left. The leftward “sideways posture” is defined as a state in which the leftward movement amount ΔDL of the vehicle 20 from the normal state at a specific position on the driver's face exceeds a threshold and the leftward roll angle θRL of the driver's face exceeds a threshold. A threshold value of ΔDL is, for example, 200 mm. A threshold value of θRL is, for example, 15°.

FIG. 12 exemplifies a "sideways posture" to the right. The rightward "sideways posture" is defined as a state in which the rightward movement amount ΔDR of the vehicle 20 from the normal state at a specific position on the driver's face exceeds a threshold, and the rightward roll angle θRR of the driver's face exceeds a threshold. A threshold value of ΔDR is, for example, 200 mm. A threshold value of θRR is, for example, 15°.

Irrespective of such definition, when the driver 30 takes a "sideways posture", for example, the distance d in the vertical direction of the vehicle 20 between the head feature point H and the left shoulder feature point LS tends to decrease. Therefore, it can be estimated whether the driver 30 is in a "sideways posture" by appropriately setting the threshold value dth related to the amount of decrease.

The image processing device 12 may be mounted on the vehicle 20, or may be provided as an external device capable of communicating with the vehicle 20 via a known wireless communication network. When image processing device 12 is mounted on vehicle 20, image processing by image processing device 12 and control processing by control device 13 may be performed by a common device or element. When the image processing device 12 is provided as an external device capable of communicating with the vehicle 20 via a known wireless communication network, the image information I output from the imaging device 11 is transmitted to the image processing device 12 via the wireless communication network. Control information C that can be output as a result of posture estimation processing by the processing unit 122 is transmitted to the control device 13 via a well-known wireless communication network.

The abnormality estimation system 10 can also be applied to mobile bodies other than the vehicle 20 . Examples of other mobile objects include railroads, aircraft, ships, and the like.

10: Abnormality estimation system 11: Imaging device 12: Image processing device 121: Receiving unit 122: Processing unit 13: Control device 20: Vehicle 30: Driver 31: Head 40: Controlled device H: Head feature point I: Image information IM: Image LS: Left shoulder feature point NS: Nose feature point d, d': Distance dth, dth': Threshold θP: Rotation angle in pitch direction

Claims (8)

a reception unit that receives image information corresponding to an image in which the driver of the moving object is captured;
a processing unit that estimates whether the posture of the driver is abnormal based on the image information;
and
The processing unit, based on the image information,
identifying a plurality of feature points on the driver's upper body reflected in the image;
estimating that the posture is abnormal when an amount of decrease in distance in a direction corresponding to the vertical direction of the moving body between the plurality of feature points reflected in the image exceeds a threshold;
Image processing device. The processing unit estimates that the driver is in a prone posture when the amount of decrease in the distance exceeds the threshold.
The image processing apparatus according to claim 1. The processing unit is
identifying a rotation angle of the driver's head reflected in the image in the pitch direction of the moving object;
If the amount of decrease in the distance exceeds the threshold and the rotation angle exceeds the threshold, it is estimated that the driver is in a prone posture.
The image processing apparatus according to claim 1 or 2. wherein the plurality of feature points correspond to the driver's nose and shoulders;
The image processing apparatus according to any one of claims 1 to 3. The plurality of feature points correspond to the driver's head center and shoulders.
The image processing apparatus according to any one of claims 1 to 4. configured to be mounted on the mobile body,
The image processing apparatus according to any one of claims 1 to 5. A computer program executable by a processing unit of an image processing device,
By being executed, the image processing device
receiving image information corresponding to an image in which a driver of a mobile object is captured;
Based on the image information,
specifying a plurality of feature points on the upper body of the driver reflected in the image;
estimating that the posture of the driver is abnormal when a decrease in distance in a direction corresponding to the vertical direction of the moving body between the plurality of feature points reflected in the image is below a threshold;
computer program. an imaging device that outputs image information corresponding to an image in which a driver of a mobile object is captured;
an image processing device that estimates whether the posture of the driver is abnormal based on the image information;
a control device that controls the operation of a controlled device mounted on the moving body based on the estimation result that the posture is abnormal by the image processing device;
and
The image processing device, based on the image information,
identifying a plurality of feature points on the driver's upper body reflected in the image;
estimating that the posture is abnormal when an amount of decrease in distance in a direction corresponding to the vertical direction of the moving object between the plurality of feature points captured in the image is below a threshold;
Anomaly estimation system.

Images