JP7053340B2 - Processing equipment and programs - Google Patents

Description

The present invention relates to a processing device and a program, and more particularly to a processing device and a program that perform detection processing based on a captured image.

Conventionally, there is a technique of capturing an image reflected by a vehicle occupant and processing the captured image to identify a detection target. In this type of technology, it is required to detect the driver's head for the purpose of estimating the driver's physical condition and arousal level. In order to detect the driver's head, a technique has been proposed in which a pre-learned classifier is collated with an image taken (see, for example, Patent Document 1).

The processing device of Patent Document 1 includes an image pickup unit that captures an image to be detected by an operator who operates an operation unit of an in-vehicle device that can be operated by a driver and a passenger, and image data output from the image pickup unit. A differential processing unit that performs differential processing based on this to create a horizontal differential image and a vertical differential image, and a first polar coordinate system having the operation unit as a coordinate origin are set, and the horizontal differential image and the vertical direction are set. A calculation unit that calculates the first angular and radial differentials based on the differential image, the calculated first angular differential, the first radial differential, and the first. The binary image creating unit that creates the first binary image based on the conditions, and the first and second straight lines that have a first interval and are substantially parallel based on the first binary image. An operator discrimination device is configured to include an extraction unit for extracting a straight line and a discrimination unit for discriminating the operator based on the extracted first and second straight lines.

Japanese Unexamined Patent Publication No. 2009-294843

However, in the technique of Patent Document 1, the amount of processing in the processing unit, the detection unit, the creation unit, the extraction unit, and the discrimination unit is large, and in particular, in the processing using the three-dimensional pixel group, the calculation cost is increased for the processing and the calculation. There is a problem that it is necessary.

Therefore, an object of the present invention is to provide a processing device and a program for reducing a calculation cost in a technique for detecting an occupant's head.

[1] In order to achieve the above object, an acquisition unit for acquiring a distance image in which an imaging region including a vehicle occupant is an imaging target and in which distance information to the imaging target is assigned to each pixel is used. A pixel group constituting the distance image, which is extracted according to the correspondence between the coordinate system in the vehicle interior space of the vehicle and the coordinate system in the distance image, and has a high possibility that the head of the occupant exists. Provided is a processing device having a control unit for executing a specific process for specifying the head of the occupant with respect to the high likelihood region.
[2] The control unit divides a pixel group constituting the distance image into a plurality of regions corresponding to a three-dimensional space, and within the divided regions, provides distance information indicating a distance within a specified range. The processing apparatus according to the above [1] may have a region having the largest number of pixels as the high likelihood region.
[3] Further, the control unit may be the processing device according to the above [2], which divides a pixel group constituting the distance image along a specific coordinate axis in the vehicle coordinate system.
[4] Further, the control unit is a head model specified in advance as a head identification process for specifying the crown of the occupant from the high likelihood region and a head model with the head as a base point. The above-mentioned [1] to [ The processing apparatus according to any one of 3] may be used.
[5] Further, in the crown specifying process, the control unit specifies, in the high likelihood region, the highest portion in the coordinate system in the vehicle interior space of the vehicle along the vehicle height direction as the crown. The processing apparatus according to the above [4] may be used.
[6] Further, even if the processing device according to the above [4] or [5], the control unit detects the position of the center of gravity of the head region as the position of the head in the head specifying process. good.
[7] The processing device according to any one of [1] to [6] above, wherein the control unit executes a verification process for confirming the certainty of the identified head of the occupant. May be good.
[8] Further, when the control unit detects a pixel group existing below the specified crown in the coordinate system in the vehicle interior space of the vehicle as the body portion of the occupant, the specified occupant's body portion The processing apparatus according to the above [7] may be used as the verification process for ensuring that the head is probable.
[9] Further, the control unit executes the specific process again as the verification process, assuming that the head of the specified occupant is uncertain when the body portion of the occupant is not detected. The processing apparatus according to the above [7] or [8] may be used.
[10] Further, the control unit has a specified number of pixels having distance information indicating a distance in a specified range in a body estimation area indicating a body area with the head of an occupant specified in the head specifying process as a base point. The processing device according to the above [8] or [9] may be used, which detects the body estimation area as the body portion of the occupant when is detected.
[11] In order to achieve the above object, an acquisition step of acquiring a distance image in which an imaging region including a vehicle occupant is an imaging target and a distance information to the imaging target is assigned to each pixel, and an acquisition step. A pixel group constituting the distance image, which is extracted according to the correspondence between the coordinate system in the vehicle interior space of the vehicle and the coordinate system in the distance image, and has a high possibility that the head of the occupant exists. A control step for executing a specific process for specifying the head of the occupant and a program for causing the computer to execute the specific process for specifying the head of the occupant are provided for the high likelihood region.
[12] In the control step, the pixel group constituting the distance image is divided into a plurality of regions corresponding to the three-dimensional space, and the distance information indicating the distance within the specified range is provided in the divided regions. The program according to the above [11] may be used in which the region having the largest number of pixels is set as the high likelihood region.
[13] Further, the control step may be the program according to the above [12], which divides a pixel group constituting the distance image along a specific coordinate axis in the vehicle coordinate system.
[14] Further, the control step includes a head portion specifying process for specifying the crown of the occupant from the high likelihood region, and a head model predetermined as a model of the head with the head as a base point. The above-mentioned [11] to [ 13] may be the program described in any one of.
[15] Further, in the control step, in the crown specifying process, the highest portion of the high likelihood region in the coordinate system in the vehicle interior space of the vehicle along the vehicle height direction is specified as the crown. The program described in [14] above may be used.
[16] Further, the control step may be the program according to the above [14] or [15], which detects the position of the center of gravity of the head region as the position of the head in the head specifying process. ..
[17] Further, even if the control step is the program according to any one of the above [11] to [16], the verification process for confirming the certainty of the identified head of the occupant is executed. good.
[18] Further, when the control step detects a group of pixels existing below the specified crown in the coordinate system in the vehicle interior space of the vehicle as the body portion of the occupant, the specified occupant's body portion. The program according to the above [17] may be executed as the verification process to ensure that the head is probable.
[19] Further, in the control step, if the body portion of the occupant is not detected, the head of the identified occupant is uncertain, and the specific process is re-executed as the verification process. The program described in [17] or [18] above may be used.
[20] Further, in the control step, in the fuselage estimation region indicating the region of the fuselage with the head of the occupant specified in the head identification process as the base point, the control step has a specified number of pixels having distance information indicating the distance in the specified range. Is detected, the program according to the above [18] or [19] may be used, in which the body estimation area is detected as the body portion of the occupant.

According to the processing apparatus and the program of the present invention, the calculation cost can be reduced in the technique of detecting the head of the occupant.

FIG. 1 is a coordinate diagram showing three-dimensionally the relationship of the coordinate system in a state where the processing device according to the embodiment of the present invention is mounted on a vehicle. FIG. 2 is a flowchart showing the operation of the processing apparatus according to the embodiment of the present invention. FIG. 3A is a pixel group diagram showing a three-dimensional pixel group in the vehicle coordinate system (w, l, h) in the detection area, and FIG. 3B is a region from the pixel group in the detection area. It is the extraction pixel group diagram which narrowed down. FIG. 4A is a pixel group diagram extracted by Step 2, FIG. 4B is a pixel group diagram when the pixel group is divided in the direction of the l-axis, and FIG. 4C is a high pixel group diagram. It is a pixel group diagram extracted as a likelihood region. FIG. 5 is a chart showing a head model definition example of the head width and the head height. FIG. 6A is a pixel group diagram extracted by Step 3, FIG. 6B is a pixel group diagram when the pixel group is divided in the direction of the w-axis, and FIG. 6C is a high pixel group diagram. It is a pixel group diagram extracted as a likelihood region. FIG. 7 is a diagram showing the crown extracted as the maximum value in the h direction. FIG. 8A is a pixel group diagram in the vehicle coordinate system (w, l, h) showing the extracted head top, and FIG. 8B is wh from which the head is extracted according to the head model definition example. It is a pixel group diagram in a plane, FIG. 8C is a pixel group diagram in a lh plane from which a head is extracted according to a head model definition example, and FIG. 8D is a pixel showing an extracted head region. It is a group map. FIG. 9 is a pixel group diagram showing the extracted head position (center of gravity position of the head region). FIG. 10 is a diagram showing a fuselage estimation region included in the region below the head position. FIG. 11 is a chart showing values of the w-axis, l-axis, and h-axis showing parameter examples of the fuselage estimation region. FIG. 12A is an image example showing an example in which a hand is erroneously detected at the head position, and FIG. 12B is an image example showing an example in which the hand and the head can be distinguished and the erroneous detection is improved. be.

(Embodiment of the present invention)
The processing device 1 according to the embodiment of the present invention is a distance image in which an imaging region including a vehicle occupant is an imaging target, and is an acquisition unit that acquires a distance image in which distance information to an imaging target is assigned to each pixel. The TOF camera 10 and the pixel group constituting the distance image are extracted according to the correspondence between the coordinate system in the vehicle interior space of the vehicle 8 and the coordinate system in the front image, and the head 50 of the occupant 5 exists. It is configured to include a control unit 20 that executes a specific process for specifying the head of the occupant 5 with respect to a high likelihood region, which is a pixel group having a high possibility of being generated.

The pixel group is a set of points corresponding to the pixels imaged by the TOF camera 10, and the position is displayed by the coordinate system in the vehicle interior space of the vehicle 8, the coordinate system in the TOF camera 10, and the like. It is a set of points in space. The high-likelihood region is a region in which a detection target is likely to exist. Specifically, for example, it is a region in which the number of pixel groups in the region is large or the pixel group density is high.

(TOF camera 10)
The acquisition unit can be used as long as it can recognize the image pickup target in three dimensions, but in the present embodiment, the TOF (Time Of Flight) camera 10 is used as the acquisition unit. The TOF camera 10 can detect the time until the light of the light source hits the object to be measured and returns for each pixel, and can take a three-dimensional image including the position information corresponding to the distance in the depth direction. After emitting infrared light or the like, the TOF camera 10 receives the reflected light that is reflected by the object and returns, measures the time from the light emission to the light reception, and measures the time from the light emission to the light reception, and the distance to the object to be imaged for each pixel. Is detected.

As shown in FIG. 1, for example, the TOF camera 10 is attached near the rear-view mirror, and the imaging region including the occupant 5 of the vehicle 8 is targeted for imaging. The image captured by the TOF camera 10 includes coordinates (u, v) and pixel values d (u, v) as depth information at these coordinates (u, v). The acquired pixel value d (u, v) (u = 0, 1, ... U-1, v = 0, 1, ... V-1) is an object to be imaged (the object referred to here is the driver). And in-vehicle devices). The reference numeral U in the embodiment means the horizontal width [pixel] in the captured image, and the reference numeral V means the vertical width [pixel] in the captured image. That is, a distance image is generated by assigning the distance information in the three-dimensional space to each pixel in the image captured by the TOF camera 10.

なお、（ｃ_ｘ、ｃ_ｙ）は、画像中心座標、ｆは、レンズ焦点距離である。 (Camera coordinate system obtained from distance image)
The coordinate values u, v in the range image and the pixel value d (u, v) in the coordinates (u, v) are converted into points (x, y, z) in the three-dimensional space using the following equation. Then, a three-dimensional pixel group in the camera coordinate system (x, y, z) shown in FIG. 1 can be created.

Note that (c _x , _cy ) is the image center coordinates, and f is the lens focal length.

(Vehicle coordinate system)
The camera coordinate system (x, y, z) can be converted into a vehicle coordinate system (w, l, h) whose origin is the driver (for example, the position of the hip bone) to be detected. The conversion method is a combination of general coordinate rotation conversion, translation conversion, and scale conversion. The coordinate axes after conversion are w-axis in the right direction of the driver, l-axis (el-axis) in the anterior direction, h-axis in the upper direction, and the origin is the position of the hipbone of the driver.

(Control unit 20)
The control unit 20 includes, for example, a microcomputer for performing coordinate conversion, head identification processing, head identification processing, and the like. As shown in FIG. 1, the control unit 20 is connected to the TOF camera 10. The control unit 20 includes a CPU (Central Processing Unit) 21, a semiconductor memory RAM (Random Access Memory) 22 and a ROM (Read Only Memory) 23, which perform calculations and processing on the acquired data according to a stored program. I have.

(Operation of processing device 1)
It will be described based on the flowchart which shows the operation of the processing apparatus 1 which concerns on embodiment of this invention shown in FIG. The control unit 20 executes the following operations and processes according to the flowchart.

(Step 1)
The TOF camera 10 as an acquisition unit acquires a distance image in which the image pickup region including the occupant 5 of the vehicle 8 is the image pickup target, and the distance information to the image pickup target is assigned to each pixel. This distance image can be acquired as one frame of the frames captured at predetermined time intervals. The acquired distance image functions as an input (three-dimensional pixel group) in the following processing. The distance image is converted into the camera coordinate system and the vehicle coordinate system described above. As a result, as shown in FIG. 3A, the three-dimensional pixel group 201 included in the detection area can be extracted. In the following processing, operations, processing, and the like are executed in the three-dimensional pixel group of the vehicle coordinate system (w, l, h).

(Step2)
The control unit 20 first extracts the pixel group 202 in the detection area as shown in FIG. 3B as the extraction of the crown portion. The area is narrowed down by defining the detection area and extracting the pixel group included in the space. The detection area is a space above the vehicle interior where the head 50 is considered to be present when seated. For example, the origin is the driver hipbone position 203, and the range is −510 <w <360, 280 <l <700, 430 <h <750. The unit is mm (hereinafter the same).

When the pixel group extracted by Step 1 is viewed from the w-axis direction in order to extract the region including the head 50, the control unit 20 has the highest pixel group density as shown in FIG. 4A. It is considered that the region includes the head 50 and the body portion 70 as a high likelihood region (it is considered that the body portion 70 exists below the head 50). Here, based on the width D of the head model definition (FIG. 5), the region is divided in the direction of the l-axis (FIG. 4 (b)), and the region 204 having the largest number of pixel groups on the lh plane is extracted (FIG. 4). (C)).

(Step 3)
The control unit 20 has the region where the pixel group density is highest when the region 204 (FIG. 6A) of the pixel group extracted by Step 2 is viewed from the l-axis direction in order to extract the region including the crown. Is considered to include the crown as a high likelihood region. Here, it is divided in the direction of the w-axis based on the width D defined by the head model (FIG. 6 (b)), and the region 205 having the largest number of pixel groups on the wh plane is extracted (FIG. 6 (c)). ..

(Step 4)
The control unit 20 executes a parietal region specifying process for calculating the parietal region position. As shown in FIG. 7, among the pixel groups extracted by Step 3, the h-axis, that is, the highest portion in the coordinate system in the vehicle interior space of the vehicle in the vehicle height direction is considered to be the crown 60. The crown 60 may be extracted by averaging some of the large values of the pixel group 205 (FIG. 6 (c)) extracted in the h direction. Let the coordinates of the extracted crown 60 be (wt, lt, ht).

(Step 5)
The control unit 20 extracts the head region. The head 50 is extracted from the head 60 and the head model definition (FIG. 5) extracted by the extraction process of the head. The head region 206 is obtained by extracting a pixel group as shown in FIGS. 8A to 8D with reference to the coordinates (wt, lt, ht) of the crown 60. The head region 206 is in the range of wt-w ₁ <w <wt + w ₁ , lt-l ₁ <l <lt + l ₁ , ht-h ₁ <h <ht. When the head model definition example (FIG. 5) is adapted, w ₁ = 176/2, l ₁ = 176/2, and h ₁ = 146. It should be noted that h ₁ is set to h ₁ = 206-60 = 146 by subtracting 60 from the height H defined in the head model, assuming the lack of information on the hair portion due to the characteristics of the TOF camera.

(Step 6)
The control unit 20 executes a head identification process for calculating the head position. The head position C (wc, lc, hc) of the region 206 extracted in Step 5 is defined as the head 50. That is, as shown in FIG. 9, the head position C is determined by obtaining the center or the center of gravity of the pixel group of the head region 206 included in the region defined by the above ranges w, l, and h. Can be done.

(Step 7)
As shown in FIG. 10, the control unit 20 counts the pixel groups included in the body estimation region 100 below the head 50 (wc, lc, hc). Since the body portion 70 is included below the head 50 in the coordinate system in the vehicle interior space of the vehicle, it is possible to verify whether or not the extracted head 50 is correct by counting the pixel groups. The body estimation area shall be large enough to include the body part. For example, as shown in FIG. 11, the fuselage estimation area can be determined based on the human body size database or the like published by the National Institute of Advanced Industrial Science and Technology.

(Step 8)
The control unit 20 determines whether or not the pixel group density is equal to or higher than the threshold value. When a specified number of pixels having distance information indicating a distance in a specified range are detected in the body estimation area, the body estimation area is detected as the body portion of the occupant. When the pixel group counted in Step 7 is equal to or larger than the threshold value, (wc, lc, hc) is determined as the head (Step 8: Yes). If the pixel group counted in Step 7 is less than the threshold value, the process proceeds to Step 9 (Step 8: No). By the judgment of Step 8, it is possible to prevent the hand 80 as shown in FIG. 12A from being erroneously detected in the head 50, and as shown in FIG. 12B, the hand 80 and the head 50 can be distinguished.

(Step 9)
The control unit 20 excludes the head region extracted by Step 5, returns to Step 1 again, and executes the process again.

The above processing execution can be repeatedly executed as a series of operations shown above.

(Embodiment as a program)
An acquisition step of acquiring a distance image, which is a distance image of an imaging region including a vehicle occupant, which is shown by the processing device 1 and has distance information to the imaging target assigned to each pixel, to a computer. A pixel group constituting the distance image, which is extracted according to the correspondence between the coordinate system in the vehicle interior space of the vehicle and the coordinate system in the distance image, and has a high possibility that the head of the occupant is present. A program for causing a computer to execute a control step for executing a specific process for specifying the head of an occupant with respect to a high likelihood region is also one of the embodiments of the present invention.

Step 1 described in the operation of the processing device 1 is an example of an acquisition step for acquiring a distance image, Step 2 to Step 6 are an example of a control step for executing a specific process, and process control for executing the flowchart shown in FIG. 2 is performed. , Can be an embodiment of a program to be executed by a computer.

A computer-readable recording medium on which the above program is recorded is also one of the embodiments of the present invention.

(Effect of embodiment)
According to the embodiment of the present invention, it has the following effects.
(1) The processing apparatus according to the embodiment of the present invention acquires a distance image in which an image pickup area including a vehicle occupant is an image pickup target, and distance information to the image pickup target is assigned to each pixel. The TOF camera 10 as an acquisition unit and the pixel group constituting the distance image are extracted according to the correspondence between the coordinate system in the vehicle interior space of the vehicle 8 and the coordinate system in the front image, and the head 50 of the occupant 5 It is configured to have a control unit 20 that executes a specific process for specifying the head of the occupant 5 with respect to a high likelihood region which is a pixel group in which there is a high possibility that the occupant 5 exists. Thereby, by utilizing the three-dimensional information, the head position can be detected regardless of the direction in which the face is photographed. Further, the narrowing process using the three-dimensional pixel group is repeated, and the head position can be easily detected with a small amount of processing.
(2) It is verified whether or not the three-dimensional pixel group exists in the area below the detected head position. By considering the presence of the body part under the head as a feature, it is possible to distinguish between the head and other objects, and it is possible to distinguish between the head and other objects by a simple method. Become. This makes it possible to suppress erroneous detection and prevent erroneous detection in head detection using a TOF camera.
(3) From the above, it becomes possible to provide a processing device and a program that reduce the calculation cost in the technique of detecting the head of the occupant.

Although some embodiments of the present invention have been described above, these embodiments are merely examples and do not limit the invention according to the claims. In addition, these novel embodiments can be implemented in various other embodiments, and various omissions, replacements, changes, etc. can be made without departing from the gist of the present invention. Moreover, not all combinations of features described in these embodiments are essential as means for solving the problems of the invention. Further, these embodiments are included in the scope and gist of the invention, and are also included in the scope of the invention described in the claims and the equivalent scope thereof.

1 ... Processing device, 5 ... Crew, 8 ... Vehicle, 10 ... TOF camera, 20 ... Control unit, 50 ... Head, 60 ... Head top, 70 ... Body part, 80 ... Hand, 100 ... Body estimation area, 201, 202, 204, 205, 206 ... Area, 203 ... Driver lumbar position

Claims (12)

An acquisition unit that acquires a distance image in which an imaging region including a vehicle occupant is an imaging target and distance information to the imaging target is assigned to each pixel corresponding to the coordinates of a given two-dimensional coordinate system .
The coordinates of the pixels in the two-dimensional coordinate system of the distance image acquired by the acquisition unit and the distance information of the coordinates are converted into points on the three-dimensional space to create a three-dimensional pixel group, and the three-dimensional The first coordinate system of the pixel group is converted into the second coordinate system with the occupant as the origin, and the head of the occupant is extracted from the three-dimensional pixel group in the converted second coordinate system. A control unit that executes a specific process for specifying the head of the occupant in a high-probability region, which is a group of pixels that are likely to exist, and a control unit.
Have,
As the specific process, the control unit has a first axis in which the right direction of the vehicle is positive and a negative direction in the left direction from the origin in the second coordinate system, and a second axis in which the front direction of the vehicle is positive. When the axis of the above and the third axis whose upward direction of the vehicle is positive are used.
Based on the predetermined width according to the predetermined head model, the extracted high-likelihood region is divided in the direction of the second axis, and the first region having the largest number of pixel groups is extracted. ,
The first region is divided in the direction of the first axis based on the width, and the second region having the largest number of pixel groups is extracted.
Further, as the crown specifying process in the specifying process, the highest portion of the extracted second region along the positive direction of the third axis is specified as the crown of the occupant.
Processing equipment. The control unit specifies the head region of the occupant by adapting the head model to the crown of the occupant specified in the crown identification process as the head identification process in the identification process. The processing device according to claim 1 , wherein the position of the center of gravity of the head region is detected as the position of the head of the occupant . The processing device according to claim 2 , wherein the control unit executes a verification process for confirming the certainty of the identified head of the occupant. When the control unit detects, in the second coordinate system, a pixel group existing below the specified crown of the occupant as the body portion of the occupant, the identified head of the occupant is detected. The processing apparatus according to claim 3 , wherein the verification process is performed to ensure certainty. The control unit executes the specific process again as the verification process, assuming that the head of the identified occupant is uncertain when the body portion of the occupant is not detected. The processing apparatus according to 3 or 4 . The control unit has a specified number of distance information indicating a distance within a specified range in a body estimation area indicating a region of the body portion of the occupant with the head of the occupant specified in the head specifying process as a base point. The processing device according to claim 4 or 5 , wherein when a pixel is detected, the body estimation area is detected as the body portion of the occupant. An acquisition step of acquiring an image pickup area including a vehicle occupant and assigning distance information to the image pickup target for each pixel corresponding to the coordinates of a given two-dimensional coordinate system, and an acquisition step.
The coordinates of the pixels in the two-dimensional coordinate system of the distance image acquired in the acquisition step and the distance information of the coordinates are converted into points on the three-dimensional space to create a three-dimensional pixel group, and the three-dimensional The first coordinate system of the pixel group is converted into the second coordinate system with the occupant as the origin, and the head of the occupant is extracted from the three-dimensional pixel group in the converted second coordinate system. A control step for executing a specific process for specifying the head of the occupant in a high likelihood region, which is a group of pixels that are likely to exist, and a control step.
Have,
In the control step, as the specific process, in the second coordinate system, the first axis having the right direction of the vehicle as positive and the left direction as negative from the origin, and the second axis having the front direction of the vehicle as positive. When the axis of the above and the third axis whose upward direction of the vehicle is positive are used.
Based on the predetermined width according to the predetermined head model, the extracted high-likelihood region is divided in the direction of the second axis, and the first region having the largest number of pixel groups is extracted. ,
The first region is divided in the direction of the first axis based on the width, and the second region having the largest number of pixel groups is extracted.
Further, as the crown specifying process in the specifying process, the computer is made to specify the highest portion of the extracted second region along the positive direction of the third axis as the crown of the occupant. Program. The control step identifies the occupant's head region by adapting the head model to the occupant's crown identified in the crown identification process as a head identification process in the identification process. The program according to claim 7 , wherein the position of the center of gravity of the head region is detected as the position of the head of the occupant . The program according to claim 8 , wherein the control step executes a verification process for confirming the certainty of the identified head of the occupant. When the control step detects, in the second coordinate system, a group of pixels existing below the identified crown of the occupant as the body portion of the occupant, the identified head of the occupant is detected. The program according to claim 9 , which executes the verification process to make it probable. The control step is claimed to execute the identification process again as the verification process, assuming that the head of the identified occupant is uncertain when the body portion of the occupant is not detected. The program according to 9 or 10. The control step is a predetermined number of pixels having the distance information indicating the distance in the specified range in the body estimation region indicating the region of the occupant's torso with the occupant's head specified in the head identification process as a base point. The program according to claim 10 or 11 , wherein when is detected, the fuselage estimation area is detected as the fuselage portion of the occupant.

Images