JP6829988B2 - Momentum estimation device, momentum estimation program, and momentum estimation system - Google Patents

Description

The present invention relates to a momentum estimation device, a momentum estimation program, and a momentum estimation system, and more particularly, to a momentum estimation device, a momentum estimation program, and a momentum estimation system that estimate the effect of exercise on a pedestrian walking in a building. ..

In recent years, the concept of "health management" that realizes improvement of employee vitality and productivity by strategically tackling employee health management from a management perspective has attracted attention.

From the viewpoint of health management, it is preferable to increase the amount of exercise of employees during work, and for this reason, for example, measures may be taken such as devising the flow lines of employees to encourage the use of stairs.

On the other hand, in order to carry out health management efficiently, it is preferable to grasp the amount of exercise of each employee. Therefore, for example, a wearable device that is worn on the body of a working employee to measure basic data on exercise such as calories burned may be used.

Japanese Unexamined Patent Publication No. 2016-144560

However, some employees forget to wear the wearable device, or the power supply of the wearable device is exhausted, which may cause a situation in which basic data on the employee's exercise cannot be obtained.

The present invention has been made in view of the above facts, and is an exercise amount estimation device and an exercise amount that can estimate the effect of exercise on a pedestrian walking in a building without wearing an estimation device for estimating the amount of exercise on the body. It is an object of the present invention to provide an estimation program and a momentum estimation system.

The momentum estimation device of the present invention is acquired by the distance image acquisition unit that detects the detection target space and acquires the distance image obtained, and the distance image acquisition unit in a state where no pedestrian exists in the detection target space. A storage unit that stores a distance image as a background distance image, a pedestrian image acquisition unit that acquires a pedestrian distance image from the difference between the distance image acquired by the distance image acquisition unit and the background distance image, and the pedestrian. on the basis of the pedestrian distance image acquired by the image acquisition unit includes an estimation unit that estimates at least one of the pedestrian exercise intensity and amount of activity is represented by the pedestrian range image, the estimator may , The pedestrian's pace and movement displacement represented by the pedestrian distance image are estimated, and the estimated pace and the estimated movement displacement, and the relationship between the previously obtained pace and the exercise intensity corresponding to the estimated movement displacement are described above. exercise intensity pedestrian represented by pedestrians distance image estimated. As a result, the exercise intensity of the pedestrian can be estimated from the pedestrian distance image.

Further, the momentum estimation device of the present invention is acquired by the distance image acquisition unit that detects the detection target space and acquires the obtained distance image, and the distance image acquisition unit in a state where no pedestrian exists in the detection target space. A storage unit that stores the obtained distance image as a background distance image, a pedestrian image acquisition unit that acquires a pedestrian distance image from the difference between the distance image acquired by the distance image acquisition unit and the background distance image, and the above. Based on the pedestrian distance image acquired by the pedestrian image acquisition unit, the estimation includes an estimation unit that estimates at least one of the exercise intensity and the activity amount of the pedestrian represented by the pedestrian distance image. parts, the pedestrian distance pedestrian pace represented by the image, and estimates the moving displacement and size, estimated cadence, estimated the displacement, of the estimated previously determined cadence and exercise intensity corresponding to the displacement has to estimate the amount of activity of the pedestrian from physique relationship and estimated represented by the pedestrian range image. This makes it possible to estimate the amount of pedestrian activity from the pedestrian distance image.

Further, in the present invention, the relationship between the movement displacement, the pace and the exercise intensity measured in advance for the stairs walking of a plurality of subjects by a method such as breath gas analysis may be used.

Further, in the present invention, the estimation unit acquires the difference in pixel values between the plurality of pedestrian distance images and the distance image acquired by the distance image acquisition unit corresponding to each of the pedestrian distance images. The pedestrian pace included in each of the pedestrian distance images is estimated using the interval.

Further, in the present invention, in the estimation unit, a region corresponding to the lower limb portion of the pedestrian is extracted from the plurality of pedestrian distance images, the difference in pixel values between the extracted regions, and the pedestrian distance image. The pace of the pedestrian included in each of the pedestrian distance images is estimated by using the acquisition interval of the distance image acquired by the distance image acquisition unit corresponding to each.

Therefore, the pedestrian's pace can be estimated without equipping the pedestrian with a device for estimating the pace.

Further, the present invention further includes a display unit that displays at least one of the pedestrian's exercise intensity and activity amount represented by the pedestrian distance image estimated by the estimation unit, and the estimation unit includes a plurality of the walking. A region corresponding to the pedestrian's waist is extracted from the pedestrian distance image, and the movement displacement of the extracted area corresponding to the pedestrian's waist indicates the quality of the walking posture of the pedestrian included in each of the pedestrian distance images. The evaluation value may be calculated and the evaluation value may be displayed on the display unit. By providing the display unit, it is possible to convey the evaluation value of one's own walking posture to a walking pedestrian, so that improvement of the walking posture can be promoted.

On the other hand, in order to achieve the above object, the momentum estimation program of the present invention uses a computer as a distance image acquisition means for detecting a detection target space and acquiring a distance image obtained, and a pedestrian in the detection target space. A pedestrian distance image from the difference between a storage means that stores a distance image acquired by the distance image acquisition means in a non-existent state as a background distance image and a distance image acquired by the distance image acquisition means and the background distance image. Based on the pedestrian image acquisition means for acquiring the image and the pedestrian distance image acquired by the pedestrian image acquisition means, at least one of the exercise intensity and the activity amount of the pedestrian represented by the pedestrian distance image is obtained. It functions as an estimation means for estimation .

Therefore, according to the exercise amount estimation program of the present invention, at least one of the exercise intensity and the activity amount of the pedestrian is estimated using the pedestrian distance image acquired from the distance image, so that at least one of the exercise intensity and the activity amount is estimated. It is possible to estimate the effect of exercise on a pedestrian walking in a building without attaching an estimation device for estimating the above to the pedestrian.

Further, the momentum estimation system of the present invention is acquired by the distance image acquisition unit that detects the detection target space and acquires the obtained distance image, and the distance image acquisition unit in a state where no pedestrian exists in the detection target space. A storage unit that stores the distance image as a background distance image, a pedestrian image acquisition unit that acquires a pedestrian distance image from the difference between the distance image acquired by the distance image acquisition unit and the background distance image, and the walking. A pedestrian distance estimation device having an estimation unit that estimates at least one of the pedestrian's exercise intensity and activity represented by the pedestrian distance image based on the pedestrian distance image acquired by the pedestrian image acquisition unit, and walking. A pedestrian who possesses the transmitting device is identified from the information of the receiving device that receives the radio wave emitted from the transmitting device possessed by the person and the transmitting device represented by the radio wave received by the receiving device, and the identified pedestrian. The identification device associates the identification information of the person with at least one of the exercise intensity and the activity amount of the pedestrian represented by the pedestrian distance image estimated by the exercise amount estimation device. It is provided with a storage device that stores the identified pedestrian identification information and at least one of exercise intensity and activity amount.

As described above, according to the momentum estimation system of the present invention, the identification device associates the identification information emitted from the transmission device possessed by the pedestrian with at least one of the pedestrian's exercise intensity and activity amount and stores the device. Since it is memorized in, the effect of exercise can be managed for each pedestrian.

Further, the momentum estimation system of the present invention is acquired by the distance image acquisition unit that detects the detection target space and acquires the obtained distance image, and the distance image acquisition unit in a state where no pedestrian exists in the detection target space. A storage unit that stores the distance image as a background distance image, a pedestrian image acquisition unit that acquires a pedestrian distance image from the difference between the distance image acquired by the distance image acquisition unit and the background distance image, and the walking. An exercise amount estimation device having an estimation unit that estimates at least one of the exercise intensity and the activity amount of the pedestrian represented by the pedestrian distance image based on the pedestrian distance image acquired by the pedestrian image acquisition unit, and the above. An imaging device that captures the face of a pedestrian walking in the detection target space, and a storage device that stores in advance information in which an image of the pedestrian's face and identification information of the pedestrian corresponding to the face are associated with each other. When the pedestrian face image captured by the imaging device is included in the face image stored in the storage device, it is associated with the pedestrian face image captured by the imaging device. An identification device that associates the identification information with at least one of the pedestrian's exercise intensity and activity amount represented by the pedestrian distance image estimated by the exercise amount estimation device and stores the identification information in the storage device. It has.

As described above, according to the momentum estimation system of the present invention, the identification device compares the pedestrian face image captured by the imaging device with the pedestrian face image stored in the storage device. When the pedestrian face image stored in the storage device includes the pedestrian face image captured by the imaging device, it is associated with the pedestrian face image stored in the storage device in advance. Since the identification information is stored in the storage device in association with at least one of the pedestrian's exercise intensity and activity amount, the pedestrian exercises for each pedestrian without possessing a transmission device for transmitting the identification information. You can manage the effect of.

In addition, the momentum estimation system according to the present invention may store attributes related to the physique of a pedestrian in advance in the storage device. As a result, when estimating the amount of pedestrian activity with the exercise amount estimation device, the attributes related to the pedestrian's physique stored in the storage device can be used. Therefore, the pedestrian's physique estimated by the exercise amount estimation device is used. Compared with the case of estimating the amount of activity, the amount of activity can be estimated more accurately.

According to the present invention, it is possible to estimate the effect of exercise on a pedestrian walking in a building without wearing an estimation device for estimating the amount of exercise on the body.

It is a figure which shows the structural example of the momentum estimation system which concerns on 1st Embodiment. It is a schematic diagram explaining the method of measuring the distance to a subject by a distance image sensor. It is a figure which shows an example of a distance image. It is a figure which shows an example of a pedestrian distance image. It is a figure which shows the configuration example of the momentum estimation device using a computer. It is a flowchart which shows an example of the flow of the momentum estimation process in the momentum estimation device. It is a figure which shows the flow from the distance image to the acquisition of a pedestrian distance image. It is a schematic diagram explaining the method of estimating the height of a pedestrian represented by a pedestrian distance image. It is a schematic diagram explaining the method of estimating the body surface area of a pedestrian represented by a pedestrian distance image. It is a figure which shows an example of the height, weight and body surface area of a pedestrian. It is a graph which shows an example of the relationship between the estimated height of a pedestrian and the actual height, and the relationship between the estimated body surface area of a pedestrian and the calculated value of the body surface area. It is a figure which shows an example of the movement displacement in the width direction in the pedestrian walking on the stairs. It is a figure which shows an example of the movement displacement in the height direction in the pedestrian walking on the stairs. It is a figure which shows an example of the frequency characteristic of the movement displacement in the width direction and the height direction of a pedestrian. It is a figure which shows an example of exercise intensity information. It is a figure which shows the configuration example of the momentum estimation system including the momentum estimation device provided with the display part. It is a figure which shows the structural example of the momentum estimation system which concerns on 2nd Embodiment. It is a figure which shows an example of the estimation information managed by a database. It is a figure which shows the structural example of the momentum estimation system which concerns on 3rd Embodiment.

Hereinafter, examples of embodiments for carrying out the present invention will be described in detail with reference to the drawings.

<First Embodiment>
First, the configuration of the momentum estimation system 1 according to the first embodiment will be described with reference to FIG. As shown in FIG. 1, the momentum estimation system 1 includes a distance image sensor 7 and a momentum estimation device 10, and the momentum estimation device 10 is connected to the network 9.

Some distance image sensors 7 use several distance measuring principles. Here, as an example, m lasers in the horizontal direction and n lasers in the vertical direction are irradiated toward the subject, and the lasers are irradiated. TOF that measures the distance from the distance image sensor 7 to the subject for each (m × n) pixel based on the difference between the time and the time when the laser reflected by the subject is received.

It is assumed that the (Time Of Flight) type distance image sensor 7 is used.

Specifically, as shown in FIG. 2, the distance from the distance image sensor 7 to the specific part 6A of the pedestrian 6 which is an example of the subject is R, the time when the laser is irradiated and the laser reflected by the subject are received. Assuming that the difference from time is Δt and the light speed is c, the distance image sensor 7 measures the distance from the distance image sensor 7 to the subject according to the equation (1) for each laser.

The distance image sensor 7 sequentially acquires an image (distance image) represented by (m × n) pixels whose value is the distance from the distance image sensor 7 to the subject at a predetermined frame rate (for example, 30 fps). To do.

Since each pixel value in the distance image represents the distance to the subject, the distance image 2 as shown in FIG. 3 can be obtained by changing the display form according to the distance to the subject. In the example of FIG. 3, the brightness is lowered (blackened) as the distance from the distance image sensor 7 to the subject increases, and the brightness is increased (whitened) as the distance from the distance image sensor 7 to the subject decreases. , The distance from the distance image sensor 7 to the subject is visualized.

The method of visualizing the distance image is not limited to this, and for example, it is displayed in a reddish color as the distance from the distance image sensor 7 to the subject increases, and becomes bluish as the distance from the distance image sensor 7 to the subject decreases. The color may be changed according to the distance, such as displaying in a different color.

Further, as long as the distance image 2 can be acquired, there is no limitation on the method of acquiring the distance image in the distance image sensor 7. For example, if the distance detection area has a predetermined illuminance or more, the subject can be viewed from different directions. The distance image 2 may be acquired by using a stereo camera that captures and records the depth information of the subject.

Hereinafter, the space for detecting the distance by the distance image sensor 7 is set to the stairs 8 in which the exercise intensity is higher than when walking in another passage and the effect of the pedestrian's exercise is likely to appear. That is, a mode will be described in which the distance image sensor 7 is installed above the pedestrian 6 and on a wall surface or the like where the stairs 8 can be seen, and the distance image 2 of the pedestrian 6 walking on the stairs 8 is acquired. The location of the distance image sensor 7 is an example, and it may be installed on the upper part of a wall surface such as a corridor where a flat passage can be seen.

The momentum estimation device 10 includes a distance image acquisition unit 11, an analysis unit 12, a communication unit 13, and a storage unit 14.

A distance image sensor 7 is connected to the distance image acquisition unit 11, the distance image 2 is acquired from the distance image sensor 7, the acquired distance image 2 is stored in the storage unit 14, and the distance image 2 is stored from the distance image sensor 7. The notification (distance image acquisition notification) obtained is notified to the analysis unit 12.

The analysis unit 12 includes the pedestrian image acquisition unit 15 and the estimation unit 16, and when the distance image acquisition notification is received from the distance image acquisition unit 11, the pedestrian image acquisition unit 15 starts from the distance image 2 stored in the storage unit 14. A region corresponding to the pedestrian 6 is extracted, a pedestrian distance image 5 as shown in FIG. 4 is acquired, and the acquired pedestrian distance image 5 is stored in the storage unit 14. Further, the pedestrian image acquisition unit 15 stores the pedestrian distance image 5 in the storage unit 14, and then notifies the estimation unit 16 of the notification that the pedestrian distance image 5 has been acquired (pedestrian distance image acquisition notification).

The method of acquiring the pedestrian distance image 5 by the pedestrian image acquisition unit 15 will be described in detail later.

The estimation unit 16 that received the pedestrian distance image acquisition notification from the pedestrian image acquisition unit 15 uses the pedestrian distance image 5 stored in the storage unit 14 to represent the pedestrian 6 represented by the pedestrian distance image 5. Estimate pace and physique. Then, the estimation unit 16 estimates the exercise intensity of the pedestrian 6 walking on the stairs 8 from the estimated pace, and estimates the activity amount of the pedestrian 6 from the estimated exercise intensity.

Here, the "pedestrian" refers to the walking tempo of the pedestrian 6 represented by the moving speed of the pedestrian or the number of steps per unit time.

In addition, "exercise intensity" is an index showing the intensity of physical activity based on the state of sitting and resting, and is expressed by a unit called METS. The strength of physical activity indicates the strength of physical activity itself regardless of age, weight, and gender, and the strength of physical activity against various exercises, that is, exercise intensity, is announced by the Ministry of Health, Labor and Welfare of Japan. ing. Specifically, when the intensity of physical activity in a sitting and resting state is 1 MET, exercise such as gymnastics is equivalent to about 3.5 METs.

The product of exercise intensity and exercise time is an index showing the amount of physical activity and is also called "exercise", and the unit is [METs / hour]. That is, the stronger the exercise intensity, the shorter the time required for one exercise.

The "activity amount" is an index showing the effect of the amount of physical activity on the person who performed the physical activity. Even with the same amount of physical activity, the effect of exercise is greater for heavier people, so the product of exercise intensity, exercise time, and the weight of the person who exercised, that is, exercise The amount of activity of the pedestrian 6 is defined using the product of the weight of the person who exercised. The amount of activity of the pedestrian 6 is expressed as, for example, calories burned [kcal].

Hereinafter, "exercise amount" will be used as a generic term for at least one of "exercise intensity" and "activity amount".

After estimating the momentum of the pedestrian 6, the estimation unit 16 notifies the communication unit 13 of the estimated momentum.

The communication unit 13 that receives the estimated momentum of the pedestrian 6 from the estimation unit 16 transmits the received momentum to a server (not shown) connected to the network 9 and manages the momentum on a server (not shown). A server (not shown) provides services such as transmitting exercise information including an estimated amount of exercise to an information terminal (not shown) such as a computer or a smartphone connected to the network 9.

The distance image acquisition unit 11, the storage unit 14, the pedestrian image acquisition unit 15, and the estimation unit 16 are examples of the distance image acquisition means, the storage means, the pedestrian image acquisition means, and the estimation means in the disclosed technology, respectively.

The momentum estimation device 10 described above can be realized by using, for example, the computer 100 shown in FIG. In the computer 100, a CPU (Central Processing Unit) 102, a RAM (Random Access Memory) 104, a ROM (Read Only Memory) 106, a hard disk (HDD) 108, and an input / output interface (I / O) 110 take a bus 112. It has a configuration in which each is connected via.

For example, the CPU 102 reads a program stored in advance in the ROM 106, causes the computer 100 to function as the momentum estimation device 10, and uses data such as a distance image 2 for estimating the momentum of the pedestrian 6 as a work area at the time of executing the program. It is stored in the RAM 104 used.

Further, the computer 100 reads various parameters used in estimating the momentum of the pedestrian 6 stored in advance in the HDD 108, and stores the data to be stored in the HDD 108 even when the power of the computer 100 is turned off.

Then, the input device 20, the output device 22, and the communication device 24 are connected to the I / O 110.

The input device 20 includes, for example, at least one of switches for giving instructions to the momentum estimation device 10 by the user, a data reading device for reading data recorded on an optical disc such as a keyboard and a DVD, and an input device such as a mouse and a touch panel. included.

The output device 22 includes, for example, at least one output device such as a light emitting element such as an LED (Light Emitting Diode) indicating the movable state of the momentum estimation device 10 and a display device and a speaker represented by a liquid crystal display.

The communication device 24 includes a communication protocol for transmitting and receiving data to and from a server and an information terminal (not shown) connected to the network 9.

The input device 20 and the output device 22 do not necessarily have to be connected to the I / O 110, but are connected as needed. Further, when the momentum estimation device 10 is used in a form not connected to the network 9, the communication device 24 is also unnecessary.

Further, the HDD 108 is not always necessary for the computer 100, and the computer 100 may use a storage device connected via, for example, a network 9 instead of the HDD 108.

Next, the operation of the momentum estimation device 10 according to the present embodiment will be described with reference to FIG. Note that FIG. 6 is a flowchart showing the process of the momentum estimation program executed by the CPU 102 when the pedestrian 6 is walking on the stairs 8, that is, the flow of the momentum estimation process, and the program is stored in the ROM 106 in advance. Has been done.

The storage unit 14 contains a distance image (hereinafter referred to as "background distance image 3") acquired in advance in a state where the pedestrian 6 does not exist in the distance detection target space by the distance image sensor 7, and the distance image sensor 7. It is assumed that the distance image 2 acquired last time and a plurality of pedestrian distance images 5 (details of the pedestrian distance image 5 will be described later) acquired while the pedestrian 6 is walking on the stairs 8 are stored. To do. Further, in order to make the explanation easy to understand, a case where only one pedestrian 6 is included in the detection target space will be described.

First, in step S10 of FIG. 6, the distance image 2 is acquired from the distance image sensor 7.

In step S20, the distance image 2 acquired in step S10 is compared with the previously acquired distance image 2 stored in the storage unit 14, and the distance image 2 acquired in step S10 is the distance image 2 acquired last time. Is the same as or not.

When the determination process in step S20 is affirmative, the pedestrian 6 can be considered not to be included in the detection target space of the distance image sensor 7. Therefore, the distance image 2 acquired in step S10 is stored in the storage unit 14. Then, the process proceeds to step S110.

In step S110, it is determined whether or not the user has received the end instruction for terminating the operation of the momentum estimation device 10, which is notified by, for example, operating the input device 20.

If the determination process in step S110 is a negative determination, the process proceeds to step S10, and the process of acquiring the distance image 2 from the distance image sensor 7 is repeated until the end instruction is received. If the determination process in step S110 is affirmative, the process of the momentum estimation program shown in FIG. 6 is terminated.

On the other hand, when the determination process in step S20 is a negative determination, it can be considered that the pedestrian 6 is walking on the stairs 8, so that the distance image 2 acquired in step S10 is stored in the storage unit 14. The process proceeds to step S30, and the amount of exercise of the pedestrian 6 is estimated in the subsequent processing.

The same distance image 2 means, for example, a case where the number of pixels having different pixel values is equal to or less than a preset threshold value among the corresponding pixels of different distance images 2. The threshold value is obtained in advance by an experiment using the actual momentum estimation device 10 or a computer simulation based on the design specifications of the momentum estimation device 10, and is stored in advance in, for example, a predetermined area of the HDD 108.

In step S30, the background distance image 3 is acquired from the storage unit 14.

In step S40, a difference image 4 showing the difference between the distance image 2 acquired in step S10 and the background distance image 3 acquired in step S30 is acquired.

As described above, since the background distance image 3 is a distance image acquired in a state where the pedestrian 6 does not exist in the distance detection target space by the distance image sensor 7, the difference from the distance image 2 acquired in step S10. By taking the above, the difference image 4 shows a distance image corresponding to the pedestrian 6 walking in the detection target space.

However, since the difference image 4 is simply a distance image showing the difference between the distance image 2 acquired in step S10 and the background distance image 3, it includes various noise components in addition to the distance image corresponding to the pedestrian 6. May be Therefore, in some cases, it may be difficult to specify the area corresponding to the pedestrian 6.

Therefore, in step S50, known image processing is applied to the difference image 4 acquired in step S40 to reduce the noise component contained in the difference image 4, and the pedestrian 6 is dealt with more clearly than the difference image 4. The distance image from which the region is extracted, that is, the pedestrian distance image 5, is acquired from the difference image 4.

The image processing applied to the difference image 4 in step S50 includes an opening processing for smoothing the image by using the expansion processing and the contraction processing of the image, a median filter processing for removing noise by using a median filter, and the like. Can be mentioned. However, the image processing applied to the difference image 4 in step S50 is not limited to the opening processing and the median filter processing, and any other image processing can be applied as long as the image processing can remove the noise component from the difference image 4. Needless to say, you can do it.

FIG. 7 shows the flow from the distance image 2 to the acquisition of the pedestrian distance image 5 using the background distance image 3.

In S60, the pedestrian distance image 5 acquired in step S50 is used to estimate the physique of the pedestrian 6 represented by the pedestrian distance image 5, and more specifically, the height and weight of the pedestrian 6.

FIG. 8 is a diagram showing an example of a method for estimating the height of the pedestrian 6 represented by the pedestrian distance image 5.

As shown in FIG. 8, the height H of the pedestrian 6 is the distance r _H from the distance image sensor 7 to the crown of the pedestrian 6, and the distance r _L from the distance image sensor 7 to the feet of the pedestrian 6. if the distance angle between the image sensor 7 and the straight line L _H and the range image sensor 7 connecting the top of the pedestrian 6 and the straight line L _L connecting the feet of the walker 6 theta Togawakare, using the cosine theorem ( It can be estimated from Eq. 2).

Of pedestrian range image 5, the pixel value of the pixel P _H is located uppermost in the height direction of the pedestrian 6 (Z-axis direction in FIG. 8) corresponds to the distance r _H, the height direction of the pedestrian 6 the most pixel values of the pixels P _L located below corresponds to the distance r _L in.

Further, the distance detection range of the distance image sensor 7 in the height direction of the pedestrian 6, that is, the vertical angle of view A _L is predetermined by the distance image sensor 7, and the distance image sensor 7 in the height direction of the pedestrian 6 The number of pixels n of is also predetermined. Therefore, when the number of pixels along the straight line connecting the pixels P _H and the pixels P _L is measured and the number of pixels is n ₁ , the angle θ can be estimated from the equation (3).

That is, since the distance r _H , the distance r _L, and the angle θ can be obtained from the pedestrian distance image 5, the height H of the pedestrian 6 represented by the pedestrian distance image 5 is estimated using the equation (2). can do.

Since the stairs 8 have a step, the pedestrian distance image 5 in a state where the feet of the pedestrian 6 are hidden by the step of the stairs 8 may be obtained. In this case, since the estimated height H of the pedestrian 6 is lower than the actual height of the pedestrian 6, the estimated height H of the pedestrian 6 may be corrected.

For example, a correction value corresponding to the length of the feet hidden by the step of the stairs 8 may be added to the estimated height H of the pedestrian 6. The correction value is obtained by having a plurality of pedestrians 6 (subjects) who know the actual height _HA walk on the stairs 8 with the momentum estimation device 10 installed on the stairs 8 in the detection target space. _A summary statistic of the difference between the height H of each subject estimated by the estimation device 10 and the actual height _HA , for example, an average value or the like may be stored in advance in the HDD 108 as a correction value.

Alternatively, the estimated height of the pedestrian 6 is corrected to match the highest height or the average height among the heights estimated from the plurality of pedestrian distance images 5 that can be regarded as including the same pedestrian 6. You may.

Since the distance image sensor 7 acquires the distance image 2 at a predetermined frame rate, the distance image 2 acquired in step S10 and the distance image 2 previously acquired in step S20 while the pedestrian 6 walks on the stairs 8 Therefore, a plurality of pedestrian distance images 5 can be obtained at predetermined frame rate intervals for the same pedestrian 6. Therefore, when the pedestrian distance image 5 is acquired at a predetermined frame rate for a predetermined period or longer, the pedestrian distance image 5 is acquired within the predetermined period from the time when the latest pedestrian distance image 5 is acquired in step S50. It can be considered that the same pedestrian 6 is included in the pedestrian distance image 5 obtained.

The predetermined period used for selecting the pedestrian distance image 5 which can be considered to include the same pedestrian 6 is determined by an experiment or the like with an actual machine of the momentum estimation device 10, and is stored in advance in, for example, HDD 108. You should keep it. Hereinafter, the pedestrian distance image 5 that can be regarded as including the same pedestrian 6 is referred to as "the previous pedestrian distance image 5 acquired in step S50".

Next, a method of estimating the weight of the pedestrian 6 represented by the pedestrian distance image 5 will be described.

First, the body surface area of the pedestrian 6 is estimated from the number of pixels of the pedestrian 6 represented by the pedestrian distance image 5. Since a certain relationship is observed between the body surface area and the body weight, information indicating the relationship between the body surface area and the body weight (hereinafter referred to as "body surface area body weight information") and the estimated body surface area are used. The weight of the pedestrian 6 represented by the pedestrian distance image 5 is estimated.

FIG. 9 is a diagram showing an example of a method for estimating the body surface area of the pedestrian 6 represented by the pedestrian distance image 5. The distance image sensor 7 irradiates (m × n) lasers, but in FIG. 9, for convenience of explanation, one laser corresponding to one pixel irradiates the pedestrian 6 from the distance image sensor 7. The situation is schematically shown.

Let R be the distance from the distance image sensor 7 to the irradiation point of the pedestrian 6 by the laser, and let ΔS be the irradiation area of the pedestrian 6 irradiated by the laser. In this case, the irradiation area ΔS is calculated by the equation (4).

Here, Δθ _H is the width of the laser along the width direction of the pedestrian 6 (X-axis direction in FIG. 9), that is, the horizontal laser width, and Δθ _V is the height direction of the pedestrian 6 (Z in FIG. 9). The width of the laser along the axial direction), that is, the width of the laser in the vertical direction.

The distance R can be obtained from the pixel value of the pedestrian distance image 5. Further, the detection range of the distance in the horizontal direction of the distance image sensor 7, that is, the horizontal angle of view A _H is predetermined by the distance image sensor 7 like the vertical angle of view A _L, and the distance image in the width direction of the pedestrian 6 The number of pixels m of the sensor 7 is also predetermined. Therefore, the horizontal laser width Δθ _H per laser can be calculated in advance using the equation (5).

Further, since the vertical laser width Δθ _V per laser corresponds to the case where n ₁ = 1 in the above-mentioned equation (3), it can be calculated in advance using the equation (3).

Therefore, for each of the pixels of the pedestrian 6 represented by the pedestrian distance image 5, the irradiation area ΔS of the laser corresponding to each pixel is calculated using the equation (4), and all the calculated irradiation areas ΔS By adding, the body surface area of the pedestrian 6 on the surface facing the distance image sensor 7 can be estimated.

The horizontal laser width Δθ _H and the vertical laser width Δθ _V may be stored in advance in, for example, the HDD 108.

Then, for example, the body weight of the pedestrian 6 represented by the pedestrian distance image 5 is estimated from the estimated body surface area by using the body surface weight information stored in the HDD 108 in advance. The body surface area and body weight information may be specified by a function having body surface area as an input value and body weight as an output value, or may be specified in a tabular form in which body surface area and body weight are associated with each other.

In addition, since the pedestrian 6 tends to wear thicker clothes in the winter period than in the summer period, the estimated body surface area is larger than the actual body surface area in the winter period. Tend. That is, the estimation accuracy of the weight of the pedestrian 6 may vary depending on the season. Therefore, for example, a temperature sensor (not shown) for measuring the temperature in the detection target space of the distance image sensor 7 is connected to the I / O 110 of FIG. 5, and if the temperature measured by the temperature sensor is less than this temperature, it is during the winter period. If the temperature falls below a predetermined temperature that can be determined to be present, a process for correcting the estimated body surface surface may be performed.

In addition, since there is a tendency that a certain correlation is observed between the height of the pedestrian 6 and the body surface area, a process of correcting the estimated body surface area using the estimated height of the pedestrian 6 is performed. May be good. Then, the estimated height and weight are stored in, for example, a predetermined area of the RAM 104.

FIG. 10 shows the Fujimoto formula, which is one of the calculation formulas for estimating the actual height, actual weight, and body surface area of a subject walking on the stairs 8 (Non-patent document: Kaoruki Fujimoto, 4 outsiders, "Japanese body". The calculated values using "Study on body surface area", Journal of the Japanese Society of Health, Japan Society of Health, 1968, 23 (5), p.443-456) are shown, and FIG. 11 shows the values of each subject estimated in step S60. A graph 26 showing the relationship between the height and the actual height and a graph 28 showing the relationship between the body surface area of each subject estimated in step S60 and the calculated value of the body surface area are shown.

The body surface area of each subject in FIG. 10 is a calculated body surface area calculated from the actual height and actual weight of the subject using the above-mentioned Fujimoto formula. For example, it has been shown that the theoretical body surface area of subject A, who is 172 [cm] tall and weighs 64 [kg], is 1.709 [m ² ].

When the heights of the subjects A to E were estimated by the process of step S60, the estimated height of the subject C was about 1499 [mm], the estimated height of the subject B was about 1546 [mm], and the estimated height of the subject A was about 1546 [mm]. The estimated height of the subject was about 1627 [mm], the estimated height of the subject D was about 1645 [mm], and the estimated height of the subject E was about 1656 [mm].

Graph 26 can be obtained from the estimated height by, for example, the minimum square method, and the graph 26 shows that the estimated height also increases with a certain inclination as the actual height increases, which is the pedestrian distance in step S60. It shows that the height of the pedestrian 6 represented by the image 5 can be estimated.

The reason why the height estimated to be lower than the actual height is that the subject's feet are hidden by the steps of the stairs 8. Therefore, if the height estimated by the above method is corrected, the estimated height will be closer to the actual height.

Further, when the body surface areas of the subjects A to E were estimated by the process of step S60, the estimated body surface area of the subject C was about 302101 [mm ² ], and the estimated body surface area of the subject B was about 338665 [mm. ² ], the estimated body surface area of subject A is about 348292 [mm ² ], the estimated body surface area of subject E is about 357332 [mm ² ], and the estimated body surface area of subject D is about 386345 [mm ² ]. Met.

A graph 28 can be obtained from the estimated body surface area by, for example, the minimum square method, and the graph 28 shows that the estimated body surface area also increases with a constant slope as the calculated value of the body surface area increases, which is in step S60. It shows that the body surface area of the pedestrian 6 represented by the pedestrian distance image 5, that is, the body weight can be estimated.

The estimated body surface area represents the body surface area of the pedestrian 6 on the surface facing the distance image sensor 7, and the calculated body surface area indicates the area of the entire body of the subject. The surface is smaller than the calculated value on the body surface.

In step S70, the pace of the pedestrian 6 represented by the pedestrian distance image 5 acquired in step S50 is estimated.

Specifically, first, the difference between the pixel values of the pixels corresponding to a specific position is calculated between the pedestrian distance image 5 acquired in step S50 and the previous pedestrian distance image 5 acquired in step S50. To do.

Here, the pixel corresponding to a specific position is preferably a pixel located at the center of gravity of the region corresponding to the pedestrian 6 in the pedestrian distance image 5, for example. The center of gravity of the region corresponding to the pedestrian 6 is easier to identify than the region corresponding to the arm, head or leg of the pedestrian 6, and the position of the pedestrian 6 can be represented more accurately. A known calculation method can be used to calculate the center of gravity of the region corresponding to the pedestrian 6.

Next, since the frame rate of the distance image sensor 7 is predetermined, the acquisition interval of each pedestrian distance image 5 for which the pixel values are compared corresponds to each of the pedestrian distance images 5 for which the pixel values are compared. It is calculated from the frame rate of the distance image 2.

The difference between the pixel values corresponding to a specific position in each pedestrian distance image 5 for which the pixel values are compared, that is, the moving distance of the pedestrian 6 and each pedestrian distance image 5 for which the pixel values are compared. Since the time interval is known, the moving speed of the pedestrian 6 represented by the pedestrian distance image 5 can be estimated.

Since it is known that there is a certain relationship between the moving speed of the pedestrian 6 and the number of steps per unit time, for example, information indicating the relationship between the moving speed of the pedestrian 6 and the number of steps per unit time is provided. If stored in the HDD 108 in advance, it is possible to estimate the number of steps per unit time of the pedestrian 6 from the moving speed of the pedestrian 6 estimated by referring to the information.

Further, the lower limbs of the pedestrian 6 are extracted from the pedestrian distance image 5 by using a known pattern matching method for the pedestrian distance image 5 acquired in step S50, and the pixel values of the extracted lower limbs and the step. By analyzing the difference from the pixel value of the lower limbs of the pedestrian 6 extracted from the previous pedestrian distance image 5 acquired in S50 in time series, the pedestrian 6 in the period corresponding to the frame rate of the distance image sensor 7. You can get the change of the movement distance of.

The change in the movement distance of the lower limbs during the period when the pedestrian 6 lands both feet on the stairs 8 is the movement distance of the lower limbs when either the left or right foot is separated from the stairs 8 in order to advance to the next step. There is a tendency for it to be less than the change in. That is, each time the pedestrian 6 advances one step at a time on the stairs 8, the change in the moving distance of the pedestrian 6 is expressed as a periodic pattern.

Therefore, the number of steps per unit time of the pedestrian 6 represented by the pedestrian distance image 5 can be estimated from the changes in the frame rate of the distance image sensor 7 and the moving distance of the pedestrian 6.

If the change in the moving distance of the pedestrian 6 during the period corresponding to the frame rate of the distance image sensor 7 is known, the moving distance of the pedestrian 6 per unit time can be obtained, so that the moving speed of the pedestrian 6 can be obtained. Can be done. Therefore, as described above, for example, even if the information indicating the relationship between the moving speed of the pedestrian 6 and the number of steps per unit time stored in advance in the HDD 108 is referred to, the pedestrian 6 represented by the pedestrian distance image 5 The number of steps per unit time can be estimated.

The estimated movement speed and the number of steps per unit time are stored in, for example, a predetermined area of the RAM 104.

In step S80, in order to estimate the inclination of the stairs 8 on which the pedestrian 6 represented by the pedestrian distance image 5 is walking, the center of gravity of the pedestrian 6 walking on the stairs 8 in the width direction and the height direction. The time-series change of the movement speed of, that is, the movement displacement is estimated from the pedestrian distance image 5.

Therefore, a known method of calculating the center of gravity is applied to the pedestrian distance image 5 acquired in step S50, and the center of gravity of the region corresponding to the pedestrian 6 is specified. Then, using the position of the specified center of gravity and the position of the center of gravity in the region corresponding to the pedestrian 6 specified from the previous pedestrian distance image 5 acquired in step S50, the adjacent pedestrian distance images along the time series. The difference in the position of the center of gravity between the five is calculated, and the moving displacement of the pedestrian 6 represented by the pedestrian distance image 5 is estimated.

The estimated movement displacement is represented by each component in the X-axis, Y-axis, and Z-axis directions of the three-dimensional coordinate system set in the detection target space with reference to the position of the image distance sensor 7 as shown in FIG. The inclination angle of the stairs 8 can be estimated based on the moving displacement in the direction.

Then, the estimated inclination angle of the stairs 8 is stored in, for example, a predetermined area of the RAM 104.

12 and 13 show an example of the movement displacement actually estimated by having four subjects A to E shown in FIG. 10 walk on the stairs 8. Of these, FIG. 12 shows the movement displacement V _X in the width direction of the subject (X-axis direction in FIG. 10), and FIG. 13 shows the height direction of the subject (Z-axis direction in FIG. 10). The moving displacement V _Z is shown.

As shown in FIG. 12, the movement displacement of the pedestrian 6 tends to change periodically in the positive direction and the negative direction. Here, the forward movement displacement in FIG. 12 indicates the movement displacement when the center of gravity of the pedestrian 6 moves in either the left or right direction from the predetermined origin in the detection target space of the distance image sensor 7. The negative movement displacement in FIG. 12 indicates the movement displacement when the center of gravity of the pedestrian 6 moves in the direction opposite to the direction corresponding to the positive movement displacement in FIG.

As a result of the experiment using the subject, it was found from the moving displacement shown in FIG. 12 that the adjacent maximum value or the adjacent minimum value corresponds to two steps. It is considered that this is because the pedestrian 6 tends to move the body alternately in either the left or right direction each time the pedestrian 6 takes one step, and the direction of the movement displacement becomes the same when the pedestrian 6 takes two steps.

Further, as shown in FIG. 13, regarding the movement displacement of the pedestrian in the height direction, the movement displacement of the pedestrian 6 tends to change periodically in the positive direction and the negative direction. Here, the forward movement displacement in FIG. 13 indicates the movement displacement when the center of gravity of the pedestrian 6 moves in any direction above or below a predetermined origin in the detection target space of the distance image sensor 7. The negative movement displacement in FIG. 13 indicates the movement displacement when the center of gravity of the pedestrian 6 moves in the direction opposite to the direction corresponding to the positive movement displacement in FIG.

As a result of the experiment using the subject, it was found from the moving displacement shown in FIG. 13 that the adjacent maximum value or the adjacent minimum value corresponds to one step. It is considered that this is because the pedestrian 6 tends to sink the body once every step.

Therefore, the number of steps per unit time of the pedestrian 6, that is, the pace can be estimated from the movement displacement of the pedestrian 6 estimated in step S80.

Further, as shown in FIG. 14, the moving speed V _X and the moving speed V _Z of the pedestrian 6 obtained by the experiment are subjected to a fast Fourier transform (FFT) to obtain a peak at the frequency corresponding to the pace. Since the frequency characteristics possessed can be obtained, the pace can be estimated.

In the example of FIG. 14, in the graph 32 showing the frequency characteristics of the moving speed V _Z , the frequency at which the value of the frequency component becomes the largest (2.8 Hz in the example of FIG. 14) represents the frequency corresponding to the pace of the pedestrian 6. In the graph 34 showing the frequency characteristics of the moving speed V _X, the frequency at which the value of the frequency component is the largest (1.4 Hz in the example of FIG. 14) represents the frequency corresponding to 1/2 of the pace of the pedestrian 6.

When the inclination angle of the stairs 8 is known in advance and the pace estimated in step S70 is used, the process of step S80 does not necessarily have to be executed.

In step S90, the movement of the pedestrian 6 represented by the pedestrian distance image 5 acquired in step S50 is used by using the pace estimated in step S70 and the inclination of the stairs 8 obtained from the movement displacement estimated in step S80. Estimate the strength.

Specifically, information indicating the relationship between the pace and exercise intensity created for each inclination of the stairs 8 (hereinafter referred to as “exercise intensity information 30”) is stored in advance in, for example, HDD 108, and the inclination of the stairs 8 is obtained. The exercise intensity corresponding to the pace estimated in step S70 is estimated with reference to the exercise intensity information 30 corresponding to the above, and the estimated exercise intensity is stored in a predetermined area of the RAM 104.

The reason why the exercise intensity information 30 exists for each inclination of the stairs 8 is that the exercise intensity differs depending on the degree of inclination of the stairs 8 even if the pace is the same.

If the inclination of the stairs 8 is known in advance, the inclination of the stairs 8 that is known in advance may be used instead of the inclination of the stairs 8 obtained from the movement displacement estimated in step S80.

FIG. 15 is a diagram showing an example of exercise intensity information 30 corresponding to a specific inclination of the stairs 8. Since there is a correlation between exercise intensity and maximal oxygen uptake, the pace and maximal oxygen uptake of the subject can be measured, and exercise intensity can be defined from the measured maximal oxygen uptake.

Although FIG. 15 shows an example in which the pace and exercise intensity are proportional, the pace and exercise intensity may not always be proportional. The exercise intensity information 30 may be defined by a function having the slope and pace of the stairs 8 as an input value and the exercise intensity as an output value, and may be in a tabular form in which the pace and exercise intensity are associated with each slope of the stairs 8. May be specified in.

In step S100, the amount of activity of the pedestrian 6 represented by the pedestrian distance image 5 is used by using the weight estimated in step S60, the exercise intensity of the pedestrian 6 estimated in step S90, and the exercise time of the pedestrian 6. To estimate. Here, if the weight of the pedestrian 6 is w [kg], the exercise intensity u [METs], and the exercise time is _tu [hours], the activity amount Q [kcal] of the pedestrian 6 is calculated by using equation (6). It is known to estimate.

Therefore, it can be considered that the pedestrian 6 represented by the pedestrian distance image 5 acquired in step S50 has performed the exercise of the exercise intensity estimated in step S90 for the time corresponding to the interval of the frame rate. If the rate is 30 fps, the activity amount of the pedestrian 6 at one frame interval of the pedestrian distance image 5 can be estimated by setting the exercise time _tu to 1 / (30 × 3600) [time].

By integrating the estimated activity amount of the pedestrian 6 for each frame interval, the activity amount during the period when the pedestrian 6 is walking on the stairs 8 can be estimated.

The amount of activity of the pedestrian 6 represented by the estimated pedestrian distance image 5 is stored in, for example, a predetermined area of the RAM 104.

After estimating the activity amount of the pedestrian 6, the process proceeds to step S10, and the momentum estimation device 10 executes the momentum estimation process shown in FIG. 6 until the end instruction is received.

As described above, according to the exercise amount estimation device 10 according to the first embodiment, the pedestrian distance image 5 corresponding to the pedestrian 6 is extracted from the distance image 2 acquired by the distance image sensor 7, and the extracted pedestrian distance image 5 From the pace and physique of the pedestrian 6 estimated using the above, at least one of the exercise intensity and the amount of activity of the pedestrian 6 is estimated.

Therefore, it is possible to estimate the effect of exercise on the pedestrian 6 walking in the distance detection target space by the distance image sensor 7 without attaching the device for measuring the calorie consumption or the like to the pedestrian 6.

Since the momentum of the pedestrian 6 going down the stairs 8 is smaller than the momentum of the pedestrian 6 going up the stairs 8, the momentum estimation device 10 walks up the stairs 8. It is preferable to estimate the effect of exercise on the person 6. Whether the pedestrian 6 is climbing or descending the stairs can be determined, for example, by whether the distance to the center of gravity of the pedestrian 6 represented by the pedestrian distance image 5 becomes longer or shorter with the passage of time. ..

<Modified example of the first embodiment>
FIG. 16 is a diagram showing a configuration example of a momentum estimation system 1A using a momentum estimation device 10A in which a display unit 17 is added to the momentum estimation device 10. That is, the momentum estimation device 10A has the same configuration as the momentum estimation device 10 except that the display unit 17 is added to the momentum estimation device 10.

The display unit 17 includes a display device such as a liquid crystal display, an organic EL (Electro-Luminescence), or electronic paper, and receives estimated information such as the estimated pace, physique, exercise intensity, and activity amount from the estimation unit 16. The estimated information is displayed on the display device.

Since the display device is installed in a place visible to the pedestrian 6 walking on the stairs 8, it is possible to provide the pedestrian 6 walking on the stairs 8 with information such as the number of calories burned by walking on the stairs 8. it can. Therefore, it is possible to visually convey the effect of exercise to the pedestrian 6, and it is expected that the pedestrian 6 is encouraged to use the stairs 8 and the pedestrian 6 is promoted to raise awareness of health.

In addition, among the pedestrians 6, there may be a pedestrian 6 who pays attention to the walking posture due to an increase in awareness of health.

As a technique for quantitatively evaluating the quality of walking posture, for example, as described in Japanese Patent No. 5504810, a method of measuring the time change of acceleration of the waist portion of pedestrian 6 and calculating an index showing walking posture is known. Has been done. Therefore, it is considered that the good and bad of the walking posture correlates with the movement of the waist.

Therefore, a method such as determining the quality of the walking posture by measuring the amount of movement of the lumbar region of the pedestrian 6 in time series using a known method and comparing it with the amount of movement of the lumbar region in ideal walking is possible. Conceivable.

Therefore, the estimation unit 16 extracts the waist portion of the pedestrian 6 from the pedestrian distance image 5 by using a known pattern matching method for the pedestrian distance image 5 acquired in step S50 of FIG. 6, and the extracted waist portion. By analyzing in chronological order the difference between the position of and the position of the waist of the pedestrian 6 extracted from the previous pedestrian distance image 5 acquired in step S50, the change in the amount of movement of the waist of the pedestrian 6 is acquired. To do. Then, the estimation unit 16 compares the change in the acquired movement amount of the lumbar region of the pedestrian 6 with the change in the movement amount of the lumbar region in the ideal walking stored in the HDD 108 or the like in advance, so that the walking posture is good or bad. Is evaluated, and the evaluation result is notified to the display unit 17 as an evaluation value (hereinafter referred to as “score”).

The score is represented by, for example, a maximum of 100 points, and the closer the score is to 100 points, the better the walking posture. The display unit 17, which has received the score from the estimation unit 16, displays the score on the display device and notifies the pedestrian 6 walking on the stairs 8.

As described above, the momentum estimation device 10A includes a display unit 17 that notifies the pedestrian 6 who is walking of the estimated momentum. Therefore, the pedestrian 6 can visually confirm the effect of the exercise, and the effect of encouraging the pedestrian 6 to use the stairs 8 can be expected.

<Second Embodiment>
In the second embodiment, the pedestrian 6 walking in the distance detection target space by the distance image sensor 7 is identified, and the estimated information such as the pace, physique, exercise intensity, and activity amount estimated by the exercise amount estimation device 10A is obtained. The momentum estimation system 1B managed for each pedestrian 6 will be described.

FIG. 17 is a diagram showing a configuration example of the momentum estimation system 1B according to the second embodiment. As shown in FIG. 17, the momentum estimation system 1B includes a distance image sensor 7, a momentum estimation device 10A, an antenna 40, an identification device 50, a database 60, and an electronic tag 70.

The electronic tag 70 is an example of a transmission device that transmits information for identifying who the pedestrian 6 is (hereinafter, referred to as “identification information”) to the outside using radio waves, for example, and is added to, for example, an employee ID card. It is possessed by each pedestrian 6. The identification information includes information that identifies an individual, such as gender, date of birth, department to which the person belongs, and identification ID. The attachment destination of the electronic tag 70 is not limited to the employee ID card, and may be attached to other articles such as a wristband.

The antenna 40 is installed within the reach of radio waves emitted from the electronic tag 70 possessed by the pedestrian 6 walking in the detection target space, receives the radio waves emitted from the electronic tag 70, and is notified via the radio waves. The identification information is converted into an electric signal and output to the identification device 50.

The identification device 50 is connected to the network 9, the antenna 40, and the database 60, and estimates information such as the pace, physique, exercise intensity, and activity of the pedestrian 6 notified from the exercise amount estimation device 10A via the network 9, and the antenna. It is stored in the database 60 in association with the identification information of the pedestrian 6 received from 40.

The antenna 40 is an example of a receiving device in the disclosed technology, the database 60 is an example of a storage device in the disclosed technology, and the electronic tag 70 is an example of a transmitting device in the disclosed technology.

FIG. 18 is a diagram showing an example of estimated information for each pedestrian 6 managed by the database 60. In the example of FIG. 18, the identification ID of the pedestrian 6 (also referred to as “pedestrian ID”), the time when the pedestrian 6 is detected by the distance image sensor 7, and the pedestrian 6 estimated by the exercise amount estimation device 10A. The height and weight are associated with the activity amount of the pedestrian 6 estimated by the exercise amount estimation device 10A and managed in the database 60. It goes without saying that the content of the estimated information for each pedestrian 6 managed in the database 60 is not limited to the example shown in FIG. 18, and for example, a score indicating the good or bad of the walking posture may be included. ..

In this way, in the momentum estimation system 1B using the electronic tag 70, the pedestrian 6 carrying the electronic tag 70 is identified only by walking on the stairs 8, and the estimated information is managed for each pedestrian 6. can do.

Therefore, since information such as the number of times the pedestrian 6 has used the stairs 8 and the change in the weight of the pedestrian 6 can be obtained, it is compared with the case where the pedestrian 6 whose momentum or the like is estimated by the momentum estimation device 10A is unknown. Therefore, each pedestrian 6 can be given specific guidance regarding exercise based on the information recorded in the database 60.

If attributes related to physique such as height and weight declared by each pedestrian 6 are registered in the database 60 in advance, the exercise amount estimation device 10A obtains the height and weight of the pedestrian 6 represented by the identification information from the database 60. Can be obtained. Since the attributes related to the physique such as height and weight are registered in response to the declaration of the pedestrian 6, they are often more accurate values than the pedestrian 6's physique estimated by the momentum estimation device 10A. Therefore, the activity amount of the pedestrian 6 can be estimated more accurately than the case where the activity amount is estimated using the physique of the pedestrian 6 estimated by the momentum estimation device 10A.

Further, it goes without saying that in the momentum estimation system 1B, the momentum estimation device 10 that does not include the display unit 17 may be used instead of the momentum estimation device 10A.

<Third Embodiment>
In the exercise amount estimation system 1B according to the second embodiment, the pedestrian 6 is identified by having the pedestrian 6 carry the electronic tag 70, but in the third embodiment, the pedestrian 6 is like the electronic tag 70. The exercise amount estimation system 1C for identifying the pedestrian 6 will be described without possessing a device containing the identification information of the pedestrian 6.

FIG. 19 is a diagram showing a configuration example of the momentum estimation system 1C according to the third embodiment. The difference between the momentum estimation system 1C and the momentum estimation system 1B according to the second embodiment is that the antenna 40 is replaced by the image pickup device 80 and the identification device 50 is replaced by the identification device 50A. It is the same as the momentum estimation system 1B.

The image pickup device 80 is attached at a position where the face of a pedestrian 6 walking in the detection target space can be imaged. Then, the image pickup device 80 images the detection target space at predetermined intervals, and notifies the identification device 50A of the captured image.

On the other hand, in the database 60, the face image of each pedestrian 6 captured in advance and the identification information are associated with each pedestrian 6 and stored in advance.

The identification device 50A is connected to the network 9, the image pickup device 80, and the database 60, and when the image is received from the image pickup device 80, the received image includes some pedestrian 6's face using a known pattern matching method. If the face of the pedestrian 6 is included, the face image of the pedestrian 6 is compared with the face image of each pedestrian 6 stored in advance in the database 60, and the face image having the most matching features is selected. To do.

Then, the identification device 50A acquires the identification information of the pedestrian 6 associated with the selected face image from the database 60, and the pace of the pedestrian 6 notified from the exercise amount estimation device 10A via the network 9. Estimated information such as physique, exercise intensity, and activity amount is associated with the acquired identification information and stored in the database 60.

In this way, in the momentum estimation system 1C using the image pickup device 80, the pedestrian 6 is identified only by walking on the stairs 8 without possessing the device including the identification information, and the estimated information is used as the pedestrian 6 It can be managed for each.

Needless to say, in the momentum estimation system 1C, the momentum estimation device 10 that does not include the display unit 17 may be used instead of the momentum estimation device 10A.

Although the present invention has been described above using each embodiment, the present invention is not limited to the scope described in each embodiment. Various changes or improvements can be made to each embodiment without departing from the gist of the present invention, and the modified or improved forms are also included in the technical scope of the present invention. For example, the order of processing may be changed without departing from the gist of the present invention.

In each of the above-described embodiments, as an example, a mode in which the momentum estimation process in the momentum estimation devices 10 and 10A is realized by software has been described, but the process equivalent to the flowchart shown in FIG. 6 is processed by hardware. You may. In this case, the speed of the processing can be increased as compared with the case where the momentum estimation processing is realized by software.

Further, in each of the above-described embodiments, the mode in which the momentum estimation program is installed in the ROM 106 has been described, but the present invention is not limited to this. The momentum estimation program according to the present invention can also be provided in a form recorded on a computer-readable recording medium. For example, the momentum estimation program according to the present invention may be provided in a form recorded on an optical disk such as a CD (Compact Disc) -ROM or a DVD-ROM, or recorded on a semiconductor memory such as a USB memory and a flash memory. May be provided in the form provided. Further, it may be downloaded from another device connected to the network 9 via the communication device 24.

1 (1A, 1B, 1C) ... Exercise amount estimation system, 2 ... Distance image, 3 ... Background distance image, 4 ... Difference image, 5 ... Pedestrian distance image, 6 ... Pedestrian, 7 ... distance image sensor, 8 ... stairs, 9 ... network, 10 (10A) ... momentum estimation device, 11 ... distance image acquisition unit, 12 ... analysis unit, 13 ... communication unit, 14 ... storage unit, 15 ... pedestrian image acquisition unit, 16 ... estimation unit, 17 ... display unit, 30 ... exercise intensity information, 40 ... Antenna, 50 (50A) ... identification device, 60 ... database, 70 ... electronic tag, 80 ... imaging device, 100 ... computer, 102 ... CPU, 104 ... RAM, 106 ... ROM

Claims (10)

A distance image acquisition unit that detects the detection target space and acquires the distance image obtained,
A storage unit that stores a distance image acquired by the distance image acquisition unit as a background distance image in a state where no pedestrian is present in the detection target space.
A pedestrian image acquisition unit that acquires a pedestrian distance image from the difference between the distance image acquired by the distance image acquisition unit and the background distance image, and
Based on the acquired pedestrian distance image by the pedestrian image acquisition unit, it viewed including the an estimation unit that estimates at least one of the pedestrian exercise intensity and amount of activity is represented by the pedestrian distance image,
The estimation unit estimates the pedestrian's pace and movement displacement represented by the pedestrian distance image, and includes the estimated pace and estimated movement displacement, and the pace and exercise intensity obtained in advance corresponding to the estimated movement displacement. An exercise amount estimation device that estimates the exercise intensity of a pedestrian represented by the pedestrian distance image from the above relationship . A distance image acquisition unit that detects the detection target space and acquires the distance image obtained,
A storage unit that stores a distance image acquired by the distance image acquisition unit as a background distance image in a state where no pedestrian is present in the detection target space.
A pedestrian image acquisition unit that acquires a pedestrian distance image from the difference between the distance image acquired by the distance image acquisition unit and the background distance image, and
Based on the pedestrian distance image acquired by the pedestrian image acquisition unit, an estimation unit that estimates at least one of the exercise intensity and the amount of activity of the pedestrian represented by the pedestrian distance image is included.
The estimation unit estimates the pedestrian's pace, movement displacement, and physique represented by the pedestrian distance image, and the estimated pace, estimated movement displacement, and pre-determined pace and exercise intensity corresponding to the estimated movement displacement. An exercise amount estimation device that estimates the amount of activity of a pedestrian represented by the pedestrian distance image from the relationship with and the estimated physique . The exercise amount estimation device according to claim 1 or 2, wherein the relationship between the pace and exercise intensity is determined in advance using the pace measured for a plurality of subjects and the maximum oxygen uptake. The estimation unit uses the difference in pixel values between the plurality of pedestrian distance images and the acquisition interval of the distance image acquired by the distance image acquisition unit corresponding to each of the pedestrian distance images. motion estimation apparatus according to any one of claims 1 to 3 for estimating the pace of the pedestrian included in each of the pedestrian distance image. The estimation unit extracts a region corresponding to the lower limb of the pedestrian from the plurality of pedestrian distance images, the difference in pixel values between the extracted regions, and the distance corresponding to each of the pedestrian distance images. by using the acquisition interval of the acquired range image by the image acquisition unit, momentum according to any one of claims 1 to 3 for estimating the pace of the pedestrian included in each of the pedestrian range image Estimator. Further, a display unit for displaying at least one of the pedestrian's exercise intensity and the amount of activity represented by the pedestrian distance image estimated by the estimation unit is provided.
The estimation unit extracts a region corresponding to the pedestrian's waist from the plurality of pedestrian distance images, and walks included in each of the pedestrian distance images from the movement displacement of the extracted region corresponding to the pedestrian's waist. The exercise amount estimation device according to any one of claims 1 to 5 , which calculates an evaluation value indicating the quality of a person's walking posture and displays the evaluation value on the display unit. Computer,
A distance image acquisition means for detecting a detection target space and acquiring a distance image obtained,
A storage means for storing a distance image acquired by the distance image acquisition means as a background distance image in a state where no pedestrian is present in the detection target space.
A pedestrian image acquisition means that acquires a pedestrian distance image from the difference between the distance image acquired by the distance image acquisition means and the background distance image, and
An estimation means for estimating at least one of the exercise intensity and the amount of activity of the pedestrian represented by the pedestrian distance image based on the pedestrian distance image acquired by the pedestrian image acquisition means.
To luck momentum estimating program to function. A distance image acquisition unit that detects a detection target space and acquires a distance image obtained, and stores a distance image acquired by the distance image acquisition unit in a state where no pedestrian exists in the detection target space as a background distance image. A storage unit, a pedestrian image acquisition unit that acquires a pedestrian distance image from the difference between the distance image acquired by the distance image acquisition unit and the background distance image, and the pedestrian image acquisition unit acquired by the pedestrian image acquisition unit. An exercise amount estimation device having an estimation unit that estimates at least one of the exercise intensity and the activity amount of the pedestrian represented by the pedestrian distance image based on the person distance image.
A receiving device that receives radio waves emitted from a transmitting device owned by a pedestrian,
A pedestrian possessing the transmitter is identified from the information of the transmitter represented by the radio wave received by the receiver, and the identified pedestrian identification information and the pedestrian estimated by the momentum estimation device are used. An identification device that associates at least one of the pedestrian's exercise intensity and activity amount represented by the distance image,
A storage device that stores the identification information of the identified pedestrian associated with the identification device and at least one of the exercise intensity and the amount of activity.
Momentum estimation system equipped with. A distance image acquisition unit that detects a detection target space and acquires a distance image obtained, and stores a distance image acquired by the distance image acquisition unit in a state where no pedestrian exists in the detection target space as a background distance image. A storage unit, a pedestrian image acquisition unit that acquires a pedestrian distance image from the difference between the distance image acquired by the distance image acquisition unit and the background distance image, and the pedestrian image acquisition unit acquired by the pedestrian image acquisition unit. An exercise amount estimation device having an estimation unit that estimates at least one of the exercise intensity and the activity amount of the pedestrian represented by the pedestrian distance image based on the person distance image.
An imaging device that captures the face of a pedestrian walking in the detection target space,
A storage device that stores in advance information in which an image of a pedestrian's face and pedestrian's identification information corresponding to the face are associated with each other.
When the image of the pedestrian's face captured by the imaging device is included in the image of the face stored in the storage device, it is associated with the image of the pedestrian's face captured by the imaging device. An identification device that associates the identification information with at least one of the pedestrian's exercise intensity and activity amount represented by the pedestrian distance image estimated by the exercise amount estimation device and stores the identification information in the storage device.
Momentum estimation system equipped with. The momentum estimation system according to claim 8 or 9 , wherein attributes related to the pedestrian's physique are stored in advance in the storage device.