JP6909986B2 - Identification device - Google Patents

Description

The present invention relates to an identification device.

In recent years, a technique for identifying a person's posture has been known.

For example, Patent Document 1 describes a device that detects the positions of the feature points of a person's hand, elbow, and shoulder from a camera image and corrects the positions of the feature points of the hand, elbow, and shoulder so as to match a joint model created in advance. Is disclosed.

Japanese Unexamined Patent Publication No. 2010-211705

However, in the device described in Patent Document 1, when a person wears clothes having a low reflectance to near-red light such as a leather jacket, the pixel values of the body and arms cannot be normally obtained, and the shoulders and elbows cannot be obtained normally. Since it is not possible to detect the joints of the person, there is a problem that the posture of the person cannot be identified.

An object of the present invention is to provide an identification device capable of identifying the posture of a person.

The identification device according to one aspect of the present invention has an input unit that receives image information of a person photographed by a camera, identifies the person based on the image information, and detects at least the head and hand parts of the person. The motion model includes the person identified by performing, the detected portion, and a control unit that identifies the motion of the person based on the motion model in which the motion of the person is registered for each person. only including a first behavior model is generated based on the operation of the person identified by the control unit, and a second operation model stored is previously generated, the said control unit, said first operation model The first operation model and the second operation model are switched and used based on the accumulated amount of .
Further, the identification device according to one aspect of the present invention includes an input unit that receives image information of a person photographed by a camera, identification of the person based on the image information, and at least head and hand parts of the person. The motion model comprises a control unit that detects and identifies the person, the detected part, and the motion of the person based on the motion model registered for each person. , A first motion model generated based on the motion of the person identified by the control unit, and a second motion model generated and stored in advance, the control unit includes the first motion. The first motion model and the second motion model are used in combination based on the accumulated amount of the model.

The term "person" as used herein is not limited to the meaning of a specific individual, but also means a person of any age, gender, and physique in general.

According to the present invention, the posture of a person can be identified.

A block diagram showing an example of the configuration of the identification device according to the embodiment of the present invention. Block diagram showing an example of the configuration of the control unit Explanatory drawing showing an example of skeleton detection processing Explanatory drawing showing an example of an operation model Explanatory drawing which shows how person information and part information are compared with the motion model in embodiment of this invention. The flow diagram explaining the main operation of the identification apparatus which concerns on embodiment of this invention. Explanatory drawing which shows how person information and part information are compared with motion model in modification 2. The figure which shows an example of the hardware composition of the computer which realizes the function of each part by a program.

Hereinafter, each embodiment of the present invention will be described in detail with reference to the drawings. However, in each embodiment, configurations having the same function are designated by the same reference numerals, and duplicate description will be omitted.

First, an example of the configuration of the identification device 100 according to the embodiment of the present invention will be described with reference to FIG. 1a. FIG. 1a is a block diagram showing an example of the configuration of the identification device 100 according to the embodiment of the present invention.

The identification device 100, the camera 200, and the device 300 shown in FIG. 1a are mounted on a moving body such as an automobile, a railroad vehicle, a ship, or an aircraft. In the present embodiment, a case where the identification device 100, the camera 200, and the device 300 are mounted on an automobile will be described as an example.

The camera 200 is, for example, a distance measuring camera such as a stereo camera or a TOF (Time Of Flight) type camera, and is a device that photographs a range in which a person exists in a vehicle interior to acquire image information and identifies the image information. Output to 100. The image information includes at least a near-red image showing the intensity of reflection intensity of near-infrared light. Further, in the case of a stereo camera or a TOF camera, a distance image showing the perspective of the distance may be further included.

The device 300 is a device used in an automobile. The device 300 is, for example, a car navigation device, an audio device, an air conditioner device, a tablet, a smartphone, a rear camera, a side camera, a front camera, and an ETC on-board unit. Alternatively, the device 300 includes, for example, a handle, a shift lever (gear lever), a winker lever (winker switch), a wiper lever (wiper switch), a door lock switch, a power window elevating switch, an engine start switch, an accelerator pedal, a brake pedal, and the like. It may be a clutch pedal or the like.

The device 300 is connected to the identification device 100 via a wired network (for example, CAN: Controller Area Network), a wireless network (for example, a wireless LAN: Local Area Network), or a network in which a wired network and a wireless network are mixed. .. Then, the device 300 is controlled based on the control signal received from the identification device 100.

The identification device 100 includes an input unit 101 and a control unit 102.

The input unit 101 receives image information of a person from the camera 200 and outputs it to the control unit 102.

The control unit 102 identifies the person and detects at least the head and hand parts of the person based on the image information received from the input unit 101, and the identified person, the detected part, and the person. The movement of the person is identified based on the movement model registered for each person. With such a configuration, the posture of the person can be identified.

Next, an example of the configuration of the control unit 102 will be described with reference to FIG. 1b. FIG. 1b is a block diagram showing an example of the configuration of the control unit.

As shown in FIG. 1b, the control unit 102 includes a detection control unit 110, a skeleton detection unit 120, a skeleton motion identification unit 130, a person identification unit 140, an operation model unit 150, a site detection unit 160, a site motion identification unit 170, and an apparatus. A control unit 180 is provided.

The detection control unit 110 receives the image information from the camera 200, analyzes the image information, and outputs the image information to the skeleton detection unit 120, the person identification unit 140, and the site detection unit 160. Specifically, the detection control unit 110 determines whether or not skeleton detection is possible by analyzing image information, and if it is determined that skeleton detection is possible, an image is sent to the skeleton detection unit 120 and the person identification unit 140. Output information. On the other hand, if it is determined that the skeleton detection is not possible, it is further determined whether or not the motion model described later is available, and if it is determined that the motion model is available, the person identification unit 140 and the site detection unit are used. Image information is output to 160. The determination of whether or not skeleton detection is possible will be described later.

The skeleton detection unit 120 detects the skeleton position of a person based on the image information received from the detection control unit 110, and outputs the skeleton information indicating the detection result to the skeleton motion identification unit 130. The skeletal position of a person indicates the position of a representative point representing an area such as the head, neck, hips, shoulders, elbows, and hands. The position of the representative point is specified as two-dimensional coordinates on the image.

FIG. 2 shows an example of the skeleton detection process. Image 1000 is an example of an image of a person, and is a schematic view of an image of a person driving a car holding a steering wheel. Image 1001 is an example of a result image in which the skeleton position of a person in image 1000 is detected, and the positions of nine points which are representative points of the head, neck, waist, left shoulder, left elbow, left hand, right shoulder, right elbow, and right hand are displayed. Indicates the detected state. The skeleton information is the position coordinates of the detected representative points, and is obtained as the position coordinates (X1, Y1), (X2, Y2) ... (X9, Y9) of the nine representative points on the image. Skeleton detection can be realized by using an existing method as shown in Patent Document 1, for example.

Further, in the image 1001, since the positions of nine points, which are the representative points of the head, neck, waist, left shoulder, left elbow, left hand, right shoulder, right elbow, and right hand, can be detected, the detection control unit 110 can detect the skeleton. Judge that detection is possible. On the other hand, image 1002 shows an example of a result image obtained by detecting the skeleton position of a person obtained when the photographed person wears clothes having a low reflectance to near-red light such as a leather jacket. In image 1002, it can be seen that only the positions of three points, which are the representative points of the head, the left hand, and the right hand, can be detected. In this case, the detection control unit 110 determines that skeleton detection is not possible.

The skeleton motion identification unit 130 identifies the occupant's motion based on the skeleton information received from the skeleton detection unit 120. The occupant's movements include, for example, the movement of operating a device such as an air conditioner device, a steering wheel, and an audio device, a gesture such as searching for an object while driving, and turning around.

Identification of behavior based on skeletal information is performed by using an existing method, for example, a method such as a so-called random forest that identifies one of the predetermined patterns from multiple magnitude comparisons according to the tree structure of the parameters to be compared. be able to.

The skeleton motion identification unit 130 outputs motion information and skeleton information indicating the identification result of the occupant's motion as described above to the motion model unit 150. Further, the skeleton motion identification unit 130 outputs operation information to the device control unit 180.

The person identification unit 140 identifies who the person is based on the image information received from the detection control unit 110, and outputs the person information indicating the identification result to the motion model unit 150 and the part motion identification unit 170. do.

For example, existing face authentication technology can be used to realize person identification. For example, the face images of a plurality of people driving a car are registered in advance, the face images are acquired when those people actually drive the car, and the acquired face images and the registered face images are combined with the existing technology. By collating, it is possible to identify which registrant. As an existing technique for face recognition, there is a method of acquiring facial features by a method such as LBP (Local Binary Pattern) or HOG (Histogram of Oriented Gradient) and collating the features.

The motion model unit 150 generates and stores an motion model based on the motion information from the skeleton motion identification unit 130, the skeleton information, and the person information from the person identification unit 140, and stores the stored motion model as a part motion. Output to the identification unit 170.

The motion model is a database in which person information (registration number corresponding to a registered person), skeleton information (positional coordinates of feature points of the skeleton), and motion information (motion number corresponding to the type of motion) are stored in association with each other. Make it a form. Here, as the skeleton information, instead of registering all the feature point information output from the skeleton detection unit 120, the head, the left hand, and the head, the left hand, which are the detection targets in the part detection process by the part detection unit 160 described later, and Only the information of the feature point of the right hand is registered. As an example, FIG. 3 shows a relational database 1003 of person information, head / left / right hand skeleton information, and motion information.

In the present embodiment, the motion model database becomes available after accumulating a certain number or more of skeleton information, person information, and motion information detected in a state where the skeleton can be detected by the skeleton detection unit 120. For example, in order for a database of motion models to be available, it is necessary to accumulate 10,000 or more combinations of person information, skeleton information, and motion information.

The operation model unit 150 outputs the state information of whether or not the operation model is available to the detection control unit 110. The detection control unit 110 branches the process according to whether or not the operation model is available.

The site detection unit 160 detects the head position, the left hand position, and the right hand position as the site information of the person based on the image information received from the detection control unit 110, and transmits the detection result site information to the site motion identification unit 170. Output.

The skeleton detection unit 120 described above also detects the position coordinates of a plurality of feature points including the head position, the left hand position, and the right hand position, but here, when the image information of the body part of the person cannot be normally obtained. Use the detection method in.

For example, as a method of realizing detection of site information by combining existing methods, a mass of distance information existing at equal distances is detected using the distance information included in the image information, and the position of the mass is tracked between a plurality of frames. It is possible to use the method of

The part motion identification unit 170 identifies the movement of the occupant based on the person information from the person identification unit 140, the motion model from the motion model unit 150, and the site information from the site detection unit 160.

A method of the site motion identification process for identifying the motion of the occupant in the site motion identification unit 170 will be described with reference to FIG.

In the site motion identification process, the person information and the site information in the input information 2001 of the site motion identification unit 170 are collated with the person information and the skeleton information in the registration information 2002 already registered as the motion model, and the registered information 2002 is used. , The data whose personal information matches the personal information in the input information and whose skeletal information is closest to the part information in the input information is searched.

As a method of searching the registered information having the skeleton information closest to the part information of the input information, for example, the following method can be adopted. First, the distances between the corresponding parts (for example, the head positions, the left hand positions, and the right hand positions) between the input information and the registration information are calculated. Then, it is determined that the registration information having the smallest total distance between the calculated parts is the registration information having the skeleton information closest to the part information of the input information.

When collating the input information with the registration information, considering the identification accuracy of the operation, search for data in which the person information matches the person information in the input information and the skeleton information matches the part information in the input information. Is ideal. However, since the movement of a person when performing a predetermined movement is not always the same, it is conceivable that the skeleton information in the registration information does not match the part information in the input information.

In consideration of such a case, in the present embodiment, when collating the input information and the registration information, the person information matches the person information in the input information, and the skeleton information is the part information in the input information. Find the data closest to.

In the case of the example of FIG. 4, since the input information 2001 is the closest to the data of 2003 in the registration information 2002, the site motion identification unit 170 identifies that the motion information is "6" and displays the site motion information. Output to the device control unit 180.

The device control unit 180 generates control information for controlling the device 300 (not shown) based on the motion information from the skeletal motion identification unit 130 or the motion information from the site motion identification unit 170, and generates the generated control information. Output to device 300.

The device 300 is controlled based on the control information received from the device control unit 180.

In the example of FIG. 1b, the control unit 102 includes the skeleton detection unit 120 and the skeleton motion identification unit 130, but the control unit 102 may not include the skeleton detection unit 120 and the skeleton motion identification unit 130. good. Details will be described later.

Further, in the example of FIG. 1b, the control unit 102 is configured to include the device control unit 180, but the device control unit 180 may be provided separately from the control unit 102.

That is, the control unit 102 may have a configuration including at least a detection control unit 110, a person identification unit 140, an operation model unit 150, a site detection unit 160, and a site motion identification unit 170.

Next, an example of the operation of the control unit 102 according to the present embodiment will be described with reference to FIG. FIG. 5 is a flow chart showing an example of the operation flow of the control unit 102 according to the present embodiment.

First, in step S101, the detection control unit 110 receives image information from the camera 200. As described above, the image information is an image of the occupant.

Subsequently, in step S102, the detection control unit 110 determines whether or not the image can detect the skeleton based on the image information. Specifically, when all the representative point positions to be detected can be detected from the image information, it is determined that the image can detect the skeleton, and one of the representative point positions to be detected from the image information. When only the position of the representative point of the part can be detected, it is determined that the image is not capable of detecting the skeleton.

If it is determined in step S102 that the image is capable of detecting the skeleton (step S102: YES), the process proceeds to step S103, and in step S103, the skeleton detection unit 120 performs the skeleton detection process.

Subsequently, in step S104, the person identification unit 140 performs the person identification process, in step S105, the skeleton motion identification unit 130 performs the motion identification process, and in step S106, the device control unit 180 generates the device control information. Perform processing.

In step S107 following step S106, based on the skeleton information identified by the skeleton detection process of step S103, the person information identified by the person identification process of step S104, and the operation information identified by the operation identification process of step S105. The operation model unit 150 updates the operation model. Specifically, the work of adding the obtained skeleton information, person information, and motion information to the database of motion models is performed. The order of performing steps S106 and S107 may be reversed, or steps S106 and S107 may be performed at the same time.

If it is not determined in step S102 that the image is capable of detecting the skeleton (step S102: NO), the process proceeds to step S108, and in step S108, the detection control unit 110 determines whether or not the operation model is available.

If it is determined in step S108 that the operation model is available (step S108: YES), the process proceeds to step S109, and in step S109, the site detection unit 160 performs the site detection process.

Subsequently, in step S110, the person identification unit 140 performs the person identification process.

In step S111 following step S110, the site motion identification unit 170 is based on the site information identified by the site detection process in step S109, the person information identified by the person identification process in step S110, and the motion model stored in advance. , Performs site motion identification processing.

In step S112 following step S111, the device control unit 180 performs device control information generation processing.

If it is not determined in step S108 that the operation model is available (step S108: NO), the process proceeds to step S113, and in step S113, the detection control unit 110 determines that the operation identification process is impossible. , The execution of the operation identification process is prohibited, and the process is terminated by presenting error information.

As described above, in the identification device 100 of the present embodiment, the person is identified and at least the head and hand parts of the person are detected based on the image information received from the input unit 101, and the identification is performed. The movement of the person, the detected part, and the movement of the person are identified based on the movement model registered for each person.

Therefore, if the person is wearing clothes with low reflectance to near-infrared light such as a leather jacket, the distance between the body and arms is not output normally, and the shoulder and elbow joints cannot be detected, only the part information is available. However, the identification device 100 of the present embodiment can correctly identify the motion of a person even in such a case.

Further, for example, even if the positions of the head, left hand, and right hand of a person are the same, the type of movement may differ due to the habit of the person's gesture or the difference in physique. Even in such a case, since the identification device 100 of the present embodiment identifies the movement of the person based on the person information, the part information, and the basic movement model, it absorbs the difference in the movement due to the difference in the person and operates. Can be correctly identified.

Although the embodiment of the present invention has been described above, the present invention is not limited to the description of the embodiment. Hereinafter, each modification will be described.

<Modification example 1>
In the above-described embodiment, the identification device 100, the camera 200, and the device 300 have been described with reference to an example provided in the automobile, but the present invention is not limited thereto. The identification device 100, the camera 200, and the device 300 may be provided in a moving body other than an automobile (for example, a railroad vehicle, a ship, an aircraft, etc.), or may be provided in a place other than the moving body.

<Modification 2>
In the above-described embodiment, in the part motion identification process, the motion information in the registered information in which the person information is the same as the person information in the input information and the skeleton information is closest to the site information in the input information is the operation of the input information. I tried to identify it as information, but it is not limited to this.

For example, if the person information matches between the input information and the registration information, but there is no match between the part information in the input information and the skeleton information in the registration information, a predetermined number of registration information is provided in the order of proximity to the part information in the input information. When the search is performed and the operation information in the predetermined number of registered information is all the same, the operation information may be used as the operation information in the input information. The number of registration information to be searched is determined in advance.

Hereinafter, a specific embodiment of the second modification will be described with reference to FIG. In addition, in FIG. 6, in order to facilitate the explanation, only the registered information whose personal information is "2" is shown. Further, in the following description, an example in which the number of registered information to be searched is three is shown.

In FIG. 6, among the registration information 3002, as the registration information close to the part information in the input information 3001, the data of 3005 which is the closest data, 3007 which is the second closest data, and 3003 which is the third closest data. The data is searched. Then, in this case, since the operation information in the three searched data is all "1", it is determined that the operation information in the input information is "1".

In this way, the motion identification accuracy is improved as compared with simply using the skeleton information in the registered information closest to the part information in the input information. In particular, when the amount of registered information is small, there is a concern that the motion identification accuracy will decrease if the skeleton information in the registered information closest to the part information in the input information is used. Therefore, estimation of the motion information using the above method is effective. Is the target.

<Modification example 3>
In the above-described embodiment, the motion model database cannot be used until, for example, 10,000 or more combinations of person information, skeleton information, and motion information are accumulated, but the usage restrictions of the motion model are not limited to this.

Even if the accumulated amount of combinations of person information, skeleton information, and motion information is less than 10,000 pairs, it is possible to make the database of motion models available at the stage when information that can guarantee reliability to some extent is accumulated. be. Further, even when the accumulated amount of the combination of the person information, the skeleton information, and the motion information is small, it is possible to make the database of the motion model available only for a specific motion.

In particular, by combining with the method shown in the second modification, the database of the motion model can be used while ensuring the motion identification accuracy even when the accumulated amount of the combination of the person information, the skeleton information, and the motion information is small. ..

<Modification example 4>
In the above-described embodiment, the motion model is created and stored based on an image capable of detecting the skeleton, but the mode in which the motion model is stored is not limited to this. It is also possible to store the basic operation model (hereinafter referred to as “basic operation model”) in the identification device 100 in advance at the time of shipping the identification device 100.

In the basic motion model, standard skeletal information and motion information are registered for each predetermined age, gender, and physique. This basic motion model is pre-built with data obtained from various experiments and / or calculations.

When using the basic motion model, the age, gender, physique, etc. of the occupant are identified based on the image information, and an appropriate basic motion model is selected from a plurality of stored basic motion models.

Further, in order to improve the identification accuracy, it is preferable to store a large number of basic motion models having different ages, genders, and physiques in the identification device 100.

<Modification 5>
In the modified example 5, in the modified example 4, the database of the basic motion model and the database of the motion model created based on the image capable of detecting the skeleton are separately prepared, and the basic motion model and the skeleton can be detected. The operation model created based on the various images is switched and used.

Specifically, when the accumulated amount of the motion model created based on the image capable of detecting the skeleton is small, the accumulated amount of the motion model created based on the image capable of detecting the skeleton using the basic motion model is large. When the amount exceeds a predetermined amount, the motion model to be used is switched from the basic motion model to the motion model created based on the image capable of skeleton detection.

By doing so, it is possible to use the motion model database while ensuring the motion identification accuracy even when the accumulation amount of the motion model created based on the image capable of skeleton detection is small.

<Modification 6>
In the modified example 6, in the modified example 4, the basic motion model and the motion model created based on the image capable of detecting the skeleton are used in combination.

Specifically, when the basic motion model is used and the motion model is created based on the image capable of detecting the skeleton while the accumulated amount of the motion model created based on the image capable of detecting the skeleton is small. In addition, the database of the basic operation model is updated using the created operation model. Specifically, the work of adding the obtained skeleton information, person information, and motion information to the database of the basic motion model is performed.

By doing so, it is possible to use the motion model database while ensuring the motion identification accuracy even when the accumulation amount of the motion model created based on the image capable of skeleton detection is small. Furthermore, the reliability of the motion model can be further improved by incorporating highly reliable motion model data created based on an image capable of detecting the skeleton into the basic motion model.

<Modification 7>
In the above-described embodiment, the motion model is created based on an image capable of detecting the skeleton, but the creation of the motion model is not limited to this. For example, when the motion information is identified by the site motion identification unit 170 using the motion model, the person information, the site information, and the motion information are fed back to the motion model unit 150.

Specifically, in the flow diagram of FIG. 5, after step S111, the person information and the site information used in the site motion identification process and the motion information identified in the site motion identification process are output to the motion model unit 150. , The operation model unit 150 updates the operation model.

By doing so, in addition to creating a motion model based on an image capable of detecting the skeleton, a motion model can be created from person information, part information, and motion information identified based on the motion model, so that the motion model can be created. The amount of accumulated information increases dramatically. Of the data output from the site motion identification unit, only data with high reliability of motion identification, such as data in which a plurality of site information in the input information and skeleton information in the registration information match, is fed back. Therefore, it is possible to prevent the reliability of the operation model from being lowered.

<Modification 8>
In the above-described embodiment, it is assumed that the driver, who is an occupant, is wearing long-sleeved clothing having a low reflectance to near-red light, and the part to be detected by the part detection unit 160 is exposed to the head. , Left hand and right hand, but the part detected by the part detection unit 160 is not limited to this. The number of detected parts may be less than three or more than three. Alternatively, instead of or in addition to the hand (wrist to fingertip), a predetermined location on the upper limb, such as the elbow or shoulder, may be detected.

For example, considering the case where the driver wears short-sleeved clothes and gloves having low reflectance to near-red light, in the above-described embodiment, the head can be detected but both hands can be detected. It is not possible to perform the site motion identification process. However, if the elbow is detected instead of the hand, the site motion identification process can be performed.

Further, when the driver's seat is on the right side in the vehicle traveling direction, the driver often operates the device 300 with the left hand. Therefore, if the parts to be detected are one or more of the head and the upper left limb, even if the number of detection parts is reduced, the movement of the detection part is large, so that the driver's movement can be correctly identified.

<Modification example 9>
In the above-described embodiment, as the skeleton information stored in the motion model, only the skeleton information of the site to be detected by the site detection process in the site detection unit 160 is registered, but the skeleton stored in the motion model. The type of information is not limited to this.

For example, the motion model may store the entire skeleton information. In that case, the part motion identification unit 170 includes the part information of the part that could be detected from the image information and the skeleton information of the part corresponding to the part that could be detected from the image information among the skeleton information in the motion model. To match.

In this way, even if the combination of detectable parts changes, the part motion identification process can be performed.

The embodiments and modifications of the present invention have been described above.

FIG. 7 is a diagram showing a hardware configuration of a computer that programmatically realizes the functions of each part in the above-described embodiment and modification.

As shown in FIG. 7, the computer 2100 includes an input device 2101 such as an input button and a touch pad, an output device 2102 such as a display and a speaker, a CPU (Central Processing Unit) 2103, a ROM (Read Only Memory) 2104, and a RAM (Random). Access Memory) 2105 is provided. The computer 2100 is a reading device that reads information from a hard disk device, a storage device 2106 such as an SSD (Solid State Drive), a recording medium such as a DVD-ROM (Digital Versatile Disk Read Only Memory), or a USB (Universal Serial Bus) memory. 2107, a transmission / reception device 2108 that communicates via a network is provided. The above-mentioned parts are connected by the bus 2109.

Then, the reading device 2107 reads the program from the recording medium on which the program for realizing the functions of the above parts is recorded, and stores the program in the storage device 2106. Alternatively, the transmission / reception device 2108 communicates with the server device connected to the network, and stores the program downloaded from the server device for realizing the functions of the above parts in the storage device 2106.

Then, the CPU 2103 copies the program stored in the storage device 2106 to the RAM 2105, and sequentially reads and executes the instructions included in the program from the RAM 2105, thereby realizing the functions of the above-mentioned parts. Further, when the program is executed, the information obtained by the various processes described in each embodiment is stored in the RAM 2105 or the storage device 2106, and is appropriately used.

The identification device according to the present invention is useful for identifying a person who operates a predetermined device.

100 Identification device 101 Input unit 102 Control unit 110 Detection control unit 120 Skeletal detection unit 130 Skeletal motion identification unit 140 Person identification unit 150 Operation model unit 160 Part detection unit 170 Part operation identification unit 180 Equipment control unit 200 Camera 300 Equipment 1000, 1001 , 1002 Image 1003 Database 2001 Input Information 2002 Registration Information 2003 Data 2100 Computer 2101 Input Device 2102 Output Device 2103 CPU
2104 ROM
2105 RAM
2106 Storage device 2107 Reader 2108 Transmission / reception device 2109 Bus 3001 Input information 3002 Registration information 3003, 3004, 3005, 3006, 3007 Data

Claims (5)

An input unit that receives image information of a person taken by a camera,
Based on the image information, the person is identified and at least the head and hand parts of the person are detected, and the identified person, the detected part, and the movement of the person are registered for each person. A control unit that identifies the movement of the person based on the model is provided.
The operation model, viewed contains a first behavior model is generated based on the operation of the person identified by the control unit, and a second operation model stored are generated in advance, a,
The control unit is an identification device that switches between the first operation model and the second operation model based on the accumulated amount of the first operation model. The control unit uses the second operation model when the accumulated amount of the first operation model is less than the predetermined amount, and uses the operation model when the accumulated amount of the first operation model is equal to or more than the predetermined amount. The identification device according to claim 1 , wherein the second operation model is switched to the first operation model. When the control unit identifies the motion of the person, the control unit feeds back the person information of the person, the part information of the person, or the motion information indicating the identification result of the motion of the person to the first motion model. The identification device according to 1. An input unit that receives image information of a person taken by a camera,
Based on the image information, the person is identified and at least the head and hand parts of the person are detected, and the identified person, the detected part, and the movement of the person are registered for each person. A control unit that identifies the movement of the person based on the model is provided.
The motion model includes a first motion model generated based on the motion of the person identified by the control unit, and a second motion model generated and stored in advance.
Wherein the control unit is used by combining the second operation model and the first behavior model based on the accumulated amount of the first operation model, identification device. The control unit uses the second operation model when the accumulated amount of the first operation model is less than a predetermined amount, and uses the generated operation model when the first operation model is generated. The identification device according to claim 4, wherein the second operation model is updated.

Images