Information processing device, information processing method and program

Abstract

To provide an information processing device, a method, and a program that can predict a fall risk of a pedestrian and notify the pedestrian.SOLUTION: A fall prevention device (information processing device) 1 comprises: an image acquisition unit 202 that acquires an image of the ambient environment of a pedestrian captured by an imaging unit 106; a sensor information acquisition unit 201 that acquires and detects information on a body state of the pedestrian, such as acceleration, angular velocity, and azimuth, obtained from a sensor unit 105; a risk object detection unit 203 that detects a risk object based on the captured image; a walking state estimation unit 204 that estimates the walking state of the pedestrian based on the information on the body state; an approaching situation estimation unit 205 that estimates whether the pedestrian is approaching the risk object based on information from the sensor information acquisition unit and a detection result from the risk object detection unit; a fall risk determination unit 206 that determines a fall risk of the pedestrian based on estimation results from the approaching situation estimation unit and the walking state estimation unit; and a notification processing unit 207 that controls a notification unit 107 based on the determination result from the fall risk determination unit.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program that are capable of determining a risk of a pedestrian falling and notifying the pedestrian of the risk. [Background technology]

[0002] In recent years, the elderly population has been increasing, and the risk of tripping or falling while walking due to decreased cognitive ability, as well as traffic accidents, is increasing. This creates a vicious cycle in which long-term hospitalization due to injuries or fractures from accidents leads to decreased cognitive function, which in turn increases the risk of accidents. In addition, the risk of falling is not limited to elderly people, but also increases in situations where it is difficult to see where one is standing, such as when transporting luggage. The main causes of falls are sudden loss of balance due to overlooking the ground conditions, such as unevenness, slipping on water, etc. The risk of falling also increases when you are not paying attention or are concentrating on something else. Patent Document 1 proposes a fall prevention device (portable electronic device) that uses information from an acceleration sensor attached to the body to detect whether or not there is a risk of falling. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2013-232806 A Summary of the Invention [Problem to be solved by the invention]

[0004] In Patent Document 1, when a person falls, a lot of acceleration occurs in the body, and the posture is different from that in the standing position. Taking advantage of this, the risk of falling is determined by detecting physical changes. However, such technology does not inherently prevent injuries from occurring or their aftermath: the physical impact of a fall can actually occur several seconds after the risk of a fall is detected. For this reason, it is necessary to detect the risk of falling while taking into account the pedestrian's own condition and provide warnings to prevent falls from actually occurring.

[0005] The present invention has been made in consideration of the above problems, and aims to propose an information processing device etc. that is capable of determining a pedestrian's risk of falling and notifying the pedestrian. [Means for solving the problem]

[0006] In order to solve the above problems, an information processing device in an embodiment of the present invention is characterized by comprising a first acquisition means for acquiring an image of the pedestrian's surrounding environment, a second acquisition means for acquiring and detecting information related to the physical condition of the pedestrian, a hazard object detection means for detecting a hazard based on the image acquired by the first acquisition means, a walking state estimation means for estimating the walking state of the pedestrian based on the information acquired by the second acquisition means, an approach situation estimation means for estimating whether the pedestrian is approaching the hazard object based on the information acquired by the second acquisition means and the detection result by the hazard object detection means, a fall risk determination means for determining a risk of the pedestrian falling based on the estimation result by the approach situation estimation means and the estimation result by the walking state estimation means, and an alarm control means for controlling an alarm to be issued based on the determination result by the fall risk determination means. Effect of the Invention

[0007] According to the present invention, it is possible to realize an information processing device or the like that is capable of determining a risk of a pedestrian falling and notifying the pedestrian. [Brief description of the drawings]

[0008] [Figure 1] 1 is a diagram showing a state in which the fall prevention device 1 is used. [Diagram 2]1 is a diagram illustrating a hardware configuration of a tip-over prevention device 1. FIG. [Diagram 3] FIG. 2 is a diagram illustrating a functional configuration of the fall prevention device 1. [Figure 4] FIG. 13 is a flowchart showing a method for preventing a fall. [Diagram 5] 5A and 5B are diagrams showing images 501 and 502 captured in an environment with a step 503. [Figure 6] 1 shows images 601 and 602 captured in an environment where a road 603 is present. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, the embodiments will be described in detail with reference to the attached drawings. Note that the following embodiments do not limit the invention according to the claims. Although the embodiments describe a number of features, not all of these features are essential to the invention, and the features may be combined in any manner. Furthermore, in the attached drawings, the same reference numbers are used for the same or similar configurations, and duplicated descriptions are omitted.

[0010] <Fall prevention device 1> FIG. 1 is a diagram showing a state in which the fall prevention device 1 is in use. The fall prevention device (information processing device) 100 is a device carried by a user (pedestrian F) such as an elderly person while walking. As shown in Fig. 1, a fall prevention device 1 is carried by hanging it around the neck of a pedestrian F. When the pedestrian F approaches a hazard K such as a step and the risk of falling increases, the fall prevention device 1 alerts the pedestrian F with a warning sound, audio guidance, etc. The fall prevention device 1 is preferably equipped with a battery such as a lithium ion battery as a driving power source because it operates while the user is walking. The fall prevention device 1 may be, for example, a wearable device or a smartphone.

[0011] <Hardware configuration> FIG. 2 is a diagram showing a hardware configuration of the fall prevention device 1. As shown in FIG. The fall prevention device 1 includes a CPU 101 , a ROM 102 , a RAM 103 , a secondary storage device 104 , a sensor unit 105 , an imaging unit 106 , a notification unit 107 , a communication unit 108 , and a bus 109 .

[0012] The CPU 101 controls the entire fall prevention device 1 . The CPU 101 executes instructions according to a program (the program of the present invention) stored in the ROM 102 or RAM 103 . A fall prevention method (information processing method) is executed by executing each step of a flowchart described later. The ROM 102 is a non-volatile memory and stores various programs and data. The RAM 103 is a volatile memory and stores temporary data such as frame image data and sensor information. The secondary storage device 104 is a rewritable secondary storage device such as a hard disk drive or flash memory, and stores image information, programs, various settings, etc. This information is transferred to the RAM 103 and used by the CPU 101 to execute programs and use data.

[0013] The sensor unit 105 is an acceleration sensor, and outputs the direction of movement, speed, acceleration, and the like. The imaging unit 106 is composed of an imaging lens, a CMOS sensor, a video signal processing unit, and the like, and captures an image of the surrounding environment in front of the pedestrian F (in the traveling direction). The notification unit (notification means) 107 is a speaker, an LED, or the like, and notifies the user (pedestrian F) of the fall prevention device 1 of the risk of falling by sound or light. A communication unit (communication means) 108 connects to a network such as the Internet, an intranet, etc. The network may be the Internet, a wired LAN (Local Area Network), a wireless LAN, a WAN (Wide Area Network), etc. The bus 109 connects the CPU 101 and other components and enables mutual input and output of data.

[0014] <Function configuration> FIG. 3 is a diagram showing the functional configuration of the fall prevention device 1. As shown in FIG. The fall prevention device 1 includes a sensor information acquisition unit 201 , an image acquisition unit 202 , a hazard object detection unit 203 , a walking state estimation unit 204 , an approach situation estimation unit 205 , a fall risk determination unit 206 , and a notification processing unit 207 .

[0015] The sensor information acquisition unit (second acquisition means) 201 acquires information such as acceleration, angular velocity, and direction from the sensor unit 105. The sensor information acquisition unit 201 detects the movement state of the pedestrian F. The image acquisition unit (first acquisition means) 202 acquires an image captured by the imaging unit 106. The image acquisition unit 202 acquires an image of the surrounding environment in the travel direction of the pedestrian F.

[0016] The hazard detection unit (hazard detection means) 203 detects hazards that pose a risk of falling from the multiple images acquired by the image acquisition unit 202. The hazard detection unit 203 determines the position of the hazard K in the image.

[0017] The walking state estimation unit (walking state estimation means) 204 estimates the walking state of the walker F from the movement state of the walker F detected by the sensor information acquisition unit 201. The walking state estimation unit 204 estimates dangerous walking states such as unsteadiness, sudden changes in posture, etc. The walking state estimation unit 204 does not estimate a fall of the pedestrian F itself, but estimates body sway caused by poor physical condition, etc. For example, if pedestrian F is fatigued, the muscular strength in the legs is reduced and the walking rhythm is disturbed, resulting in swaying from side to side and longer amplitude compared to normal walking. The difference from normal walking can be determined from the difference in the time-series pattern change of acceleration and angular velocity. For example, by applying frequency analysis such as FFT to the signal waveforms of short-term acceleration and angular velocity, it is possible to extract abnormal frequency components. By calculating the difference from the frequency distribution during normal walking registered in advance, if the deviation exceeds a predetermined threshold, it can be estimated that the walking state is abnormal. It should be noted that the sensor unit 105 is not limited to detecting acceleration, angular velocity, etc. It is sufficient if it is possible to determine the difference in output between normal walking and other walking depending on the type of sensor used. In addition to FFT, a method using deep learning may be used as an analysis method. In that case, a model having a recurrent network is used to learn signal information using learning data of sensor output values ​​labeled as normal walking or other.

[0018] An approaching situation estimating section (approaching situation estimating means) 205 estimates whether the pedestrian F is approaching a dangerous object based on information from the sensor section 105 and the dangerous object detecting section 203. The approach situation estimation unit 205 obtains the positional relationship between the hazard object K and the pedestrian F and the traveling direction of the pedestrian F based on the position information of the hazard object detected by the hazard object detection unit 203 and the information obtained from the sensor unit 105. Then, the approach situation estimation unit 205 estimates whether the pedestrian F is approaching the hazard object K. It also estimates the time (arrival time) until the pedestrian F arrives at the hazard object K. This estimation can be performed using a motion prediction algorithm such as a Kalman filter.

[0019] The fall risk determination unit (fall risk determination means) 206 predicts the risk of the pedestrian F falling based on the information from the approach situation estimation unit 205 and the walking state estimation unit 204. The fall risk determination unit 206 predicts whether or not there is a high risk of the pedestrian falling in the future, which makes it possible to determine whether or not a warning (notification) should be issued to the pedestrian F. The fall risk determination unit 206 determines whether the pedestrian F is approaching an area where there is a risk of falling, based on the walking state of the pedestrian F and the time it takes for the pedestrian F to reach the danger target K. The fall risk determination unit 206 also determines whether the state is such that the risk of falling can be avoided. Then, the fall risk determination unit 206 adjusts the timing of the command to the notification processing unit 207.

[0020] The determination of whether or not there is a risk of falling is based on the relationship between the direction of movement of pedestrian F and changes in acceleration. The risk of falling can be determined by storing and registering in advance values ​​that indicate conditions such as unsteadiness or sudden changes in posture as dangerous conditions. Another method is to measure the change in the direction of movement and acceleration when the user starts walking, set the measured values ​​as standard values, and determine that any deviation from the standard values ​​is a dangerous state. Alternatively, there is also a method in which the standard values ​​are learned in advance according to the pedestrian F. When the risk of pedestrian F falling becomes dangerous, the fall risk determination unit 206 transmits a notification processing command to the notification processing unit 207.

[0021] The notification processing unit (notification control means) 207 transmits a notification signal to the notification unit 107 based on the notification processing command received from the fall risk determination unit 206. The notification processing unit 207 has a plurality of audio signal files related to prerecorded warnings, and selects a notification type (warning type) according to the level of the fall risk. The notification processing unit 207 is not limited to audio, and may use warning sounds or the like.

[0022] Since the alarm sound may be drowned out by noise in the surrounding environment, a vibration stimulus may be presented using a vibrator. Also, a visual stimulus may be presented by flashing a warning light such as an LED. It is sufficient that the pedestrian F can reliably receive the notification selected by the notification processing unit 207 according to the degree of danger. Stimulus information capable of calling the pedestrian F's attention to a fall risk that the pedestrian F is not aware of is selected. Since the sensitivity to sensory stimuli varies from user to user, the type of notification may be selectable by the user. Specifically, the type of voice or warning sound to be presented may be selectable by a GUI or the like. For example, a smartphone may be connected via the communication unit 108, allowing the user to select the type of notification from the smartphone.

[0023] <Fall prevention method (information processing method)> FIG. 4 is a flow chart showing the fall prevention method. The fall prevention method includes an imaging step S401, a sensing step S402, a hazard object detection step S403, a walking state estimation step S404, an approach situation estimation step S405, a fall risk determination step S406, and a notification step S407.

[0024] In the imaging step (first acquisition step) S401, the imaging unit 106 captures an image of the surrounding environment of the walker (user) F, and this image is acquired by the image acquisition unit 202. Since the fall prevention device 1 is hanging from the neck of the walker F, the environment in front of the walker F (in the traveling direction) is captured by the imaging unit 106.

[0025] In the sensing step (second acquisition step) S402, the sensor unit 105 detects the physical state of the walker F. The motion state of the walker F is detected by the sensor unit 105. Then, the sensor information acquisition unit 201 detects the motion state of the walker F from information such as acceleration, angular velocity, and direction obtained from the sensor unit 105.

[0026] In the hazard object detection step S403, a hazard object K is detected from the images obtained in the imaging step S401. The hazard object detection unit 203 detects, from the multiple images obtained by the image acquisition unit 202, a hazard object K that poses a risk of falling. The method for detecting the dangerous object K will be described in detail later.

[0027] In the walking state estimation step S404, the walking state of the walker F is estimated based on the information obtained in the sensing step S402. The walking state estimation unit 204 estimates the walking state of the walker F from the movement state of the walker F detected by the sensor information acquisition unit 201.

[0028] In the approach situation estimation step S405, it is estimated whether the pedestrian F is approaching the hazard object K based on the information obtained in the sensing step S402 and the hazard object detection step S403. The approach situation estimation unit 205 determines the positional relationship between the hazard object and the pedestrian F and the traveling direction of the pedestrian F based on the position information of the hazard object K detected by the hazard object detection unit 203 and the information obtained from the sensor unit 105. The approach situation estimation unit 205 also estimates the arrival time for the pedestrian F to reach the hazard object K.

[0029] In a fall risk determination step S406, a fall risk of the pedestrian F is determined based on the information obtained in the approach situation estimation step S405 and the walking state estimation step S404. The fall risk determination unit 206 determines a fall risk of the pedestrian F based on the information from the approach situation estimation unit 205 and the walking state estimation unit 204. The method for determining the risk of falling will be described in detail below.

[0030] In the notification step (notification control step) S407, the pedestrian F is notified of a fall risk based on the information obtained in the fall risk determination step S406. The notification processing unit 207 transmits a notification signal to the notification unit 107 based on the notification processing command received from the fall risk determination unit 206. Then, the notification unit 107 notifies the pedestrian F with an alarm sound or the like.

[0031] The detection of the hazardous object K in the hazardous object detection step S403 will be described in detail. Image recognition technology can be applied to detect dangerous objects. Specifically, this is achieved by collecting a large amount of learning data and pre-learning so that the system can detect dangerous objects K, such as steps and puddles, from images of the environment. Deep learning object recognition technology can estimate whether an object is dangerous from an image of the environment by learning a huge number of combinations of answers for input images using a learning model with many parameters.

[0032] The determination of the risk of falling in the fall risk determination step S406 will be described in detail. FIG. 5(a) is a diagram showing an image 501 captured in an environment with a step 503. FIG. 5B is a diagram showing an image 502 captured in an environment where a step 503 and a handrail 505 exist. In image 501, a handrail 505 is provided at a step 504. On the other hand, in image 502, a handrail is not provided at a step 503. In an environment equipped with facilities such as a barrier-free environment, the pedestrian F can secure the stability of the body by placing his / her hands on the handrail as an auxiliary measure. This reduces the risk of the pedestrian F falling. Therefore, the environment shown in the image 501 is determined to have a low risk of falling. On the other hand, the environment shown in image 502 is determined (labeled) as having a high risk of falling because there are no handrails.

[0033] When setting the risk of falling, the accuracy of the judgment can be adjusted efficiently by quantifying the degree of danger as a numerical index. For example, the degree of danger can be set to a numerical value (real number) between "0" and "1", with the higher the value indicating a higher degree of danger. The score can be set to change according to the degree of danger. Taking into consideration the stability of the learning system, the real number of the degree of danger is normalized and expressed in the range from "0" to "1".

[0034] FIG. 6(a) shows an image 601 captured in an environment where a road 603 exists. FIG. 6B shows an image 602 captured in an environment with a road 603 and a puddle 604 . Image 601 shows only a flat road 603. On the other hand, image 602 shows a puddle 604 in part of the flat road 603. Even in such an environment, the risk of falling is set using the same method as that used for image 501.

[0035] By preparing a large number of such images and training a deep learning model that outputs the level of danger for each image input, it is possible to estimate the level of danger of object K. In addition to deep learning, multiple pattern recognition methods may be used using geometric features, color features, etc. of an image. A method of integrating and judging the recognition results may also be used.

[0036] A major cause of many fall risks due to environmental factors is that the individual (pedestrian F) does not recognize in advance environmental factors that are different from normal conditions. Dangerous objects K, such as steps, exist everywhere in our living environment. Even if there are steps, falls can be prevented by walking carefully. Even if pedestrian F is notified of the risk of falling for all of these dangerous objects K, pedestrian F's reaction may become distracted. In contrast, the fall prevention device 1 of this embodiment detects hazardous objects K and predicts the risk of falling by taking into account the physical condition of the pedestrian F. This makes it possible to appropriately warn (warn) the pedestrian F to avoid the risk of falling. This makes it possible to prevent the risk of the pedestrian F falling before it actually happens. Furthermore, the reliability of the fall prevention device 1 by the pedestrian F is also increased.

[0037] <Modification> In the above-described embodiment, an acceleration sensor is used as the sensor unit 105, but other sensors may also be used. A vital sensor that detects vital signs such as pulse, respiration, body temperature, and blood pressure of the walker (user) F may be used as the sensor unit 105. Specifically, a blood pressure sensor, an electromyographic sensor, or the like may be used. The walking state estimation unit 204 can estimate an abnormal walking state, etc. by acquiring various vital signs from the sensor information acquisition unit 201. It can estimate an abnormal state of the walker F by comparing with the normal vital signs. For example, if the body temperature is higher than normal (standard value), it can be estimated that the person is in a dangerous state. Also, if the blood pressure is high, it can be estimated that the heart is under strain, the body is overloaded, and the person is in a distracted state.

[0038] <Modification> In the above-described embodiment, a notification is given to the pedestrian F to avoid the risk of falling, but a notification may be given to a third party or the like when the pedestrian F falls. The walking state estimation unit 204 detects a fall of the pedestrian F from information such as acceleration obtained from the sensor unit 105. When a fall of the pedestrian F is detected, the notification processing unit 207 transmits a notification to a pre-registered contact via the communication unit 108. This enables emergency response to the fall of the pedestrian F.

[0039] <Modification> In the above-described embodiment, a notification is given to the pedestrian F to avoid the risk of falling, but further prevention and measures against the risk of falling may be taken. Specifically, the fall risk determination unit 206 accumulates situations where a fall risk is predicted to be high, and creates a database. That is, information (fall risk information) such as an image of the pedestrian F immediately before the pedestrian F falls and the walking state is transmitted to a database server (storage means) (not shown) and accumulated therein. Then, a third party can access this database server and confirm situations where a fall risk is high. This can be useful for first aid and treatment of injuries sustained by the pedestrian F. Furthermore, if a fall accident is caused by another person, the video of the accident can serve as evidence.

[0040] When the fall risk determination unit 206 predicts that the risk of falling is high, it instructs the image acquisition unit 202, the walking state estimation unit 204, etc. to transmit and store the walking state and images (fall risk information) of the walker F at that time to the secondary storage device 104. In addition, when the walking state estimation unit 204 detects a fall of the walker F, it transmits the images etc. at that time (fall risk information) to the secondary storage device 104 and stores them.

[0041] <Modification> The fall prevention device 1 in the above-described embodiment is worn by the user, but is not limited to this. For example, a pedestrian wears a device having at least a sensor unit 105, an imaging unit 106, a notification unit 107, and a communication unit 108. The fall prevention device 1 may be configured to be externally connected to the device worn by the pedestrian wirelessly or via a wire. In this case, the fall prevention device 1 acquires various information from the device worn by the pedestrian, executes the above-mentioned processing, and controls so that a notification is issued from the device worn by the pedestrian. When the fall prevention device 1 is externally connected to a device worn by a pedestrian, the fall prevention device 1 may be, for example, a PC, a smartphone, etc. When the fall prevention device 1 is a smartphone, a sensor and an imaging unit of the smartphone may function as the sensor unit 105 and the imaging unit 106.

[0042] [Other embodiments] The fall prevention device 1 is not limited to being carried by the pedestrian F (user). The fall prevention device 1 may be any device that notifies the pedestrian F with an alarm sound or the like when the risk of falling increases. For example, the sensor unit 105, the imaging unit 106, and the notification unit 107 may be disposed in a location or device (housing) different from the pedestrian F. In other words, the sensor unit 105, the imaging unit 106, and the notification unit 107 may be configured to be externally connected to the fall prevention device 1.

[0043] The present invention can be embodied as, for example, a system, an apparatus, a method, a program, or a recording medium (storage medium), etc. Specifically, the present invention may be applied to a system composed of multiple devices (for example, a host computer, an interface device, an imaging device, a web application, etc.), or may be applied to an apparatus composed of a single device. Needless to say, the object of the present invention can be achieved by the following: That is, a recording medium (or storage medium) on which is recorded a program code (computer program) of software that realizes the functions of the above-mentioned embodiments is supplied to a system or device. The storage medium is, of course, a computer-readable storage medium. Then, a computer (or a CPU or MPU) of the system or device reads and executes the program code stored in the recording medium. In this case, the program code itself read from the recording medium realizes the functions of the above-mentioned embodiments, and the recording medium on which the program code is recorded constitutes the present invention.

[0044] The present invention can also be realized by a process in which a program for realizing one or more functions of the above-mentioned embodiments is supplied to a system or device via a network or a storage medium, and one or more processors in a computer of the system or device read and execute the program. The present invention can also be realized by a circuit (e.g., ASIC) for realizing one or more functions. The present invention is not limited to the above-described embodiments, and various modifications and variations are possible without departing from the spirit and scope of the invention. Therefore, the following claims are appended to disclose the scope of the invention.

[0045] The disclosure of this embodiment includes the following anti-tip device. (Configuration 1) A first acquisition means for acquiring an image of an environment surrounding a pedestrian; A second acquisition means for acquiring and detecting information regarding a physical condition of the pedestrian; a hazard detection means for detecting a hazard based on the captured image acquired by the first acquisition means; a walking state estimation means for estimating a walking state of the pedestrian based on the information acquired by the second acquisition means; an approach status estimation means for estimating whether the pedestrian is approaching the hazard based on the information acquired by the second acquisition means and the detection result by the hazard detection means; a fall risk determination means for determining a fall risk of the pedestrian based on the estimation result by the approach situation estimation means and the estimation result by the walking state estimation means; A notification control means for controlling the notification means to notify based on the result of the determination by the fall risk determination means; An information processing device comprising: (Configuration 2) 2. The information processing device according to claim 1, wherein the second acquisition means acquires information indicating a motion state of the walker as the information relating to the physical condition based on information acquired by an acceleration sensor. (Configuration 3) 3. The information processing device according to claim 1, wherein the second acquisition means acquires information indicating a physiological state of the walker as the information relating to the physical condition based on information acquired by a vital sensor. (Configuration 4) 4. The information processing device according to any one of claims 1 to 3, wherein the notification control means switches the type of notification depending on the content of the risk of falling. (Configuration 5) 4. The information processing device according to any one of configurations 1 to 3, wherein the notification control means controls the notification means having at least one of a speaker, a vibrator, and a warning light. (Configuration 6) 6. An information processing device according to any one of configurations 1 to 5, further comprising a communication means for communicating information from the notification control means to an outside. (Configuration 7) 7. The information processing device according to any one of claims 1 to 6, further comprising a storage means for receiving and storing fall risk information via the communication means. (Configuration 8) 8. The information processing device according to any one of claims 1 to 7, wherein the first acquisition means acquires a captured image from an externally connected imaging device. (Configuration 9) 9. The information processing device according to any one of configurations 1 to 8, wherein the second acquisition means acquires information regarding the physical condition based on information acquired from an externally connected sensor. (method) a first acquisition step of acquiring an image of an environment surrounding a pedestrian; A second acquisition step of acquiring and detecting information regarding a physical condition of the pedestrian; a hazard detection step of detecting a hazard based on the captured image acquired by the first acquisition step; a walking state estimating step of estimating a walking state of the pedestrian based on the information acquired in the second acquisition step; an approach situation estimation step of estimating whether the pedestrian is approaching the hazard based on the information acquired in the second acquisition step and the detection result in the hazard detection step; a fall risk determination step of determining a fall risk of the pedestrian based on an estimation result in the approach situation estimation step and an estimation result in the walking state estimation step; a notification control step of controlling the notification means to notify based on the result of the fall risk determination step; 13. An information processing method comprising: (program) A program for causing a computer to function as the information processing device according to any one of claims 1 to 9.

[0046] Although the preferred embodiments of the present invention have been described above, the present invention is not limited to these embodiments, and various modifications and changes are possible within the scope of the gist of the present invention. [Explanation of symbols]

[0047] 1. Fall prevention device (information processing device) 101 CPU 104 Secondary storage device 105 Sensor section 106 Imaging unit 107 Notification unit (notification means) 108 Communication Department (Communication Means) 201 Sensor information acquisition unit (second acquisition means) 202 Image acquisition unit (first acquisition means) 203 Danger object detection unit (danger object detection means) 204 Walking state estimation unit (walking state estimation means) 205 Approach situation estimation unit (approach situation estimation means) 206 Fall risk determination unit (fall risk determination means) 207 Notification processing unit (notification control means) 501,502 images 503,504 Steps (hazards) 601,602 images 604 Puddle (Danger) F Pedestrians (users) K Dangerous object

Claims

1. A first acquisition means for acquiring an image of an environment surrounding a pedestrian; A second acquisition means for acquiring and detecting information regarding a physical condition of the pedestrian; a hazard detection means for detecting a hazard based on the captured image acquired by the first acquisition means; a walking state estimation means for estimating a walking state of the pedestrian based on the information acquired by the second acquisition means; an approach status estimation means for estimating whether the pedestrian is approaching the hazard based on the information acquired by the second acquisition means and the detection result by the hazard detection means; a fall risk determination means for determining a fall risk of the pedestrian based on the estimation result by the approach situation estimation means and the estimation result by the walking state estimation means; A notification control means for controlling the notification means to notify based on the result of the determination by the fall risk determination means; An information processing device comprising:

2. The information processing apparatus according to claim 1 , wherein the second acquisition means acquires information indicating a motion state of the walker as the information relating to the physical condition, based on information acquired by an acceleration sensor.

3. The information processing apparatus according to claim 1 , wherein the second acquisition means acquires information indicating a physiological state of the walker as the information relating to the physical condition based on information acquired by a vital sensor.

4. The information processing device according to claim 1 , wherein the notification control means switches the type of notification depending on the content of the risk of falling.

5. 2. The information processing apparatus according to claim 1, wherein the notification control means controls the notification means having at least one of a speaker, a vibrator, and a warning light.

6. 2. The information processing apparatus according to claim 1, further comprising a communication means for communicating the information from said notification control means to an outside.

7. 7. The information processing apparatus according to claim 6, further comprising a storage unit that receives and stores the fall risk information via the communication unit.

8. The information processing apparatus according to claim 1 , wherein the first acquisition means acquires a captured image from an externally connected imaging device.

9. The information processing apparatus according to claim 1 , wherein the second acquisition means acquires the information regarding the physical condition based on information acquired from an externally connected sensor.

10. a first acquisition step of acquiring an image of an environment surrounding a pedestrian; A second acquisition step of acquiring and detecting information regarding a physical condition of the pedestrian; a hazard detection step of detecting a hazard based on the captured image acquired by the first acquisition step; a walking state estimating step of estimating a walking state of the pedestrian based on the information acquired in the second acquisition step; an approach situation estimation step of estimating whether the pedestrian is approaching the hazard based on the information acquired in the second acquisition step and a detection result in the hazard detection step; a fall risk determination step of determining a fall risk of the pedestrian based on an estimation result in the approach situation estimation step and an estimation result in the walking state estimation step; a notification control step of controlling the notification means to notify based on the result of the fall risk determination step; 13. An information processing method comprising:

11. A program for causing a computer to function as the information processing device according to any one of claims 1 to 9.

Images