CN111899470A - Human body falling detection method, device, equipment and storage medium - Google Patents
Human body falling detection method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111899470A CN111899470A CN202010867487.9A CN202010867487A CN111899470A CN 111899470 A CN111899470 A CN 111899470A CN 202010867487 A CN202010867487 A CN 202010867487A CN 111899470 A CN111899470 A CN 111899470A
- Authority
- CN
- China
- Prior art keywords
- human body
- falling
- fall
- detected
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
- G08B21/043—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis detecting an emergency event, e.g. a fall
Landscapes
- Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- General Health & Medical Sciences (AREA)
- Gerontology & Geriatric Medicine (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a human body falling detection method, a human body falling detection device, human body falling detection equipment and a storage medium, and belongs to the technical field of security and protection monitoring. The method comprises the steps of shooting a human body to be detected to obtain a human body activity image set of the human body to be detected; classifying the motion state of each human body moving image in the human body moving image set based on a pre-constructed falling detection model to obtain the motion state type corresponding to each human body moving image; the human body to be detected is subjected to falling detection according to the motion state type, the motion state classification is carried out on the human body activity image set of the human body to be detected through a pre-constructed falling detection model, the falling detection is carried out on the human body to be detected according to the motion state type corresponding to each human body activity image, the falling detection can be accurately carried out on the human body according to a plurality of human body activity images with different motion state types, the problem of low detection accuracy is solved, and the accuracy of the human body falling detection is improved.
Description
Technical Field
The invention relates to the technical field of security monitoring, in particular to a human body falling detection method, a human body falling detection device, human body falling detection equipment and a storage medium.
Background
With the development of economic technology and the improvement of living standard of people, the intelligent home technology is developed greatly. The demand of human health monitoring under the house environment is increasingly urgent, and the nursing of old people is a key direction in the intelligent house field. Especially for elderly people living alone, falling down is a major health threat in their indoor activities. Therefore, the research of fall detection is attracting more and more attention, the existing old people nursing system is based on multiple sensors and profile detection, the fall detection methods are simple in algorithm, but the detection result accuracy is low, and the false alarm rate in practical application is high.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide a human body falling detection method, a human body falling detection device, human body falling detection equipment and a storage medium, and aims to solve the technical problems of low detection accuracy and high false alarm rate in the prior art.
In order to achieve the above object, the present invention provides a human fall detection method, comprising the steps of:
shooting a human body to be detected to obtain a human body activity image set of the human body to be detected;
classifying the motion state of each human body moving image in the human body moving image set based on a pre-constructed falling detection model to obtain the motion state type corresponding to each human body moving image;
and carrying out falling detection on the human body to be detected according to the motion state type.
Optionally, the step of detecting that the human body to be detected falls down according to the motion state type includes:
screening out a motion state type which accords with a first preset type from the motion state types;
acquiring the type number corresponding to the motion state type conforming to the first preset type;
determining falling confidence corresponding to the human body to be detected according to the type number;
and carrying out falling detection on the human body to be detected according to the falling confidence coefficient.
Optionally, the step of detecting that the human body to be detected falls according to the fall confidence coefficient includes:
comparing the fall confidence level with a preset confidence level;
if the falling confidence coefficient does not exceed the preset confidence coefficient, judging that the human body to be detected does not have falling risk;
and if the falling confidence coefficient exceeds the preset confidence coefficient, judging that the human body to be detected has falling risk.
Optionally, after the step of determining that the human body to be detected has a risk of falling if the fall confidence coefficient exceeds the preset confidence coefficient, the method further includes:
when the human body to be detected has a falling risk, acquiring a falling tendency image set of the human body to be detected within a preset time;
classifying falling tendency states of the falling tendency images in the falling tendency image set based on the pre-constructed falling detection model so as to obtain falling tendency state types corresponding to the falling tendency images;
screening fall tendency state types which accord with a second preset type from the fall tendency state types;
acquiring the type number corresponding to the falling tendency state type in accordance with the second preset type;
and when the type number reaches a number threshold value, judging that the human body to be detected has a falling event, and outputting a falling alarm prompt.
Optionally, before the step of classifying motion states of the human body moving images in the human body moving image set based on a pre-constructed fall detection model to obtain a motion state type corresponding to each human body moving image, the method further includes:
acquiring a sample falling video of the human body to be detected and sample falling time corresponding to the sample falling video;
generating a sample fall image queue according to the sample fall time and the sample fall video;
processing each falling image in the sample falling image queue to obtain a target sample falling image queue;
training the target sample fall image queue to obtain a fall detection model.
Optionally, the step of generating a sample fall image queue from the sample fall time and the sample fall video comprises:
sequentially dividing the sample fall video into a plurality of fall video fragments according to the time sequence corresponding to the sample fall time;
intercepting the initial frame image of each falling video clip and the final frame image of the last falling video clip;
and generating a sample falling image queue according to the initial frame image and the final frame image of each falling video clip.
Optionally, the step of processing each fall image in the sample fall image queue to obtain a target sample fall image queue comprises:
denoising each fallen image in the sample fallen image queue according to the human motion characteristics to obtain a plurality of human fallen images;
classifying the plurality of human body falling images in sequence according to the time sequence corresponding to the sample falling time;
and setting corresponding category identification for each classified human body falling image so as to obtain a target sample falling image.
Further, to achieve the above object, the present invention also proposes a human fall detection apparatus, comprising:
the system comprises an acquisition module, a detection module and a display module, wherein the acquisition module is used for shooting a human body to be detected so as to obtain a human body activity image set of the human body to be detected;
the classification module is used for classifying the motion states of all human body moving images in the human body moving image set based on a pre-constructed falling detection model so as to obtain the motion state types corresponding to all human body moving images;
and the judging module is used for carrying out falling detection on the human body to be detected according to the motion state type.
Further, to achieve the above object, the present invention also proposes a human body fall detection apparatus comprising: a memory, a processor and a human fall detection program stored on the memory and executable on the processor, the human fall detection program being configured to implement the steps of the human fall detection method as described above.
Furthermore, to achieve the above object, the present invention also proposes a storage medium having stored thereon a human fall detection program which, when executed by a processor, implements the steps of the human fall detection method as described above.
The human body to be detected is shot to obtain a human body activity image set of the human body to be detected; classifying the motion state of each human body moving image in the human body moving image set based on a pre-constructed falling detection model to obtain the motion state type corresponding to each human body moving image; the human body to be detected is subjected to falling detection according to the motion state type, the motion state classification is carried out on the human body activity image set of the human body to be detected through a pre-constructed falling detection model, the falling detection is carried out on the human body to be detected according to the motion state type corresponding to each human body activity image, the falling detection can be accurately carried out on the human body according to a plurality of human body activity images with different motion state types, and therefore the accuracy of the falling detection of the human body is improved.
Drawings
Fig. 1 is a schematic structural diagram of a human fall detection device according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a first embodiment of a human fall detection method according to the invention;
fig. 3 is a schematic flow chart of a human fall detection method according to a second embodiment of the invention;
fig. 4 is a schematic flow chart of a human fall detection method according to a third embodiment of the invention;
fig. 5 is a schematic diagram of a human body falling shooting method according to the present invention;
fig. 6 is a block diagram of the first embodiment of the human fall detection apparatus of the invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a human fall detection device in a hardware operating environment according to an embodiment of the present invention.
As shown in fig. 1, the personal fall detection apparatus may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a WIreless interface (e.g., a WIreless-FIdelity (WI-FI) interface). The Memory 1005 may be a Random Access Memory (RAM) Memory, or may be a Non-Volatile Memory (NVM), such as a disk Memory. The memory 1005 may alternatively be a storage device separate from the processor 1001.
It will be appreciated by a person skilled in the art that the arrangement shown in fig. 1 does not constitute a limitation of a human fall detection apparatus, and may comprise more or fewer components than shown, or some components in combination, or a different arrangement of components.
As shown in fig. 1, the memory 1005, which is a storage medium, may include therein an operating system, a network communication module, a user interface module, and a human fall detection program.
In the human fall detection device shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the human body fall detection apparatus of the present invention may be provided in the human body fall detection apparatus, which calls the human body fall detection program stored in the memory 1005 through the processor 1001 and executes the human body fall detection method provided by the embodiment of the present invention.
An embodiment of the present invention provides a method for detecting a human body fall, and referring to fig. 2, fig. 2 is a schematic flow diagram of a first embodiment of the method for detecting a human body fall according to the present invention.
In this embodiment, the human fall detection method includes the following steps:
step S10: shooting a human body to be detected to obtain a human body activity image set of the human body to be detected.
It should be noted that the execution subject of this embodiment may be an image data processing device for shooting an object to obtain an object moving image and performing analysis processing on the object moving image, and may also be a terminal device having functions of acquiring image data and processing image data, which is not limited in this embodiment.
In this embodiment, a human body to be detected is photographed by a camera, the camera includes an analog camera, a digital camera, a high definition camera, a Charge Coupled Device (CCD) camera, a spherical camera, and the like, in this embodiment, no limitation is imposed on a transmission signal, a resolution, a sensor model, and an appearance of the camera, and a suitable camera can be adopted according to an actual situation. In addition, the camera can collect human body moving images of a human body to be detected at a speed of 24 frames per second, and can also collect the human body moving images at other speeds.
It should be noted that, when the camera shoots the human body to be detected, the acquired image is actually the region image of the motion region where the human body to be detected is located, and the region image may have other moving objects, so the human body moving image is obtained by substantially processing the region image.
Step S20: and classifying the motion state of each human body moving image in the human body moving image set based on a pre-constructed falling detection model so as to obtain the motion state type corresponding to each human body moving image.
It should be noted that each human body moving image in the human body moving image set contains a motion state of a human body to be detected, different motion states all have corresponding motion state types, the motion state types substantially reflect different corresponding motion states in a human body falling process, the whole process from standing to falling of the human body is divided into a plurality of different stages, and a corresponding motion state type is defined for the motion state of the human body in each stage, for example, the whole falling process of the human body is divided into a standing state, a falling state a, a falling state B, a falling state C and a falling state, the motion state type corresponding to the standing state is defined as 1, the motion state type corresponding to the falling state a is defined as 2, the motion state type corresponding to the falling state B is defined as 3, the motion state type corresponding to the falling state C is defined as 4, and the motion state type corresponding to the falling state is defined as 5, the motion state type corresponding to the motion state in this embodiment may be set according to an actual situation, and is not limited in this embodiment.
In specific implementation, the motion state classification is carried out on each human body moving image through a falling detection model to obtain the motion state type corresponding to each human body moving image, and the falling detection model is constructed in advance based on a sample falling image of a human body.
Step S30: and carrying out falling detection on the human body to be detected according to the motion state type.
It is easy to understand that after the motion state types corresponding to the human body moving images are obtained, the falling detection of the human body to be detected can be performed according to the motion state types corresponding to the human body moving images, for example, if the motion state type corresponding to the standing state is 1, it can be determined that the human body to be detected has not a falling event according to the motion state type 1, and if the motion state type corresponding to the motion state is 2, it can be determined that the human body to be detected has a falling event according to the motion state type 2.
In the embodiment, a human body to be detected is shot to obtain a human body activity image set of the human body to be detected; classifying the motion state of each human body moving image in the human body moving image set based on a pre-constructed falling detection model to obtain the motion state type corresponding to each human body moving image; the human body to be detected is subjected to falling detection according to the motion state type, the motion state classification is carried out on the human body activity image set of the human body to be detected through a pre-constructed falling detection model, the falling detection is carried out on the human body to be detected according to the motion state type corresponding to each human body activity image, the falling detection can be accurately carried out on the human body according to a plurality of human body activity images with different motion state types, and therefore the accuracy of the falling detection of the human body is improved.
Referring to fig. 3, fig. 3 is a flowchart illustrating a human fall detection method according to a second embodiment of the present invention.
Based on the first embodiment described above, in the present embodiment, the step S30 includes:
step S301: and screening out the motion state type which accords with a first preset type from the motion state types.
In this embodiment, a motion state type meeting a first preset type is screened from motion state types according to a motion state type for fall detection of a human body to be detected, where the first preset type is a motion state type in which a fall event has not occurred but a fall risk exists, for example, a whole fall process of the human body is divided into a standing state, a falling state a, a falling state B, a falling state C, and a falling state, a motion state type corresponding to the standing state is defined as 1, a motion state type corresponding to the falling state a is defined as 2, a motion state type corresponding to the falling state B is defined as 3, a motion state type corresponding to the falling state C is defined as 4, and a motion state type corresponding to the falling state is defined as 5, and the first preset type may be set as 1, 2, or 1, 2, 3, and the like.
Step S302: and acquiring the type number corresponding to the motion state type conforming to the first preset type.
In a specific implementation, after the motion state type conforming to the first preset type is screened out, the corresponding type number is required to be obtained, the type number represents the number of different motion state types, for example, assuming that the first preset type is 1, 2, 3, the motion state type corresponding to the human body moving image X is 1 based on a pre-constructed fall detection model, the motion state type corresponding to the human body moving image Y is 2, the motion state type corresponding to the human body moving image Z is 3, the type number corresponding to the motion state type conforming to the first preset type is 3, if the motion state type corresponding to the human body moving image Z is 4, the type number corresponding to the motion state type conforming to the first preset type is 2, if the motion state type corresponding to the human body moving image Y and the motion state type corresponding to the human body moving image Z are both 2, the number of types corresponding to the motion state type conforming to the first preset type is 2.
Step S303: and determining the falling confidence corresponding to the human body to be detected according to the type number.
It should be noted that the motion state conforming to the first preset type indicates that the fall of the human body to be detected is likely to happen, and the more the number of types of motion state conforming to the first preset type is, the greater the possibility that the human body to be detected is likely to fall is indicated, in this embodiment, the fall confidence level represents the possibility that the human body to be detected is likely to fall, and the greater the fall confidence level is, the greater the possibility that the corresponding human body to be detected is likely to fall is indicated, so that the fall confidence level corresponding to the human body to be detected can be determined according to the obtained number of types of motion state, for example, when the obtained number of types is 1, the fall confidence level corresponding to the human body to be detected can be determined to be 50%, when the obtained number of types is 2, the fall confidence level corresponding to the human body to be detected can be determined to be 85%, and the, the present embodiment is not limited.
Step S304: and carrying out falling detection on the human body to be detected according to the falling confidence coefficient.
In a specific implementation, the falling state of the human body to be detected can be determined according to the falling confidence, so that the falling detection of the human body to be detected is realized, and the step of performing the falling detection on the human body to be detected according to the falling confidence comprises the following steps: comparing the fall confidence level with a preset confidence level; if the falling confidence coefficient does not exceed the preset confidence coefficient, judging that the human body to be detected does not have falling risk; and if the falling confidence coefficient exceeds the preset confidence coefficient, judging that the human body to be detected has falling risk.
It should be noted that the greater the falling confidence coefficient is, the greater the possibility that the human body to be detected is about to fall is, the preset confidence coefficient indicates that it can be determined that the human body to be detected is about to fall, in this embodiment, the falling confidence coefficient is compared with the preset confidence coefficient, if the falling confidence coefficient does not exceed the preset confidence coefficient, it is indicated that the possibility that the human body to be detected is about to fall is low, it can be determined that the human body to be detected does not have a falling risk, and if the falling confidence coefficient exceeds the preset confidence coefficient, it is indicated that the possibility that the human body to be detected is about to.
Further, after determining that the human body to be detected has a risk of falling, the detection is continuously performed on the human body to be detected, so as to determine whether the human body to be detected finally has a falling event, in this embodiment, after the step S304, the method further includes: when the human body to be detected has a falling risk, acquiring a falling tendency image set of the human body to be detected within a preset time; classifying falling tendency states of the falling tendency images in the falling tendency image set based on the pre-constructed falling detection model so as to obtain falling tendency state types corresponding to the falling tendency images; screening fall tendency state types which accord with a second preset type from the fall tendency state types; acquiring the type number corresponding to the falling tendency state type in accordance with the second preset type; and when the type number reaches a number threshold value, judging that the human body to be detected has a falling event, and outputting a falling alarm prompt.
It should be noted that, from the time of determining that there is a risk of falling of the human body to be detected, a falling tendency image set of the human body to be detected within a preset time is obtained, the preset time is a standard falling time in a falling detection model, after the falling tendency image set is obtained, each falling tendency image in the falling tendency image set still needs to be classified, the classification process is similar to the process of classifying the motion states of each human body moving image in the human body moving image set, the classification process is also performed based on a preset falling detection model, the falling tendency state type corresponding to each falling tendency image is obtained, then the falling tendency state type conforming to a second preset type is screened out, the second preset type is a falling tendency state type belonging to the human body falling process, and the more types of the falling tendency state type conforming to the second preset type are, the more the possibility of falling incidents of the human body to be detected is indicated, therefore, whether the human body to be detected has a falling event or not can be judged according to the type number of the falling and inclining state types.
It should be noted that the second preset type and the first preset type are related to each other, and the second preset type includes the first preset type, for example, the first preset type is 1, 2, 3, 4, and 5, and the second preset type is 1, 2, 3, 4, 5, …, and 20, where the first preset type is used to determine whether the human body to be detected has a risk of falling, and the second preset type is used to determine whether the human body to be detected finally has a fall event after determining that the human body to be detected has a risk of falling. For example, the first preset type is defined as 1, 2, 3, 4 and 5, the second preset type is defined as 1, 2, 3, 4, 5, … and 20, the preset confidence is 85% and the number threshold is 10, assuming that the motion state types corresponding to the human body moving images of the human body a to be detected are 1 and 3, the falling confidence corresponding to the human body a to be detected is 90%, it can be determined that the human body a to be detected has a falling risk, then the falling tendency image set of the human body a to be detected within the preset time is continuously obtained, the classification result obtained by classifying the falling tendency images in the falling tendency image set is the falling tendency state types 1, 3, 6, 7, 8, 10, 12, 14, 16, 17 and 18, the number of the types corresponding to the falling tendency state types is 11, and it can be determined that the human body a to be detected has a falling event.
In the embodiment, a motion state type which is in accordance with a first preset type is screened out from the motion state types; acquiring the type number corresponding to the motion state type conforming to the first preset type; determining falling confidence corresponding to the human body to be detected according to the type number; and performing fall detection on the human body to be detected according to the fall confidence coefficient, performing fall detection according to the type number corresponding to the motion state type of the human body moving image and the fall confidence coefficient corresponding to the type number, determining the fall state type of the fall image based on a second preset type when the human body to be detected has a fall risk, further judging whether a fall event occurs, and enabling the human body fall detection to be more accurate through the combination of the fall risk judgment and the fall event occurrence.
Referring to fig. 4, fig. 4 is a flowchart illustrating a human fall detection method according to a third embodiment of the present invention.
Based on the first embodiment or the second embodiment, a third embodiment of a human fall detection method of the present invention is proposed.
Taking the first embodiment as an example, in this embodiment, step S20 is preceded by:
step S201: and acquiring the sample falling video of the human body to be detected and the sample falling time corresponding to the sample falling video.
It is easily understood that a fall detection model needs to be constructed in advance before classifying moving human body images or fall tendency images based on the fall detection model.
In this embodiment, a sample fall video of a human body to be detected and a sample fall time corresponding to the sample fall video are obtained, the fall video includes a whole fall process from the beginning of the human body falling to the beginning of the falling, and the sample fall time is the total duration time from the moment when the human body starts falling to the moment when the human body finally falls and stays still.
It should also be noted that the fall detection modelThe standard fall time in (1) is T, T ═ Leσ, wherein LeIs the average value of all fall times, σ is the standard deviation of the fall times, σ can be defined according to the actual requirement, which is not limited in this embodiment, Le＝(L1+L2+L3+…+_Ln) N, wherein L1、L2、L3、…、LnFor example, as shown in fig. 5, N individuals prepare M fall modes and shoot through 8 cameras, so that 8 × M × N, L can be obtained1、L2、L3、…、LnAnd the sample fall time corresponding to each fall video in 8 × M × N is represented, and 8 × M × N is the number of fall videos.
Step S202: and generating a sample fall image queue according to the sample fall time and the sample fall video.
In a specific implementation, the fall videos are divided according to a time sequence of the fall time of the sample, and the fall images in each fall video are captured, so as to generate a sample fall image queue, where the step S202 in this embodiment includes: sequentially dividing the sample fall video into a plurality of fall video fragments according to the time sequence corresponding to the sample fall time; intercepting the initial frame image of each falling video clip and the final frame image of the last falling video clip; and generating a sample falling image queue according to the initial frame image and the final frame image of each falling video clip.
It should be noted that the sample fall videos are sequentially divided into a plurality of fall video segments according to the time sequence corresponding to the fall time of the sample (the sequence from the beginning of falling to the final falling), for example, the sample fall video V is divided into V1、V2And V3Three fall video clips, wherein after the fall video clips are obtained, the start frame image of each fall video clip and the end frame image of the last fall video clip are captured, for example, when a sample fall video V is divided into V1、V2And V3After three fall video clips, V is intercepted1、V2And V3Start frame image P1、P2And P3And the last fall recording V3End frame image P of3', start frame picture P1、P2、P3And an end frame image P3' A sample image queue can be constructed. In addition, it should be noted that the step of capturing the start frame and the end point image of the fall video segment is to decompose the whole fall process of the human body, so as to obtain human body states of the human body at different stages in the fall process, wherein the human body states include a human body activity state and a fall tendency state.
Step S203: processing each fall image in the sample fall image queue to obtain a target sample fall image queue.
In a specific implementation, denoising and classifying processing needs to be performed on each sample fall image in the sample fall image queue to obtain a target sample fall image queue, where the step S203 in this embodiment includes: denoising each fallen image in the sample fallen image queue according to the human motion characteristics to obtain a plurality of human fallen images; classifying the plurality of human body falling images in sequence according to the time sequence corresponding to the sample falling time; and setting corresponding category identification for each classified human body falling image so as to obtain a target sample falling image queue.
It should be noted that there may be other moving objects in the fall image that affect fall detection, and therefore, it is necessary to denoise each fall image according to the motion characteristics of the human body, so as to obtain a human body fall image only including the human body. Then, the human body falling images are classified in sequence according to the time sequence of the falling time of the sample, each human body falling image corresponds to one human body state, a corresponding class identifier is set for each classified human body falling image, for example, the class identifier of the human body falling image X is set to be 1, the class identifier of the human body falling image Y is set to be 2, and a target sample falling image queue can be formed by completing denoising and classifying a plurality of human body falling images.
Step S204: training the target sample fall image queue to obtain a fall detection model.
In this embodiment, training the target sample fall image queue may input the de-noised and classified target sample fall image queue into a preset neural network model, so as to obtain a fall detection model, where the preset neural network model includes a BP neural network, a Hopfield network, an ART network, a Kohonen network, and the like.
In the embodiment, the sample falling video of the human body to be detected and the sample falling time corresponding to the sample falling video are obtained; generating a sample fall image queue according to the sample fall time and the sample fall video; processing each falling image in the sample falling image queue to obtain a target sample falling image queue; training the target sample falling image queue to obtain a falling detection model, denoising the sample falling image queue to obtain a human body falling image, classifying the human body falling image to obtain a target sample falling image queue, and training based on the target sample image queue to obtain the falling detection model, so that the pre-constructed falling detection model is more accurate, and the accuracy of human body falling detection is further improved.
Furthermore, an embodiment of the present invention also provides a storage medium having a human body fall detection program stored thereon, which when executed by a processor implements the steps of the human body fall detection method as described above.
Referring to fig. 6, fig. 6 is a block diagram of the first embodiment of the human fall detection apparatus of the present invention.
As shown in fig. 6, a human fall detection apparatus according to an embodiment of the present invention includes:
the acquisition module 10 is configured to shoot a human body to be detected, so as to obtain a human body activity image set of the human body to be detected.
In this embodiment, human detection device that tumbles can gather the motion image of object and carry out analysis processes to the object motion image, human detection device that tumbles is equipped with the camera, wait to detect through the camera and detect the human body and shoot, the camera includes analog camera, digital camera, high definition digtal camera, charge-coupled device CCD camera and spherical camera etc. transmission signal, resolution ratio, sensor model and the appearance to the camera in this embodiment all do not put the restriction, can adopt suitable camera according to actual conditions. In addition, the camera can collect human body moving images of a human body to be detected at a speed of 24 frames per second, and can also collect the human body moving images at other speeds.
It should be noted that, when the camera shoots the human body to be detected, the acquired image is actually the region image of the motion region where the human body to be detected is located, and the region image may have other moving objects, so the human body moving image is obtained by substantially processing the region image.
And the classification module 20 is configured to classify motion states of the human body moving images in the human body moving image set based on a pre-constructed fall detection model, so as to obtain a motion state type corresponding to each human body moving image.
It should be noted that each human body moving image in the human body moving image set contains a motion state of a human body to be detected, different motion states all have corresponding motion state types, the motion state types substantially reflect different corresponding motion states in a human body falling process, the whole process from standing to falling of the human body is divided into a plurality of different stages, and a corresponding motion state type is defined for the motion state of the human body in each stage, for example, the whole falling process of the human body is divided into a standing state, a falling state a, a falling state B, a falling state C and a falling state, the motion state type corresponding to the standing state is defined as 1, the motion state type corresponding to the falling state a is defined as 2, the motion state type corresponding to the falling state B is defined as 3, the motion state type corresponding to the falling state C is defined as 4, and the motion state type corresponding to the falling state is defined as 5, the motion state type corresponding to the motion state in this embodiment may be set according to an actual situation, and is not limited in this embodiment.
In specific implementation, the motion state classification is carried out on each human body moving image through a falling detection model to obtain the motion state type corresponding to each human body moving image, and the falling detection model is constructed in advance based on a sample image of a human body to be detected.
And the judging module 30 is used for carrying out falling detection on the human body to be detected according to the motion state type.
It is easy to understand that after the motion state types corresponding to the human body moving images are obtained, the falling detection of the human body to be detected can be performed according to the motion state types corresponding to the human body moving images, for example, if the motion state type corresponding to the standing state is 1, it can be determined that the human body to be detected has not a falling event according to the motion state type 1, and if the motion state type corresponding to the falling state is 2, it can be determined that the human body to be detected has a falling event according to the motion state type 2.
In the embodiment, a human body to be detected is shot to obtain a human body activity image set of the human body to be detected; classifying the motion state of each human body moving image in the human body moving image set based on a pre-constructed falling detection model to obtain the motion state type corresponding to each human body moving image; the human body to be detected is subjected to falling detection according to the motion state type, the motion state classification is carried out on the human body activity image set of the human body to be detected through a pre-constructed falling detection model, the falling detection is carried out on the human body to be detected according to the motion state type corresponding to each human body activity image, the falling detection can be accurately carried out on the human body according to a plurality of human body activity images with different motion state types, and therefore the accuracy of the falling detection of the human body is improved.
In an embodiment, the determining module 30 is further configured to screen out a motion state type meeting a first preset type from the motion state types; acquiring the type number corresponding to the motion state type conforming to the first preset type; determining falling confidence corresponding to the human body to be detected according to the type number; and carrying out falling detection on the human body to be detected according to the falling confidence coefficient.
In an embodiment, the determining module 30 is further configured to compare the fall confidence level with a preset confidence level; if the falling confidence coefficient does not exceed the preset confidence coefficient, judging that the human body to be detected does not have falling risk; and if the falling confidence coefficient exceeds the preset confidence coefficient, judging that the human body to be detected has falling risk.
In an embodiment, the determining module 30 is further configured to obtain a fall tendency image set of the human body to be detected within a preset time when the human body to be detected has a fall risk; classifying falling tendency states of the falling tendency images in the falling tendency image set based on the pre-constructed falling detection model so as to obtain falling tendency state types corresponding to the falling tendency images; screening fall tendency state types which accord with a second preset type from the fall tendency state types; acquiring the type number corresponding to the falling tendency state type in accordance with the second preset type; and when the type number reaches a number threshold value, judging that the human body to be detected has a falling event, and outputting a falling alarm prompt.
In an embodiment, the human fall detection apparatus further comprises: building a module;
the construction module is used for acquiring a sample falling video of the human body to be detected and sample falling time corresponding to the sample falling video; generating a sample fall image queue according to the sample fall time and the sample fall video; processing each falling image in the sample falling image queue to obtain a target sample falling image queue; training the target sample fall image queue to obtain a fall detection model.
The construction module is further configured to sequentially divide the sample fall video into a plurality of fall video segments according to a time sequence corresponding to the sample fall time; intercepting the initial frame image of each falling video clip and the final frame image of the last falling video clip; and generating a sample falling image queue according to the initial frame image and the final frame image of each falling video clip.
The construction module is further used for denoising each falling image in the sample falling image queue according to the human motion characteristics to obtain a plurality of human falling images; classifying the plurality of human body falling images in sequence according to the time sequence corresponding to the sample falling time; and setting corresponding category identification for each classified human body falling image so as to obtain a target sample falling image.
It should be understood that the above is only an example, and the technical solution of the present invention is not limited in any way, and in a specific application, a person skilled in the art may set the technical solution as needed, and the present invention is not limited thereto.
It should be noted that the above-described work flows are only exemplary, and do not limit the scope of the present invention, and in practical applications, a person skilled in the art may select some or all of them to achieve the purpose of the solution of the embodiment according to actual needs, and the present invention is not limited herein.
In addition, the technical details that are not described in detail in this embodiment can be referred to the human body fall detection method provided in any embodiment of the present invention, and are not described herein again.
Further, it is to be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention or portions thereof that contribute to the prior art may be embodied in the form of a software product, where the computer software product is stored in a storage medium (e.g. Read Only Memory (ROM)/RAM, magnetic disk, optical disk), and includes several instructions for enabling a terminal device (e.g. a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A human fall detection method, characterized by comprising:
shooting a human body to be detected to obtain a human body activity image set of the human body to be detected;
classifying the motion state of each human body moving image in the human body moving image set based on a pre-constructed falling detection model to obtain the motion state type corresponding to each human body moving image;
and carrying out falling detection on the human body to be detected according to the motion state type.
2. A personal fall detection method according to claim 1, wherein said step of detecting a fall of the person to be detected based on the type of motion state comprises:
screening out a motion state type which accords with a first preset type from the motion state types;
acquiring the type number corresponding to the motion state type conforming to the first preset type;
determining falling confidence corresponding to the human body to be detected according to the type number;
and carrying out falling detection on the human body to be detected according to the falling confidence coefficient.
3. A personal fall detection method according to claim 2, wherein said step of detecting a fall of the person to be detected according to the fall confidence level comprises:
comparing the fall confidence level with a preset confidence level;
if the falling confidence coefficient does not exceed the preset confidence coefficient, judging that the human body to be detected does not have falling risk;
and if the falling confidence coefficient exceeds the preset confidence coefficient, judging that the human body to be detected has falling risk.
4. A method for detecting a human fall according to claim 3, wherein the step of determining that the human body to be detected has a fall risk if the fall confidence level exceeds the preset confidence level further comprises:
when the human body to be detected has a falling risk, acquiring a falling tendency image set of the human body to be detected within a preset time;
classifying falling tendency states of the falling tendency images in the falling tendency image set based on the pre-constructed falling detection model so as to obtain falling tendency state types corresponding to the falling tendency images;
screening fall tendency state types which accord with a second preset type from the fall tendency state types;
acquiring the type number corresponding to the falling tendency state type in accordance with the second preset type;
and when the type number reaches a number threshold value, judging that the human body to be detected has a falling event, and outputting a falling alarm prompt.
5. A human body fall detection method according to any one of claims 1 to 4, wherein, before the step of classifying the motion states of the respective human body moving images in the human body moving image set based on a previously constructed fall detection model to obtain the motion state types corresponding to the respective human body moving images, the method further comprises:
acquiring a sample falling video of the human body to be detected and sample falling time corresponding to the sample falling video;
generating a sample fall image queue according to the sample fall time and the sample fall video;
processing each falling image in the sample falling image queue to obtain a target sample falling image queue;
training the target sample fall image queue to obtain a fall detection model.
6. A personal fall detection method as claimed in claim 5, wherein the step of generating a sample fall image queue from the sample fall time and the sample fall video record comprises:
sequentially dividing the sample fall video into a plurality of fall video fragments according to the time sequence corresponding to the sample fall time;
intercepting the initial frame image of each falling video clip and the final frame image of the last falling video clip;
and generating a sample falling image queue according to the initial frame image and the final frame image of each falling video clip.
7. A personal fall detection method as claimed in claim 5, wherein the step of processing each fall image in the sample fall image queue to obtain a target sample fall image queue comprises:
denoising each fallen image in the sample fallen image queue according to the human motion characteristics to obtain a plurality of human fallen images;
classifying the plurality of human body falling images in sequence according to the time sequence corresponding to the sample falling time;
and setting corresponding category identification for each classified human body falling image so as to obtain a target sample falling image.
8. A human fall detection device, characterized by comprising:
the system comprises an acquisition module, a detection module and a display module, wherein the acquisition module is used for shooting a human body to be detected so as to obtain a human body activity image set of the human body to be detected;
the classification module is used for classifying the motion states of all human body moving images in the human body moving image set based on a pre-constructed falling detection model so as to obtain the motion state types corresponding to all human body moving images;
and the judging module is used for carrying out falling detection on the human body to be detected according to the motion state type.
9. A personal fall detection apparatus, comprising: a memory, a processor and a personal fall detection program stored on the memory and run on the processor, the personal fall detection program being configured to implement the steps of the personal fall detection method as claimed in any of claims 1 to 7.
10. A storage medium, characterized in that the storage medium has stored thereon a human fall detection program which, when executed by a processor, implements the steps of the human fall detection method according to any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010867487.9A CN111899470B (en) | 2020-08-26 | 2020-08-26 | Human body falling detection method, device, equipment and storage medium |
| PCT/CN2020/127346 WO2022041484A1 (en) | 2020-08-26 | 2020-11-07 | Human body fall detection method, apparatus and device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010867487.9A CN111899470B (en) | 2020-08-26 | 2020-08-26 | Human body falling detection method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111899470A true CN111899470A (en) | 2020-11-06 |
| CN111899470B CN111899470B (en) | 2022-07-22 |
Family
ID=73225616
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010867487.9A Active CN111899470B (en) | 2020-08-26 | 2020-08-26 | Human body falling detection method, device, equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111899470B (en) |
| WO (1) | WO2022041484A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112489368A (en) * | 2020-11-30 | 2021-03-12 | 安徽国广数字科技有限公司 | Intelligent falling identification and detection alarm method and system |
| WO2022041484A1 (en) * | 2020-08-26 | 2022-03-03 | 歌尔股份有限公司 | Human body fall detection method, apparatus and device, and storage medium |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117278837B (en) * | 2023-11-16 | 2024-01-26 | 新乡天辅电子科技有限公司 | Emergency rescue-oriented imaging equipment control method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955699A (en) * | 2014-03-31 | 2014-07-30 | 北京邮电大学 | Method for detecting tumble event in real time based on surveillance videos |
| US20170188895A1 (en) * | 2014-03-12 | 2017-07-06 | Smart Monitor Corp | System and method of body motion analytics recognition and alerting |
| CN109009145A (en) * | 2018-07-24 | 2018-12-18 | 西安工程大学 | A kind of tumble judgment method based on wearable device |
| CN110327050A (en) * | 2019-05-05 | 2019-10-15 | 北京理工大学 | People's tumble state embedded intelligence detection method for wearable equipment |
| CN110620905A (en) * | 2019-09-06 | 2019-12-27 | 平安医疗健康管理股份有限公司 | Video monitoring method and device, computer equipment and storage medium |
| CN110765860A (en) * | 2019-09-16 | 2020-02-07 | 平安科技（深圳）有限公司 | Tumble determination method, tumble determination device, computer apparatus, and storage medium |
| CN111052194A (en) * | 2017-05-25 | 2020-04-21 | 三星电子株式会社 | Method and system for detecting a hazardous situation |
| US20200211154A1 (en) * | 2018-12-30 | 2020-07-02 | Altumview Systems Inc. | Method and system for privacy-preserving fall detection |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105279483B (en) * | 2015-09-28 | 2018-08-21 | 华中科技大学 | A kind of tumble behavior real-time detection method based on depth image |
| US20170213081A1 (en) * | 2015-11-19 | 2017-07-27 | Intelli-Vision | Methods and systems for automatically and accurately detecting human bodies in videos and/or images |
| CN110490080B (en) * | 2019-07-22 | 2023-05-09 | 毕昇云(武汉)信息技术有限公司 | Human body falling judgment method based on image |
| CN111209848B (en) * | 2020-01-03 | 2023-07-21 | 北京工业大学 | Real-time falling detection method based on deep learning |
| CN111899470B (en) * | 2020-08-26 | 2022-07-22 | 歌尔科技有限公司 | Human body falling detection method, device, equipment and storage medium |
-
2020
- 2020-08-26 CN CN202010867487.9A patent/CN111899470B/en active Active
- 2020-11-07 WO PCT/CN2020/127346 patent/WO2022041484A1/en active Application Filing
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170188895A1 (en) * | 2014-03-12 | 2017-07-06 | Smart Monitor Corp | System and method of body motion analytics recognition and alerting |
| CN103955699A (en) * | 2014-03-31 | 2014-07-30 | 北京邮电大学 | Method for detecting tumble event in real time based on surveillance videos |
| CN111052194A (en) * | 2017-05-25 | 2020-04-21 | 三星电子株式会社 | Method and system for detecting a hazardous situation |
| CN109009145A (en) * | 2018-07-24 | 2018-12-18 | 西安工程大学 | A kind of tumble judgment method based on wearable device |
| US20200211154A1 (en) * | 2018-12-30 | 2020-07-02 | Altumview Systems Inc. | Method and system for privacy-preserving fall detection |
| US20200205697A1 (en) * | 2018-12-30 | 2020-07-02 | Altumview Systems Inc. | Video-based fall risk assessment system |
| CN111383421A (en) * | 2018-12-30 | 2020-07-07 | 奥瞳系统科技有限公司 | Privacy protection fall detection method and system |
| CN110327050A (en) * | 2019-05-05 | 2019-10-15 | 北京理工大学 | People's tumble state embedded intelligence detection method for wearable equipment |
| CN110620905A (en) * | 2019-09-06 | 2019-12-27 | 平安医疗健康管理股份有限公司 | Video monitoring method and device, computer equipment and storage medium |
| CN110765860A (en) * | 2019-09-16 | 2020-02-07 | 平安科技（深圳）有限公司 | Tumble determination method, tumble determination device, computer apparatus, and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 刘勇: "基于模式识别的跌倒检测仪研究", 《中国优秀硕士学位论文全文数据库（电子期刊） 工程科技Ⅱ辑》 * |
| 范凯波: "基于视频的跌倒人体目标识别研究", 《中国优秀博硕士学位论文全文数据库（博士）社会科学Ⅱ辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022041484A1 (en) * | 2020-08-26 | 2022-03-03 | 歌尔股份有限公司 | Human body fall detection method, apparatus and device, and storage medium |
| CN112489368A (en) * | 2020-11-30 | 2021-03-12 | 安徽国广数字科技有限公司 | Intelligent falling identification and detection alarm method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022041484A1 (en) | 2022-03-03 |
| CN111899470B (en) | 2022-07-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111899470B (en) | Human body falling detection method, device, equipment and storage medium | |
| CN111178183B (en) | Face detection method and related device | |
| EP1936576B1 (en) | Camera tampering detection | |
| CN109815787B (en) | Target identification method and device, storage medium and electronic equipment | |
| CN111354024B (en) | Behavior prediction method of key target, AI server and storage medium | |
| CN112183166A (en) | Method and device for determining training sample and electronic equipment | |
| CN111654700A (en) | Privacy mask processing method and device, electronic equipment and monitoring system | |
| CN111898581A (en) | Animal detection method, device, electronic equipment and readable storage medium | |
| CN111814776B (en) | Image processing method, device, server and storage medium | |
| CN112257643A (en) | Smoking behavior and calling behavior identification method based on video streaming | |
| KR102511287B1 (en) | Image-based pose estimation and action detection method and appratus | |
| CN111401239B (en) | Video analysis method, device, system, equipment and storage medium | |
| CN111444555B (en) | Temperature measurement information display method and device and terminal equipment | |
| CN110505438B (en) | Queuing data acquisition method and camera | |
| CN111523490A (en) | Mask wearing detection method, device, equipment and readable storage medium | |
| JP2007018324A (en) | Apparatus, method and program for detecting intruding object by image processing | |
| CN112333537B (en) | Video integration method, device and computer readable storage medium | |
| CN114764895A (en) | Abnormal behavior detection device and method | |
| US20200311401A1 (en) | Analyzing apparatus, control method, and program | |
| CN110795980A (en) | Network video-based evasion identification method, equipment, storage medium and device | |
| CN111414886A (en) | Intelligent recognition system for human body dynamic characteristics | |
| US20220122377A1 (en) | Information processing apparatus, information processing method, and program | |
| CN115346060A (en) | Picture abnormity identification method and device, electronic equipment and storage medium | |
| WO2014038924A2 (en) | A method for producing a background model | |
| CN112149451B (en) | Affinity analysis method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |