CN113591712A - AI identification-based user abnormal behavior identification method and device - Google Patents
AI identification-based user abnormal behavior identification method and device Download PDFInfo
- Publication number
- CN113591712A CN113591712A CN202110874718.3A CN202110874718A CN113591712A CN 113591712 A CN113591712 A CN 113591712A CN 202110874718 A CN202110874718 A CN 202110874718A CN 113591712 A CN113591712 A CN 113591712A
- Authority
- CN
- China
- Prior art keywords
- user
- vector
- body posture
- real
- posture description
- 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.)
- Pending
Links
- 206010000117 Abnormal behaviour Diseases 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 title claims abstract description 28
- 239000013598 vector Substances 0.000 claims abstract description 226
- 230000006399 behavior Effects 0.000 claims description 105
- 238000004422 calculation algorithm Methods 0.000 claims description 17
- 238000004364 calculation method Methods 0.000 claims description 13
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 13
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 4
- 238000003860 storage Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 2
- 210000004394 hip joint Anatomy 0.000 description 2
- 210000003127 knee Anatomy 0.000 description 2
- 210000003141 lower extremity Anatomy 0.000 description 2
- 210000001364 upper extremity Anatomy 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000391 smoking effect Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a user abnormal behavior identification method and device based on AI identification, comprising the following steps: calculating a structure vector included angle of a user real-time behavior joint point and a user real-time behavior joint point model ratio according to the user real-time behavior joint point vector, and then acquiring a user human body posture description vector set; the abnormal behavior of the user is identified by using the human body posture description vector of the user, the real-time behavior data of the user is analyzed, the real-time state of the user is judged, the key behavior characteristics are extracted, and the speed and the accuracy of identification are improved.
Description
Technical Field
The invention relates to the technical field of intelligent retail, in particular to a user abnormal behavior identification method and device based on AI identification.
Background
In the field of unattended intelligent retail, reducing goods loss and preventing theft are important issues considered by merchants. In an unattended scene, due to the fact that a commodity detection channel exists, a commodity is not easy to take out of a shop without consumption, articles in the shop are directly eaten and drunk in the shop, then the articles are easy to take out of the shop, and in the unattended shop, smoking and open fire are forbidden generally, so that whether a user smokes or not needs to be judged, and in addition, the inside of the shop is unattended. Only one person under the majority condition when the user uses, when the user falls down because of illness, the trade company also need to be reported an emergency and asked for help or increased vigilance, handles rapidly, prevents the accident, so has had the demand that discernment user eaten and drinks, smokes, falls down.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a user abnormal behavior identification method based on AI identification, which comprises the following steps:
acquiring a user human body posture description vector set according to the user real-time behavior joint point vector;
and identifying abnormal behaviors of the user by utilizing the user human body posture description vector set.
Preferably, the acquiring the user body posture description vector set according to the user real-time behavior joint point vector includes:
acquiring a user real-time behavior joint point structure vector according to the user real-time behavior joint point vector;
calculating the included angle of the structure vector of the user real-time behavior joint point according to the structure vector of the user real-time behavior joint point;
calculating a user real-time behavior joint point module ratio by using the user real-time behavior joint point vector;
and acquiring a user human body posture description vector set according to the included angle of the user real-time behavior joint point structure vector and the user real-time behavior joint point model ratio.
Further, the calculation formula for calculating the structure vector included angle of the user real-time behavior joint point according to the structure vector of the user real-time behavior joint point is as follows:
And p is an included angle of the structure vector of the user real-time behavior joint point, a is the structure vector of the user real-time behavior joint point, and b is the structure vector of the user real-time behavior joint point corresponding to a.
Further, the calculation formula for calculating the real-time behavior joint point model ratio by using the user real-time behavior joint point vector is as follows:
and r is a user real-time behavior joint point module ratio, a is a user real-time behavior joint point structure vector, and b is a user real-time behavior joint point structure vector corresponding to a.
Further, the acquiring a set of user body posture description vectors according to the included angle of the user real-time behavior joint point structure vector and the user real-time behavior joint point model ratio includes:
G＝(p1，p2…pn，r1，r2…rm)
wherein G is a user body posture description vector, p is a user real-time behavior joint point structure vector included angle, and r is a user real-time behavior joint point module ratio.
Preferably, the identifying abnormal user behavior by using the set of user body posture describing vectors includes:
and identifying abnormal behaviors of the user based on a DTW algorithm according to the user human body posture description vector set.
Further, the identifying abnormal behaviors of the user based on the DTW algorithm according to the user body posture description vector set comprises:
calculating the distortion quantity of the user human posture description vector based on a DTW algorithm by using the user human posture description vector set and a preset abnormal behavior user human posture description vector reference set;
calculating the accumulated distortion quantity of the user human body posture description vector by using the distortion quantity of the user human body posture description vector;
and identifying abnormal behaviors of the user according to the accumulated distortion quantity of the human body posture description vector of the user.
Further, the calculation formula for calculating the distortion quantity of the user body posture description vector based on the DTW algorithm by using the user body posture description vector set and the preset abnormal behavior user body posture description vector reference set is as follows:
D is a user human body posture description vector distortion quantity, T is a user human body posture description vector set, R is a preset abnormal behavior user human body posture description vector reference set, n is data in the user human body posture description vector set, i is the number of data in the user human body posture description vector set, m is data in the preset abnormal behavior user human body posture description vector reference set, and j is the number of data in the preset abnormal behavior user human body posture description vector reference set.
Further, the calculation formula for calculating the accumulated distortion quantity of the user body posture description vector by using the distortion quantity of the user body posture description vector is as follows:
H[T(ni)，R(mi)]＝d[T(ni)，R(mi)]+D[T(ni-1)，R(mi-1)]
h is the accumulated distortion quantity of the user body posture description vector, D is the distortion quantity of the user body posture description vector, T is a user body posture description vector set, R is a preset abnormal behavior user body posture description vector reference set, n is data in the user body posture description vector set, i is the quantity of data in the user body posture description vector set, and m is the data in the preset abnormal behavior user body posture description vector reference set.
Based on the same inventive concept, the invention also provides a user abnormal behavior recognition device based on AI recognition, which is characterized by comprising:
the acquisition module is used for acquiring a user human body posture description vector set according to the user real-time behavior joint point vector;
and the recognition module is used for recognizing the abnormal behaviors of the user by utilizing the user human body posture description vector set.
Compared with the closest prior art, the invention has the following beneficial effects:
acquiring a human body posture description vector of the user according to the real-time behavior joint point vector of the user; the abnormal behavior of the user is identified by utilizing the human body posture description vector of the user, the invention provides an identification method based on AI aiming at the problem of accurate identification of the user behavior in the unattended shop, and can effectively solve the identification problem of the abnormal behavior of the human body in the shop.
Drawings
FIG. 1 is a flow chart of a method for identifying abnormal behaviors of a user based on AI identification according to the present invention;
FIG. 2 is a schematic diagram of Kinect V2 according to the present invention;
FIG. 3 is a schematic diagram of a three-dimensional coordinate system of Kinect according to the present invention;
FIG. 4 is a schematic diagram of a distribution of human joint points according to the present invention;
FIG. 5 is a flow chart of behavior representation feature extraction provided by the present invention;
FIG. 6 is a schematic view of a human body structure vector of an upper limb portion according to the present invention;
FIG. 7 is a schematic view of angle information of a lower limb portion according to the present invention;
FIG. 8 is a schematic view of angle information at an intermediate connection provided by the present invention;
FIG. 9 is a schematic diagram of a vector distribution for obtaining modulus ratio information according to the present invention;
fig. 10 is a schematic diagram of an AI recognition-based user abnormal behavior recognition apparatus according to the present invention.
Detailed Description
The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1:
the invention provides a user abnormal behavior identification method based on AI identification, as shown in figure 1, comprising the following steps:
step 1: acquiring a user human body posture description vector set according to the user real-time behavior joint point vector;
step 2: and identifying abnormal behaviors of the user by utilizing the user human body posture description vector set.
In this embodiment, a Kinect V2 used in the AI identification based user abnormal behavior identification method is schematically shown in fig. 2.
In this embodiment, a three-dimensional spatial coordinate system of a Kinect related to an AI identification-based user abnormal behavior identification method is schematically shown in fig. 3.
The step 1 specifically comprises the following steps:
1-1: acquiring a user real-time behavior joint point structure vector according to the user real-time behavior joint point vector;
1-2: calculating the included angle of the structure vector of the user real-time behavior joint point according to the structure vector of the user real-time behavior joint point;
1-3: calculating a user real-time behavior joint point module ratio by using the user real-time behavior joint point vector;
1-4: and acquiring a user human body posture description vector set according to the included angle of the user real-time behavior joint point structure vector and the user real-time behavior joint point model ratio.
The step 1-1 specifically comprises:
preprocessing the joint point data, and aiming at the human body structure characteristics, as shown in fig. 4, after constructing a human body structure vector by using the joint point data, calculating a joint point structure vector of the real-time behavior of the user.
In this embodiment, in the AI-recognition-based user abnormal behavior recognition method, the joint points are named head, neck, left hand lhand, left elbow lebow, left shoulder lshoulder, right hand rhand, right elbow relaow, right shoulder rshoulder, center node torso, left hip joint lhip, left knee lknee, left foot lfoot, right hip joint rhip, right knee rknee, and right foot rfoot.
In this embodiment, as shown in fig. 5, a method for identifying abnormal behaviors of a user based on AI identification obtains real-time behavior data of the user in a store, where a point a (x1, y1, z1) is a coordinate of a real-time right shoulder of the user, a point B (x2, y2, z2) is a coordinate of a real-time right elbow of the user, and a corresponding structure vector is as follows:
the step 1-2 specifically comprises:
the calculation formula for calculating the structure vector included angle of the user real-time behavior joint point is as follows:
And p is an included angle of the structure vector of the user real-time behavior joint point, a is the structure vector of the user real-time behavior joint point, and b is the structure vector of the user real-time behavior joint point corresponding to a.
In this embodiment, a method for identifying abnormal behaviors of a user based on AI identification classifies human body structure vectors of the user into 10 groups of upper limb angle information as shown in fig. 6, 4 groups of lower limb angle information as shown in fig. 7, and 6 groups of intermediate joint angle information as shown in fig. 8.
The steps 1-3 specifically include:
the calculation formula for calculating the real-time behavior joint point model ratio of the user is as follows:
and r is a user real-time behavior joint point module ratio, a is a user real-time behavior joint point structure vector, and b is a user real-time behavior joint point structure vector corresponding to a.
In this embodiment, as shown in fig. 9, a is a vector of a center joint pointing to a head, b and c are vectors of a head pointing to a left hand and a right hand, and d and e are vectors of a center joint pointing to a left hand and a right hand, and the calculated vector modulo ratios are respectively:
the vector modulus ratio is introduced to be a further supplement to behavior identification and assist in accurate identification of abnormal behaviors.
The steps 1-4 specifically include:
acquiring a user body posture description vector set according to the user real-time behavior joint point structure vector included angle and the user real-time behavior joint point model ratio, wherein the acquisition comprises the following steps:
G＝(p1，p2…pn，r1，r2…rm)
wherein G is a user body posture description vector, p is a user real-time behavior joint point structure vector included angle, and r is a user real-time behavior joint point module ratio.
In this embodiment, in the AI identification-based method for identifying abnormal behaviors of a user, a value of n is 20, and a value of m is 4.
In this embodiment, a user abnormal behavior identification method based on AI identification sets pi as a certain angle value or a modulus value, as shown in the following table:
the step 2 specifically comprises the following steps:
2-1: and identifying abnormal behaviors of the user based on a DTW algorithm according to the user human body posture description vector set.
The step 2-1 specifically comprises:
2-1-1: calculating the distortion quantity of the user human posture description vector based on a DTW algorithm by using the user human posture description vector set and a preset abnormal behavior user human posture description vector reference set;
2-1-2: calculating the accumulated distortion quantity of the user human body posture description vector by using the distortion quantity of the user human body posture description vector;
2-1-3: and identifying abnormal behaviors of the user according to the accumulated distortion quantity of the human body posture description vector of the user.
The step 2-1-1 specifically comprises:
the calculation formula for calculating the distortion quantity of the user human body posture description vector based on the DTW algorithm is as follows:
D is a user human body posture description vector distortion quantity, T is a user human body posture description vector set, R is a preset abnormal behavior user human body posture description vector reference set, n is data in the user human body posture description vector set, i is the number of data in the user human body posture description vector set, m is data in the preset abnormal behavior user human body posture description vector reference set, and j is the number of data in the preset abnormal behavior user human body posture description vector reference set.
The step 2-1-2 specifically comprises:
the calculation formula for calculating the accumulated distortion quantity of the human body posture description vector of the user is as follows:
H[T(ni)，R(mi)]＝d[T(ni)，R(mi)]+D[T(ni-1)，R(mi-1)]
h is the accumulated distortion quantity of the user body posture description vector, D is the distortion quantity of the user body posture description vector, T is a user body posture description vector set, R is a preset abnormal behavior user body posture description vector reference set, n is data in the user body posture description vector set, i is the quantity of data in the user body posture description vector set, and m is the data in the preset abnormal behavior user body posture description vector reference set.
The steps 2-1-3 specifically comprise:
and according to the user human body posture description vector and the preset abnormal behavior, the user with the minimum accumulated distortion quantity is used as the abnormal behavior of the user.
Example 2:
based on the unified inventive concept, the present invention further provides an AI-recognition-based abnormal user behavior recognition apparatus, as shown in fig. 10, including:
the acquisition module is used for acquiring a user human body posture description vector set according to the user real-time behavior joint point vector;
and the recognition module is used for recognizing the abnormal behaviors of the user by utilizing the user human body posture description vector set.
In an embodiment of the present invention, the acquiring a set of user body posture description vectors according to a user real-time behavior joint point vector includes:
acquiring a user real-time behavior joint point structure vector according to the user real-time behavior joint point vector;
calculating the included angle of the structure vector of the user real-time behavior joint point according to the structure vector of the user real-time behavior joint point;
calculating a user real-time behavior joint point module ratio by using the user real-time behavior joint point vector;
and acquiring a user human body posture description vector set according to the included angle of the user real-time behavior joint point structure vector and the user real-time behavior joint point model ratio.
In the preferred embodiment of the present invention, the calculation formula for calculating the structure vector included angle of the user real-time behavior joint point according to the structure vector of the user real-time behavior joint point is as follows:
And p is an included angle of the structure vector of the user real-time behavior joint point, a is the structure vector of the user real-time behavior joint point, and b is the structure vector of the user real-time behavior joint point corresponding to a.
In a preferred embodiment of the present invention, the method of claim 2, wherein the calculation formula for calculating the real-time behavior joint node model ratio value of the user by using the real-time behavior joint node vector of the user is as follows:
and r is a user real-time behavior joint point module ratio, a is a user real-time behavior joint point structure vector, and b is a user real-time behavior joint point structure vector corresponding to a.
In the preferred embodiment of the present invention, the acquiring a set of user body posture description vectors according to the user real-time behavior joint point structure vector included angle and the user real-time behavior joint point model ratio includes:
G＝(p1，p2…pn，r1，r2…rm)
wherein G is a user body posture description vector, p is a user real-time behavior joint point structure vector included angle, and r is a user real-time behavior joint point module ratio.
In an embodiment of the present invention, the recognizing the abnormal behavior of the user by using the user body posture description vector set includes:
and identifying abnormal behaviors of the user based on a DTW algorithm according to the user human body posture description vector set.
In an embodiment of the present invention, the identifying, according to the set of user body posture description vectors, the abnormal behavior of the user based on the DTW algorithm includes:
calculating the distortion quantity of the user human posture description vector based on a DTW algorithm by using the user human posture description vector set and a preset abnormal behavior user human posture description vector reference set;
calculating the accumulated distortion quantity of the user human body posture description vector by using the distortion quantity of the user human body posture description vector;
and identifying abnormal behaviors of the user according to the accumulated distortion quantity of the human body posture description vector of the user.
In the preferred embodiment of the present invention, the identifying the abnormal behavior of the user based on the DTW algorithm according to the user body posture description vector set includes that the calculation formula for calculating the distortion of the user body posture description vector based on the DTW algorithm by using the user body posture description vector set and the preset abnormal behavior user body posture description vector reference set is as follows:
D is a user human body posture description vector distortion quantity, T is a user human body posture description vector set, R is a preset abnormal behavior user human body posture description vector reference set, N is data in the user human body posture description vector set, i is the number of data in the user human body posture description vector set, M is data in the preset abnormal behavior user human body posture description vector reference set, j is the number of data in the preset abnormal behavior user human body posture description vector reference set, and M is the number of data in the N.
In an embodiment of the present invention, the identifying the abnormal behavior of the user based on the DTW algorithm according to the set of user body posture description vectors includes calculating an accumulated distortion amount of the user body posture description vectors by using the distortion amount of the user body posture description vectors according to the following formula:
H[T(ni)，R(mi)]＝d[T(ni)，R(mi)]+D[T(ni-1)，R(mi-1)]
h is the accumulated distortion quantity of the user body posture description vector, D is the distortion quantity of the user body posture description vector, T is a user body posture description vector set, R is a preset abnormal behavior user body posture description vector reference set, n is data in the user body posture description vector set, i is the quantity of data in the user body posture description vector set, and m is the data in the preset abnormal behavior user body posture description vector reference set.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (10)
1. A user abnormal behavior identification method based on AI identification is characterized by comprising the following steps:
acquiring a user human body posture description vector set according to the user real-time behavior joint point vector;
and identifying abnormal behaviors of the user by utilizing the user human body posture description vector set.
2. The method of claim 1, wherein the obtaining a set of user body posture description vectors according to the user real-time behavior joint point vector comprises:
acquiring a user real-time behavior joint point structure vector according to the user real-time behavior joint point vector;
calculating the included angle of the structure vector of the user real-time behavior joint point according to the structure vector of the user real-time behavior joint point;
calculating a user real-time behavior joint point module ratio by using the user real-time behavior joint point vector;
and acquiring a user human body posture description vector set according to the included angle of the user real-time behavior joint point structure vector and the user real-time behavior joint point model ratio.
3. The method of claim 2, wherein the calculation of the user real-time behavior joint structure vector angle from the user real-time behavior joint structure vector is performed as follows:
and p is an included angle of the structure vector of the user real-time behavior joint point, a is the structure vector of the user real-time behavior joint point, and b is the structure vector of the user real-time behavior joint point corresponding to a.
4. The method of claim 2, wherein the calculating the real-time behavior joint mode ratio value of the user using the real-time behavior joint vector of the user is performed as follows:
and r is a user real-time behavior joint point module ratio, a is a user real-time behavior joint point structure vector, and b is a user real-time behavior joint point structure vector corresponding to a.
5. The method of claim 2, wherein the obtaining the set of user body posture description vectors according to the user real-time behavior joint point structure vector included angle and the user real-time behavior joint point model ratio comprises:
G＝(p1，p2…pn，r1，r2…rm)
wherein G is a user body posture description vector, p is a user real-time behavior joint point structure vector included angle, and r is a user real-time behavior joint point module ratio.
6. The method of claim 1, wherein the identifying abnormal user behavior using the set of user body gesture description vectors comprises:
and identifying abnormal behaviors of the user based on a DTW algorithm according to the user human body posture description vector set.
7. The method of claim 6, wherein the identifying abnormal user behavior based on the DTW algorithm according to the set of user body posture describing vectors comprises:
calculating the distortion quantity of the user human posture description vector based on a DTW algorithm by using the user human posture description vector set and a preset abnormal behavior user human posture description vector reference set;
calculating the accumulated distortion quantity of the user human body posture description vector by using the distortion quantity of the user human body posture description vector;
and identifying abnormal behaviors of the user according to the accumulated distortion quantity of the human body posture description vector of the user.
8. The method of claim 7, wherein the calculation formula for calculating the distortion quantity of the user body posture description vector based on the DTW algorithm by using the user body posture description vector set and the preset abnormal behavior user body posture description vector reference set is as follows:
D is a user human body posture description vector distortion quantity, T is a user human body posture description vector set, R is a preset abnormal behavior user human body posture description vector reference set, N is data in the user human body posture description vector set, i is the number of data in the user human body posture description vector set, M is data in the preset abnormal behavior user human body posture description vector reference set, j is the number of data in the preset abnormal behavior user human body posture description vector reference set, and M is the number of data in the N.
9. The method of claim 7, wherein the calculating the accumulated distortion amount of the user body posture description vector by using the distortion amount of the user body posture description vector is as follows:
H[T(ni)，R(mi)]＝d[T(ni)，R(mi)]+D[T(ni-1)，R(mi-1)]
h is the accumulated distortion quantity of the user body posture description vector, D is the distortion quantity of the user body posture description vector, T is a user body posture description vector set, R is a preset abnormal behavior user body posture description vector reference set, n is data in the user body posture description vector set, i is the quantity of data in the user body posture description vector set, and m is the data in the preset abnormal behavior user body posture description vector reference set.
10. An AI-recognition-based user abnormal behavior recognition apparatus, comprising:
the acquisition module is used for acquiring a user human body posture description vector set according to the user real-time behavior joint point vector;
and the recognition module is used for recognizing the abnormal behaviors of the user by utilizing the user human body posture description vector set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110874718.3A CN113591712A (en) | 2021-07-30 | 2021-07-30 | AI identification-based user abnormal behavior identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110874718.3A CN113591712A (en) | 2021-07-30 | 2021-07-30 | AI identification-based user abnormal behavior identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113591712A true CN113591712A (en) | 2021-11-02 |
Family
ID=78252941
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110874718.3A Pending CN113591712A (en) | 2021-07-30 | 2021-07-30 | AI identification-based user abnormal behavior identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113591712A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127125A (en) * | 2016-06-16 | 2016-11-16 | 山东大学 | Distributed DTW human body behavior intension recognizing method based on human body behavior characteristics |
| CN107080940A (en) * | 2017-03-07 | 2017-08-22 | 中国农业大学 | Body feeling interaction conversion method and device based on depth camera Kinect |
| CN109325469A (en) * | 2018-10-23 | 2019-02-12 | 北京工商大学 | A kind of human posture recognition method based on deep neural network |
| CN109460702A (en) * | 2018-09-14 | 2019-03-12 | 华南理工大学 | Passenger's abnormal behaviour recognition methods based on human skeleton sequence |
| KR101988037B1 (en) * | 2018-05-24 | 2019-06-11 | 변재형 | Method for providing sign language regognition service for communication between disability and ability |
| US20210103742A1 (en) * | 2019-10-08 | 2021-04-08 | Toyota Research Institute, Inc. | Spatiotemporal relationship reasoning for pedestrian intent prediction |
-
2021
- 2021-07-30 CN CN202110874718.3A patent/CN113591712A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127125A (en) * | 2016-06-16 | 2016-11-16 | 山东大学 | Distributed DTW human body behavior intension recognizing method based on human body behavior characteristics |
| CN107080940A (en) * | 2017-03-07 | 2017-08-22 | 中国农业大学 | Body feeling interaction conversion method and device based on depth camera Kinect |
| KR101988037B1 (en) * | 2018-05-24 | 2019-06-11 | 변재형 | Method for providing sign language regognition service for communication between disability and ability |
| CN109460702A (en) * | 2018-09-14 | 2019-03-12 | 华南理工大学 | Passenger's abnormal behaviour recognition methods based on human skeleton sequence |
| CN109325469A (en) * | 2018-10-23 | 2019-02-12 | 北京工商大学 | A kind of human posture recognition method based on deep neural network |
| US20210103742A1 (en) * | 2019-10-08 | 2021-04-08 | Toyota Research Institute, Inc. | Spatiotemporal relationship reasoning for pedestrian intent prediction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10810633B2 (en) | Generating a shoppable video | |
| CN102737235B (en) | Head posture estimation method based on depth information and color image | |
| CN111507317B (en) | Vision-based rotary equipment operation glove wearing detection method and system | |
| CN107908288A (en) | A kind of quick human motion recognition method towards human-computer interaction | |
| CN106779541A (en) | A kind of warehouse management method and system based on AR technologies | |
| JP2004054376A (en) | Method and device for estimating group attribute | |
| CN113449570A (en) | Image processing method and device | |
| CN112287865B (en) | Human body posture recognition method and device | |
| CN108960192A (en) | Action identification method and its neural network generation method, device and electronic equipment | |
| CN112215167B (en) | Intelligent store control method and system based on image recognition | |
| CN104751112B (en) | A kind of fingerprint template and fingerprint identification method based on fuzzy characteristics point information | |
| CN111445426B (en) | Target clothing image processing method based on generation of countermeasure network model | |
| CN114898249A (en) | Method, system and storage medium for confirming number of articles in shopping cart | |
| CN106897751A (en) | Human assistance method, device, system and host computer based on AR technologies | |
| Wang et al. | Gesture recognition by using kinect skeleton tracking system | |
| CN116311497A (en) | Tunnel worker abnormal behavior detection method and system based on machine vision | |
| Moutsis et al. | Fall detection paradigm for embedded devices based on YOLOv8 | |
| CN113591712A (en) | AI identification-based user abnormal behavior identification method and device | |
| CN102867112A (en) | Human posture tracking method applied to three-dimensional simulation fitting based on Kinect | |
| Chaudhary et al. | A vision-based method to find fingertips in a closed hand | |
| CN112395999A (en) | Wearing standard judging method based on image recognition and related equipment | |
| CN111339973A (en) | Object identification method, device, equipment and storage medium | |
| CN115205750A (en) | Motion real-time counting method and system based on deep learning model | |
| CN115223240A (en) | Motion real-time counting method and system based on dynamic time warping algorithm | |
| CN114882562A (en) | Image processing method and device for preventing head from distortion |
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 |