CN114495015A - Human body posture detection method and device - Google Patents
Human body posture detection method and device Download PDFInfo
- Publication number
- CN114495015A CN114495015A CN202210325258.3A CN202210325258A CN114495015A CN 114495015 A CN114495015 A CN 114495015A CN 202210325258 A CN202210325258 A CN 202210325258A CN 114495015 A CN114495015 A CN 114495015A
- Authority
- CN
- China
- Prior art keywords
- human body
- video
- image
- body posture
- frames
- 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
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The present disclosure provides a human body posture detection method and device, the method comprising: receiving a video stream of a target area sent by a monitoring terminal, and extracting a plurality of frames of video frames including human body images from the video stream; superposing the pixel values in the multi-frame video frame to generate image superposed data; inputting the image superposition data into a pre-trained posture detection model, and outputting an image comprising a human body contour and a limb change track; and determining the human body posture according to the position of the changed limb relative to the human body outline. In this way, the human body violation behaviors can be found in time through the recognition of the human body actions, so that the hidden danger of the abnormal behaviors is eliminated in time, and the safety is improved.
Description
Technical Field
Embodiments of the present disclosure relate generally to the field of image recognition technology, and more particularly, to a human body posture detection method and apparatus.
Background
With the popularization of video monitoring equipment, the cost for monitoring video contents by adopting human resources is increased sharply, so that behavior recognition technology based on video monitoring is widely concerned by people. On one hand, in public places with dense personnel, such as areas of public transportation hubs and the like, the rapid behavior detection of crowded streams of people by using limited security resources is a huge challenge; on the other hand, in some monitoring scenes with rare people, such as gas stations and the like, which need to perform operations strictly according to the regulations, the 24-hour abnormal monitoring is a time-consuming and labor-consuming task for supervisors.
In general, the existing human body posture detection methods mainly include the following methods:
the security personnel detect the monitored scene in real time through the monitoring camera, and immediately alarm and process the illegal action. However, the detection mode using human subjects has great limitation, when the number of monitoring pictures is large, a large number of security personnel are needed to complete the real-time monitoring, and the personnel cost is high; moreover, security personnel cannot keep continuous concentration, and careless omission occurs.
Disclosure of Invention
According to the embodiment of the disclosure, a human body posture detection method is provided, which is used for detecting a monitoring area needing abnormal human body behaviors, detecting the human body posture, and timely finding out the human body violation behaviors through human body action recognition, so that hidden dangers caused by the abnormal behaviors are timely eliminated, and the safety is improved.
In a first aspect of the present disclosure, a human body posture detection method is provided, including:
receiving a video stream of a target area sent by a monitoring terminal, and extracting a plurality of frames of video frames including human body images from the video stream;
superposing the pixel values in the multi-frame video frame to generate image superposed data;
inputting the image superposition data into a pre-trained posture detection model, and outputting an image comprising a human body contour and a limb change track;
and determining the human body posture according to the position of the changed limb relative to the human body outline.
In some embodiments, further comprising:
and matching the determined human body posture with a pre-established human body posture model, and classifying the human body posture, wherein the human body posture model corresponds to a human body posture category.
In some embodiments, the method further includes a process of pre-establishing a posture detection model, specifically including:
carrying out superposition processing on pixel values in a multi-frame video frame comprising a human body image with a human body contour and a limb change track marked in advance to obtain image superposition data as a training sample, training a pre-constructed convolutional neural network model, and outputting an image of the human body contour and the limb change track as a recognition result;
responding that errors of the human body contour and the limb change track in the recognition result and the pre-marked human body contour and the limb change track are larger than a preset threshold value, and adjusting parameters of each layer in the pre-constructed convolutional neural network model;
and repeating the process until the errors of the human body contour and the limb change track in the recognition result and the pre-marked human body contour and the limb change track are smaller than a preset threshold value, and finishing the training of the posture detection model.
In some embodiments, the receiving a video stream of a target area sent by a monitoring terminal, and extracting multiple frames of video frames including human body images from the video stream includes:
and receiving video streams of the target area shot from different azimuth angles and sent by the monitoring terminal, and extracting a plurality of frames of video frames including human body images for the video streams shot from each azimuth angle.
In some embodiments, the determining the body pose from the position of the varying limb relative to the body contour comprises:
and responding to the image correspondence of the limb change track corresponding to the video stream of the target area shot from different azimuth angles, and determining the human body posture according to the position of the changed limb relative to the human body contour.
In some embodiments, further comprising:
and aligning the multiple frames of video frames, and cutting the aligned video frames into video frames with the same size.
In some embodiments, said superimposing pixel values in said plurality of frames of video frames, generating image superimposition data, comprises:
and overlapping the pixel values of the same position in the video frames with the same size after cutting, and generating image overlapping data by using the numerical value obtained by overlapping as the data of the corresponding pixel, wherein the image overlapping data comprises the pixel position and the corresponding numerical value.
In a second aspect of the present disclosure, there is provided a human body posture detecting device including:
the system comprises a video frame extraction module, a video frame extraction module and a video frame acquisition module, wherein the video frame extraction module is used for receiving a video stream of a target area sent by a monitoring terminal and extracting a plurality of frames of video frames comprising human body images from the video stream;
the data generation module is used for superposing the pixel values in the multi-frame video frames to generate image superposition data;
the image identification module is used for inputting the image superposition data into a pre-trained posture detection model and outputting an image comprising a human body contour and a limb change track;
and the human body posture determining module is used for determining the human body posture according to the position of the changed limb relative to the human body outline.
In a third aspect of the present disclosure, an electronic device is provided, comprising a memory having stored thereon a computer program and a processor implementing the method as described above when executing the program.
In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, which program, when being executed by a processor, is adapted to carry out the method as set forth above.
By the human body posture detection method, the human body violation behaviors can be found in time through the recognition of the human body actions, so that the hidden danger of the abnormal behaviors is eliminated in time, and the safety is improved.
The statements made in this summary are not intended to limit key or critical features of the embodiments of the disclosure, nor are they intended to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. In the drawings, like or similar reference characters designate like or similar elements, and wherein:
fig. 1 shows a flowchart of a human body posture detection method according to a first embodiment of the present disclosure;
fig. 2 shows a schematic structural diagram of a human body posture detection device according to a second embodiment of the disclosure;
fig. 3 shows a schematic structural diagram of a human body posture detection device according to a third embodiment of the present disclosure.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are some, but not all embodiments of the present disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
In addition, the term "and/or" herein is only one kind of association relationship describing an associated object, and means that there may be three kinds of relationships, for example, a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.
According to the human body posture detection method, the human body posture can be recognized, for example, the human body gesture action, the limb behavior or the posture characteristic and the like, so that the violation behavior is detected, the hidden danger of the abnormal behavior is eliminated in time, and the safety is improved.
The technical solution of the present disclosure is explained below with reference to specific examples. Fig. 1 is a flowchart of a human body posture detection method according to a first embodiment of the present disclosure. As can be seen from fig. 1, the human body posture detection method of the present embodiment may include the following steps:
s101: receiving a video stream of a target area sent by a monitoring terminal, and extracting a plurality of frames of video frames including human body images from the video stream.
The human body posture detection method disclosed by the embodiment of the disclosure can be applied to various monitoring scenes, and the human body posture can be detected through monitoring equipment. The body posture in this embodiment may include gesture actions, limb actions, appearance features, and the like, for example, a call is made during driving, smoking is performed in a smoking banned place, a security door is pushed or pulled, illegal operation is performed, or a security facility is not worn, and the like, which are not listed in this embodiment.
When the human body posture is detected by the monitoring equipment, the background server receives the video stream of the target area sent by the monitoring terminal. In this embodiment, the monitoring terminal may monitor the target area at a fixed angle, or may monitor the target area in a rotating manner.
After receiving the video stream sent by the monitoring terminal, the background server extracts a plurality of video frames including the human body image from the video stream. Specifically, since video frames in which a human body exists usually exist continuously in a video stream, in the process of extracting video frames including human body images, a partial video stream with human body images may be cut from an original video, and then video frames are extracted frame by frame from the cut partial video stream, so as to obtain multiple frames of video frames including human body images.
S102: and superposing the pixel values in the multi-frame video frame to generate image superposed data.
In this embodiment, after obtaining a plurality of frames of video frames including a human body image, pixel values in the obtained plurality of frames of video frames may be superimposed to generate image superimposed data. Specifically, the multiple frames of video frames may be aligned first, and the aligned video frames may be cut into video frames with the same size. In the alignment process, two or more points with unchanged positions in multiple frames of video frames can be used as reference points to align the multiple frames of video frames, and then the video frames are cut based on the aligned video frames, so that a complete human body image is kept in each frame of video frames.
The pixel values at the same position in the video frames with the same size after being cut out are superposed, and the superposed numerical value is used as the data of the corresponding pixel to generate the image superposed data, wherein the image superposed data comprises the pixel position and the corresponding numerical value, so that a plurality of frames of video frames can be superposed into the image superposed data with the pixel value exceeding 255, for example, the pixel values at the same position in three images are respectively 168, 120 and 150, and the generated image superposed data and the data corresponding to the pixel are respectively 168+120+150=438, and similarly, the pixel values at the same position in the video frames after being cut out can be superposed to generate the corresponding image superposed data. The generated superimposition data does not have the characteristics of an image because the value of the position of the corresponding pixel value is generally much larger than 255.
Specifically, a matrix with the same storage digit number as the number of pixels in the clipped video frame may be created, then the pixel values at the same position in the clipped video frame with the same size are superimposed, the numerical value obtained by the superimposition is used as the data of the corresponding pixel and written into the created matrix, and thus, the value of each element in the matrix is the sum of the pixel points in the corresponding clipped video frame, and the image superimposed data is obtained.
S103: and inputting the image superposition data into a pre-trained posture detection model, and outputting an image comprising a human body outline and a limb change track.
In this embodiment, after the image overlay data is obtained, the image overlay data may be input to a pre-trained posture detection model, and an image including a human body contour and a limb change trajectory is output.
The posture detection model in this embodiment is obtained by training in the following manner:
carrying out superposition processing on pixel values in a multi-frame video frame comprising a human body image with a human body contour and a limb change track marked in advance to obtain image superposition data as a training sample, training a pre-constructed convolutional neural network model, and outputting an image of the human body contour and the limb change track as a recognition result;
responding that errors of the human body contour and the limb change track in the recognition result and the pre-marked human body contour and the limb change track are larger than a preset threshold value, and adjusting parameters of each layer in the pre-constructed convolutional neural network model;
and repeating the process until the errors of the human body contour and the limb change track in the recognition result and the pre-marked human body contour and the limb change track are smaller than a preset threshold value, and finishing the training of the posture detection model.
S104: and determining the human body posture according to the position of the changed limb relative to the human body outline.
After the image comprising the human body outline and the limb change track is obtained, the human body posture can be determined according to the position of the changed limb relative to the human body outline, whether violation behaviors exist or not is further determined according to the human body posture, hidden dangers of the abnormal behaviors are timely recognized and eliminated, and safety is improved.
According to the human body posture detection method, the human body posture can be recognized, for example, the human body gesture action, the limb behavior or the posture characteristic and the like, so that the violation behavior is detected, the hidden danger of the abnormal behavior is eliminated in time, and the safety is improved.
The human body posture detection method can be applied to human body posture detection in the field of small data monitoring, and in the prior art, continuous human body postures are difficult to identify through a single image, and efficient and rapid monitoring is difficult to realize when continuous video frames are identified, so that the human body postures in images of several frames or dozens of frames are rapidly identified in a picture overlapping mode.
In addition, as an optional embodiment of the present disclosure, on the basis of the above embodiment, the determined body posture may be matched with a body posture model established in advance, and the body posture may be classified, where the body posture model corresponds to a body posture category.
As an optional embodiment of the present disclosure, the received video stream of the target area sent by the monitoring terminal may be a video stream of the target area shot from different azimuth angles, and for the video stream shot from each azimuth angle, multiple frames of video frames including human body images are extracted. In the process of determining the human body posture, whether images of limb change tracks corresponding to video streams of the target area shot from different azimuth angles correspond or not can be judged, and then the human body posture is determined according to the position of the changed limbs relative to the human body outline.
It is noted that while for simplicity of explanation, the foregoing method embodiments have been described as a series of acts or combination of acts, it will be appreciated by those skilled in the art that the present disclosure is not limited by the order of acts, as some steps may, in accordance with the present disclosure, occur in other orders and concurrently. Further, those skilled in the art should also appreciate that the embodiments described in this specification are all alternative embodiments and that the acts and modules involved are not necessarily essential to the disclosure.
The above is a description of embodiments of the method, and the embodiments of the apparatus are further described below.
Fig. 2 is a schematic structural diagram of a human body posture detection device according to a second embodiment of the disclosure. The human body posture detection device of the embodiment includes:
the video frame extraction module 201 is configured to receive a video stream of a target area sent by a monitoring terminal, and extract a plurality of video frames including a human body image from the video stream.
And the data generating module 202 is configured to superimpose pixel values in the multiple frames of video frames to generate image superimposed data.
And the image recognition module 203 is used for inputting the image superposition data into a pre-trained posture detection model and outputting an image comprising a human body contour and a limb change track.
And a human body posture determining module 204 for determining the human body posture according to the position of the changed limb relative to the human body contour.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the described module may refer to the corresponding process in the foregoing method embodiment, and is not described herein again.
FIG. 3 shows a schematic block diagram of an electronic device 300 that may be used to implement embodiments of the present disclosure. As shown, device 300 includes a Central Processing Unit (CPU) 301 that may perform various appropriate actions and processes in accordance with computer program instructions stored in a Read Only Memory (ROM) 302 or loaded from a storage unit 308 into a Random Access Memory (RAM) 303. In the RAM 303, various programs and data necessary for the operation of the device 300 can also be stored. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An input/output (I/O) interface 305 is also connected to bus 304.
Various components in device 300 are connected to I/O interface 305, including: an input unit 306 such as a keyboard, a mouse, or the like; an output unit 307 such as various types of displays, speakers, and the like; a storage unit 308 such as a magnetic disk, optical disk, or the like; and a communication unit 309 such as a network card, modem, wireless communication transceiver, etc. The communication unit 309 allows the device 300 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.
The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a load programmable logic device (CPLD), and the like.
Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
Further, while operations are depicted in a particular order, this should be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single implementation. Conversely, various features that are described in the context of a single implementation can also be implemented in multiple implementations separately or in any suitable subcombination.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
Claims (10)
1. A human body posture detection method is characterized by comprising the following steps:
receiving a video stream of a target area sent by a monitoring terminal, and extracting a plurality of frames of video frames including human body images from the video stream;
superposing the pixel values in the multi-frame video frame to generate image superposed data;
inputting the image superposition data into a pre-trained posture detection model, and outputting an image comprising a human body contour and a limb change track;
and determining the human body posture according to the position of the changed limb relative to the human body outline.
2. The human body posture detection method according to claim 1, further comprising:
and matching the determined human body posture with a pre-established human body posture model, and classifying the human body posture, wherein the human body posture model corresponds to a human body posture category.
3. The human body posture detection method according to claim 2, further comprising a process of pre-establishing a posture detection model, specifically comprising:
carrying out superposition processing on pixel values in a multi-frame video frame comprising a human body image with a human body contour and a limb change track marked in advance to obtain image superposition data as a training sample, training a pre-constructed convolutional neural network model, and outputting an image of the human body contour and the limb change track as a recognition result;
responding that errors of the human body contour and the limb change track in the recognition result and the pre-marked human body contour and the limb change track are larger than a preset threshold value, and adjusting parameters of each layer in the pre-constructed convolutional neural network model;
and repeating the process until the errors of the human body contour and the limb change track in the recognition result and the pre-marked human body contour and the limb change track are smaller than a preset threshold value, and finishing the training of the posture detection model.
4. The human body posture detection method according to claim 3, wherein the receiving a video stream of a target area sent by a monitoring terminal, extracting a plurality of frames of video frames including human body images from the video stream, comprises:
and receiving video streams of the target area shot from different azimuth angles and sent by the monitoring terminal, and extracting a plurality of frames of video frames including human body images for the video streams shot from each azimuth angle.
5. The human body posture detection method of claim 4, wherein the determining the human body posture according to the position of the changed limb relative to the human body contour comprises:
and responding to the image correspondence of the limb change track corresponding to the video stream of the target area shot from different azimuth angles, and determining the human body posture according to the position of the changed limb relative to the human body contour.
6. The human body posture detection method according to claim 3, characterized by further comprising:
and aligning the multiple frames of video frames, and cutting the aligned video frames into video frames with the same size.
7. The human body posture detection method of claim 6, wherein the superimposing pixel values in the plurality of frames of video frames to generate image superimposed data comprises:
and overlapping the pixel values of the same position in the video frames with the same size after being cut, and generating image overlapping data by taking the numerical value obtained by overlapping as the data of the corresponding pixel, wherein the image overlapping data comprises the pixel position and the corresponding numerical value.
8. A human body posture detecting device, characterized by comprising:
the system comprises a video frame extraction module, a video frame extraction module and a video frame acquisition module, wherein the video frame extraction module is used for receiving a video stream of a target area sent by a monitoring terminal and extracting a plurality of frames of video frames comprising human body images from the video stream;
the data generation module is used for superposing the pixel values in the multi-frame video frames to generate image superposition data;
the image identification module is used for inputting the image superposition data into a pre-trained posture detection model and outputting an image comprising a human body contour and a limb change track;
and the human body posture determining module is used for determining the human body posture according to the position of the changed limb relative to the human body outline.
9. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program, wherein the processor, when executing the program, implements the method of any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210325258.3A CN114495015A (en) | 2022-03-30 | 2022-03-30 | Human body posture detection method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210325258.3A CN114495015A (en) | 2022-03-30 | 2022-03-30 | Human body posture detection method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114495015A true CN114495015A (en) | 2022-05-13 |
Family
ID=81489044
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210325258.3A Pending CN114495015A (en) | 2022-03-30 | 2022-03-30 | Human body posture detection method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114495015A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115171217A (en) * | 2022-07-27 | 2022-10-11 | 北京拙河科技有限公司 | Action recognition method and system under dynamic background |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107885324A (en) * | 2017-09-28 | 2018-04-06 | 江南大学 | A kind of man-machine interaction method based on convolutional neural networks |
WO2020232069A1 (en) * | 2019-05-15 | 2020-11-19 | Northeastern University | Video 2d multi-person pose estimation using multi-frame refinement and optimization |
CN112784765A (en) * | 2021-01-27 | 2021-05-11 | 北京百度网讯科技有限公司 | Method, apparatus, device and storage medium for recognizing motion |
CN113191319A (en) * | 2021-05-21 | 2021-07-30 | 河南理工大学 | Human body posture intelligent recognition method and computer equipment |
CN113326835A (en) * | 2021-08-04 | 2021-08-31 | 中国科学院深圳先进技术研究院 | Action detection method and device, terminal equipment and storage medium |
CN113887469A (en) * | 2021-10-15 | 2022-01-04 | 黑龙江雨谷科技有限公司 | Method, system and storage medium for pedestrian fall detection |
-
2022
- 2022-03-30 CN CN202210325258.3A patent/CN114495015A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107885324A (en) * | 2017-09-28 | 2018-04-06 | 江南大学 | A kind of man-machine interaction method based on convolutional neural networks |
WO2020232069A1 (en) * | 2019-05-15 | 2020-11-19 | Northeastern University | Video 2d multi-person pose estimation using multi-frame refinement and optimization |
CN112784765A (en) * | 2021-01-27 | 2021-05-11 | 北京百度网讯科技有限公司 | Method, apparatus, device and storage medium for recognizing motion |
CN113191319A (en) * | 2021-05-21 | 2021-07-30 | 河南理工大学 | Human body posture intelligent recognition method and computer equipment |
CN113326835A (en) * | 2021-08-04 | 2021-08-31 | 中国科学院深圳先进技术研究院 | Action detection method and device, terminal equipment and storage medium |
CN113887469A (en) * | 2021-10-15 | 2022-01-04 | 黑龙江雨谷科技有限公司 | Method, system and storage medium for pedestrian fall detection |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115171217A (en) * | 2022-07-27 | 2022-10-11 | 北京拙河科技有限公司 | Action recognition method and system under dynamic background |
CN115171217B (en) * | 2022-07-27 | 2023-03-03 | 北京拙河科技有限公司 | Action recognition method and system under dynamic background |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101271514B (en) | Image detection method and device for fast object detection and objective output | |
US10949657B2 (en) | Person's behavior monitoring device and person's behavior monitoring system | |
CN111160202A (en) | AR equipment-based identity verification method, AR equipment-based identity verification device, AR equipment-based identity verification equipment and storage medium | |
CN110705353A (en) | Method and device for identifying face to be shielded based on attention mechanism | |
CN111814776B (en) | Image processing method, device, server and storage medium | |
CN113807289A (en) | Human body posture detection method and device, electronic equipment and storage medium | |
CN113221767A (en) | Method for training living body face recognition model and method for recognizing living body face and related device | |
CN114495015A (en) | Human body posture detection method and device | |
CN113011300A (en) | Method, system and equipment for AI visual identification of violation behavior | |
CN114764895A (en) | Abnormal behavior detection device and method | |
CN112215567A (en) | Production flow compliance checking method and system, storage medium and terminal | |
CN115497152A (en) | Customer information analysis method, device, system and medium based on image recognition | |
CN116229502A (en) | Image-based tumbling behavior identification method and equipment | |
CN115578731A (en) | Method and device for generating safety control information, electronic equipment and storage medium | |
CN113469138A (en) | Object detection method and device, storage medium and electronic equipment | |
CN113536847A (en) | Industrial scene video analysis system and method based on deep learning | |
CN112699798A (en) | Traffic police action recognition method and device with vehicle-road cooperation | |
CN114091503A (en) | Face image verification method and device, electronic equipment and storage medium | |
US20210097392A1 (en) | Classification and re-identification | |
CN114038197B (en) | Scene state determining method and device, storage medium and electronic device | |
EP3837821B1 (en) | Assessing condition of real-world objects | |
CN115580830B (en) | Passenger violation path detection method and device based on AP probe multipoint positioning | |
CN114519882A (en) | Face recognition method and device and electronic equipment | |
CN117975405A (en) | Inspection track monitoring method and device based on deep learning | |
CN113792595A (en) | Target behavior detection method and device, computer equipment and storage medium |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220513 |