CN112800900A - Mine personnel land falling detection method based on visual perception - Google Patents
Mine personnel land falling detection method based on visual perception Download PDFInfo
- Publication number
- CN112800900A CN112800900A CN202110065517.9A CN202110065517A CN112800900A CN 112800900 A CN112800900 A CN 112800900A CN 202110065517 A CN202110065517 A CN 202110065517A CN 112800900 A CN112800900 A CN 112800900A
- Authority
- CN
- China
- Prior art keywords
- human body
- target human
- falling
- target
- posture
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- 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
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Psychiatry (AREA)
- Multimedia (AREA)
- Social Psychology (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Software Systems (AREA)
- Psychology (AREA)
- Gerontology & Geriatric Medicine (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a mine personnel land falling detection method based on visual perception, which comprises the following steps: acquiring a video frame image; acquiring classification information and detection frame information of a target object from a video frame image, and screening a target human body; tracking a target human body to obtain tracking frame information and position information of the target human body; assigning the same ID to the same target human body; intercepting the target human body from the video frame image according to the position information; detecting skeletal key points of a target human body and displaying the positions of the skeletal key points; and calculating a relative position relation value of the skeleton key points, judging whether the target human body is in the falling posture or not and judging whether the falling posture maintaining time meets the preset alarm condition or not, and if so, sending an alarm signal. The invention designs a monitoring method based on visual perception for personnel safety monitoring in mine scenes, and applies technologies such as target detection, behavior analysis, anomaly analysis and early warning and the like by taking a visual perception technology as a means, thereby greatly ensuring the life safety of mining personnel.
Description
Technical Field
The invention relates to the technical field of video intelligent analysis and security monitoring, in particular to a mine personnel land falling detection method based on visual perception.
Background
When mine personnel work on the spot, safety situations such as sudden falling occur occasionally. At present, in the process of supervision and control of on-site operation, due to various reasons such as manpower, material resources and the like, management personnel related to each link sometimes cannot monitor and monitor the construction site or stay on the site for only a short time, cannot monitor the whole day, and finally cannot meet the requirements of the current operation permission system specification.
At present, the aspects of nonstandard behaviors of workers on a construction site, illegal use of equipment and facilities and the like are difficult to monitor, the traditional video monitoring is generally limited to manual identification and monitoring in a control room, even only main production devices and key parts can be monitored, and the requirement of monitoring when the nonstandard operation occurs in any production area cannot be met. Therefore, the prior art cannot be applied to safety detection of mine personnel.
Disclosure of Invention
In order to overcome the technical problems, the invention provides a visual perception-based mine personnel falling-ground detection method, which can perform target detection, behavior analysis and abnormal analysis early warning aiming at falling-ground behaviors of personnel in a mine scene, and furthest ensures the life safety of mine personnel.
In order to achieve the purpose, the technical scheme of the invention is as follows:
a visual perception-based mine personnel fall detection method, the method comprising:
acquiring a video frame image;
acquiring classification information and detection frame information of at least one type of target object from a current video frame image, and screening out a target human body as a detection object;
tracking the target human body to obtain tracking frame information of the target human body and determine position information of the target human body;
assigning the same ID to the same target human body;
intercepting the target human body from the video frame image according to the position information of the target human body;
detecting skeletal key points of a target human body and displaying the positions of the skeletal key points; when detecting the skeletal key points of the target human body, the adopted characteristic joint points include, but are not limited to, left shoulder, right shoulder, left elbow, right elbow, left wrist, right wrist, left crotch and right crotch.
Calculating a relative position relation value of the skeleton key points, and judging whether the target human body is in a falling posture; the relative position relationship of the key points of the bones comprises, but is not limited to, the height difference of the head, the waist and the ankle in the vertical direction.
And if the target human body is in the falling posture, judging whether the time length for maintaining the falling posture of the target human body meets the preset alarm condition, and if so, sending an alarm signal.
Preferably, the screening out the target human body as the detection object includes:
if the classification information of the at least one type of target object comprises the human classification information and the corresponding detection frame information, the object corresponding to the information is the target human body.
Preferably, the assigning the same ID to the same target human body includes:
and comparing the characteristics of the target human body in the current frame with the target human bodies in the previous and next frames, and determining that the same target human body is assigned with the same ID (identity) when the characteristic similarity is greater than or equal to a specific threshold value.
Preferably, said calculating a relative positional relationship value of said skeletal key points comprises
Calculating the confidence coefficients of key points of the left crotch and the right crotch of the target human body;
calculating the center point shoulder midpoint coordinates of the left shoulder and the right shoulder of the target human body;
calculating hip center point coordinates of center points of the left hip and the right hip of the target human body;
the difference in position in the transverse direction and/or the longitudinal direction between the shoulder midpoint and the crotch midpoint is calculated.
Preferably, the determining whether the target human body is in a falling posture includes:
setting a standard value of the falling posture;
and comparing the relative position relation value of the bone key point with the standard value of the falling posture, and judging that the target human body is in the falling posture when the relative position relation value of the bone key point meets the standard value of the falling posture.
Preferably, the setting of the standard value of the upside-down posture includes:
the confidence coefficients of the key points of the left crotch and the right crotch of the target human body are both greater than 0.5;
center point shoulder midpoint coordinates (x1, y1) of the left and right shoulders of the target human body;
center point crotch point coordinates (x2, y2) of the left crotch and right crotch of the target human body;
a flip posture is considered to occur when y1 < y2, or | x2-x1| > 2 | y2-y1 |.
Preferably, the judging whether the duration of the target human body maintaining the falling posture meets the preset alarm includes:
acquiring position information of a target human body;
intercepting a human body frame of the same target human body in the video frame image to obtain a corresponding video sequence;
calculating the duration of the target human body for maintaining the falling posture according to the video sequence;
comparing the time length of the target human body inverting alarm posture with a preset time length standard value;
and when the time length of the target human body maintaining the falling posture is more than or equal to the preset time length standard value, the alarm preset is met.
The invention has the beneficial effects that:
the application provides a mine personnel falling-down detection method based on visual perception, which can be used for realizing real-time monitoring, automatic problem finding and active early warning in the aspects of safety precaution, supervision implementation, quality detection and production flow management by taking a visual perception technology as a means aiming at sudden falling-down behaviors of personnel in a mine scene, and applying technologies such as target detection, behavior analysis, anomaly analysis early warning and the like, and finally realizing advanced control, prevention is the main, prevention is in the future, and the life safety of workers under a mine is ensured to the maximum extent.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flowchart of an embodiment of a mine personnel falling-down detection method based on visual perception according to the present application;
FIG. 2 is a diagram illustrating exemplary detection box information according to the present application;
FIG. 3 is a diagram illustrating exemplary tracking box information according to the present application;
FIG. 4 is a schematic diagram illustrating a screenshot of an exemplary video frame image of a human target according to the present application;
FIG. 5 is a schematic diagram illustrating exemplary human body key point information according to the present application;
FIG. 6 is a schematic diagram illustrating human body key point information of a human body target falling gesture according to an exemplary embodiment of the present disclosure;
FIG. 7 is a schematic diagram illustrating exemplary human key point information of a falling gesture of another human target according to the present application;
FIG. 8 is a schematic diagram of an exemplary over-ground alert target of the present application;
fig. 9 is a schematic diagram of an exemplary video sequence of the present application.
Detailed Description
In order that the objects, aspects and advantages of the present invention will become more apparent, the invention will be further described in detail with reference to the accompanying drawings and exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. The word "if," as used herein, may be interpreted as "when or" responsive to a determination, "depending on the context.
In order to overcome the defects of the prior art, the embodiment of the invention provides a visual perception-based mine worker land falling detection method, which combines the technologies of visual perception, application target detection, behavior analysis, anomaly analysis early warning and the like, can intelligently monitor and analyze the safety of mine workers, realizes safety precaution, real-time monitoring and active early warning of mine production operation, and ensures the safety and the standard of mine production operation. The mine personnel falling detection method based on visual perception claimed by the invention is further elaborated by combining the specific embodiment.
Referring to fig. 1, a schematic flow chart of an embodiment of a visual perception-based mine personnel falling-down detection method according to the present application is shown, where the method includes the following steps:
acquiring a video frame image; in the embodiment, a video frame image of a job site can be captured by setting a camera such as a high-definition camera in a job scene such as a mine and a mine.
Acquiring classification information and detection frame information of at least one type of target object from a current video frame image, and screening out a target human body as a detection object; in the embodiment of the invention, various types of target objects in the current frame in the video frame image can be classified and identified. As will be appreciated by those skilled in the art, the classification of the objects may be based on the classification in the coco dataset or may be designed by the skilled person himself, and may include humans, mine cars, mechanical equipment, etc. And when at least one type of target object is identified, acquiring the classification information and the detection frame information of the target object.
As an embodiment, the at least one type of object may be identified based on a current video frame image, classification information and detection frame information of the at least one type of object may be obtained, where the detection frame information includes coordinate information of a detection frame, and a target human body may be screened out according to the classification information. In the current frame video image, firstly, a target object is identified, for example, 5 target objects, three workers and two mine cars can be identified, then the target object is identified, classification information of the target object can be identified, and classification information and detection frame information of the target object are obtained, wherein the detection frame information can include detection frame coordinates and the like, and a target human body is screened out from the detection frame information to be used as a detection object. As shown in fig. 2, that is, the screened target human body, the detection frame information and the classification information corresponding to the target human body are displayed in the current video frame image. For example, the blocks shown in fig. 2 represent detection blocks of a Person classification in which a Person is represented by a Person by identification. The inspection box may also display coordinate information at the same time, for example, as can be understood by those skilled in the art, the coordinate information may be displayed on the upper left corner and the lower right corner of the inspection box, for example, the coordinates (X1, Y1) may be displayed on the upper left corner and the coordinates (X2, Y2) may be displayed on the lower right corner of the inspection box.
Tracking the target human body to obtain tracking frame information of the target human body and determine position information of the target human body; in this embodiment, after the target human body is screened out, after the classification information and the detection frame information of the target human body are obtained, the track of the target human body can be tracked based on the classification information and the detection frame information of the target human body, so that the tracking frame information of the target human body is obtained, and the position information of the target human body in the video frame can be determined. The detection frame can also display the human body key point information at the same time, as shown in fig. 3, which is a tracking frame information schematic diagram, from which the human body key point information of the worker in the falling state can be seen.
Assigning the same ID to the same target human body; in this embodiment, the human body features of the target human body in the video frame image may be detected, and the human body features of the previous and subsequent frames are compared, and when the human body feature similarity of the target human body in the previous and subsequent frames reaches a specific value, it is determined that the target human body is the same, and then the same ID is assigned to the same target for distinguishing different target human bodies. As can be understood by those skilled in the art, the same target human body can be estimated when the similarity of human body features reaches 90%.
Intercepting the target human body from the video frame image according to the position information of the target human body; specifically, a tracking frame of the target human body may be located in a continuous multi-frame video image including the target human body, and the target human body may be cut out from the video frame image. Video screenshots may be taken according to the size of the tracking box to obtain a corresponding video frame image sequence, such as the video sequence shown in fig. 9. As can be understood by those skilled in the art, the person is intercepted from the original video frame image according to the detected position information of the staff, i.e., the target human body, so that redundant background information is conveniently removed to prevent background interference, and the accuracy rate and the calculation efficiency of behavior recognition can be improved.
Detecting skeletal key points of a target human body and displaying the positions of the skeletal key points; when detecting the skeletal key points of the target human body, the adopted characteristic joint points include, but are not limited to, left shoulder, right shoulder, left elbow, right elbow, left wrist, right wrist, left crotch and right crotch. In this embodiment, for an image captured from an original video frame image, a human body key point detection method is used to detect a human body bone key point, and the human body bone key point is marked in the video frame image; referring to fig. 6-8, the human body key point information can be seen when the worker is in a fallen state. And connecting the skeletal key points of the human body to represent the falling state of the target human body. And connecting the skeletal key points of the human body to represent the action state of the target human body.
Calculating a relative position relation value of the skeleton key points, and judging whether the target human body is in a falling posture; the relative position relationship of the key points of the bones comprises, but is not limited to, the height difference of the head, the waist and the ankle in the vertical direction. In this embodiment, the current motion of the target human body may be determined according to the relative position relationship of the skeletal key points. For example, when the height of the two crotch center points is smaller than the two shoulder centers, or when the lateral distance between the two crotch center points and the two shoulder centers is greater than 2 times the vertical distance, it can be determined that the worker is in the falling posture. As can be understood by those skilled in the art, any bone key point that can be used to determine the posture state of the target human body by its relative position is claimed in the present invention, and for example, other bone key points of the human body can be selected as feature points, and the height difference of the head, waist and ankle in the vertical direction, which is the relative position relationship of these bone key points of the human body, can be used to determine whether the state of the human body in the current frame is falling.
And if the target human body is in the falling posture, judging whether the time length for maintaining the falling posture of the target human body meets the preset alarm condition, and if so, sending an alarm signal. In this embodiment, by detecting the video frame in which the early warning gesture is detected and the subsequent video frame, when the action of falling to the ground lasts for a certain time, it can be determined that the event of falling to the ground of the person occurs, and a falling to the ground alarm signal needs to be sent out.
Preferably, the screening out the target human body as the detection object includes:
if the classification information of the at least one type of target object comprises the human classification information and the corresponding detection frame information, the object corresponding to the information is the target human body.
Preferably, the assigning the same ID to the same target human body includes:
and comparing the characteristics of the target human body in the current frame with the target human bodies in the previous and next frames, and when the characteristic similarity is greater than or equal to a specific threshold value, determining the target human bodies as the same target human body and distributing the same ID for the same target human body. In this embodiment, the human body features of the target human body in the video frame image may be detected, and the human body features of the previous and subsequent frames are compared, and when the human body feature similarity of the target human body in the previous and subsequent frames reaches a specific value, it is determined that the target human body is the same, and then the same ID is assigned to the same target for distinguishing different target human bodies. As can be understood by those skilled in the art, the same target human body can be estimated when the similarity of human body features reaches 90%.
Preferably, said calculating a relative positional relationship value of said skeletal key points comprises
Calculating the confidence coefficients of key points of the left crotch and the right crotch of the target human body; in the embodiment of the present invention, the confidence level represents a confidence level, that is, a category attribute of a certain object is determined, including whether the object includes a certain category and a corresponding probability value, such as whether the object belongs to a and a probability that the object belongs to a. Confidence is a value between the intervals 0, 1.
The definition of confidence is expressed in mathematical form:
Ci j＝Pr(object)*IOUpredtruth
wherein, Ci jRepresenting the confidence of the jth bounding box of the ith mesh.
Calculating the center point shoulder midpoint coordinates of the left shoulder and the right shoulder of the target human body;
calculating hip center point coordinates of center points of the left hip and the right hip of the target human body;
the difference in position in the transverse direction and/or the longitudinal direction between the shoulder midpoint and the crotch midpoint is calculated.
Preferably, the determining whether the target human body is in a falling posture includes:
setting a standard value of the falling posture;
and comparing the relative position relation value of the bone key point with the standard value of the falling posture, and judging that the target human body is in the falling posture when the relative position relation value of the bone key point meets the standard value of the falling posture.
Preferably, the setting of the standard value of the upside-down posture includes:
the confidence coefficients of the key points of the left crotch and the right crotch of the target human body are both greater than 0.5;
center point shoulder midpoint coordinates (x1, y1) of the left and right shoulders of the target human body;
center point crotch point coordinates (x2, y2) of the left crotch and right crotch of the target human body;
a flip posture is considered to occur when y1 < y2, or | x2-x1| > 2 | y2-y1 |.
In this embodiment, the current motion of the target human body may be determined according to the relative position relationship of the skeletal key points. For example, when the height of the two crotch center points is smaller than the two shoulder centers, or when the lateral distance between the two crotch center points and the two shoulder centers is greater than 2 times the vertical distance, it can be determined that the worker is in the falling posture. As can be understood by those skilled in the art, any bone key point that can be used to determine the posture state of the target human body by its relative position is claimed in the present invention, and for example, other bone key points of the human body can be selected as feature points, and the height difference of the head, waist and ankle in the vertical direction, which is the relative position relationship of these bone key points of the human body, can be used to determine whether the state of the human body in the current frame is falling.
Preferably, the judging whether the duration of the target human body maintaining the falling posture meets the preset alarm includes:
acquiring position information of a target human body;
intercepting a human body frame of the same target human body in the video frame image to obtain a corresponding video sequence;
calculating the time length of the target human body for maintaining the early warning posture according to the video sequence;
comparing the time length of the target human body for maintaining the falling posture with a preset time length standard value;
and when the time length of the target human body maintaining the falling posture is more than or equal to the preset time length standard value, the alarm preset is met.
In this embodiment, by detecting the video frame in which the falling gesture is detected and the subsequent video frame, when the falling motion continues for a certain time, it can be determined that a personal safety event occurs, and an alarm signal needs to be sent.
The application provides a mine worker land falling detection method based on visual perception, which is characterized in that a target detection method is used for detecting position information of workers and a tracking algorithm is combined, so that unique identity information is given to the detected same person; then according to the detected target human body position information, people are intercepted from the original video frame image so as to remove redundant background information and realize the purpose of real-time detection; finally, aiming at the intercepted image, estimating key points of the human body; and then, judging whether the human body of the current frame is in a falling state or not by combining the relative position relation of the key points of the human body. By analogy with the subsequent video frames, when the falling down action is judged to continue for a certain time, the falling down event of the personnel can be determined to occur, and an alarm signal needs to be sent out. Compared with the prior art, the method and the device have the advantages that the dependence degree on the external environment is small, the robustness is good, the method and the device are suitable for achieving rapid and accurate safety detection in special scenes such as mine and mine, target tracking detection is conducted on continuous multi-frame video images, judgment on dangerous behaviors is achieved, the detection rate of personnel safety accidents is improved, the false alarm rate of the personnel safety accidents is reduced, and the life safety of underground workers is guaranteed.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are given by way of illustration of the principles of the present invention, and that various changes and modifications may be made without departing from the spirit and scope of the invention as defined by the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (7)
1. A mine personnel land falling detection method based on visual perception is characterized by comprising the following steps:
acquiring a video frame image;
acquiring classification information and detection frame information of at least one type of target object from a current video frame image, and screening out a target human body as a detection object;
tracking the target human body to obtain tracking frame information of the target human body and determine position information of the target human body;
assigning the same ID to the same target human body;
intercepting the target human body from the video frame image according to the position information of the target human body;
detecting skeletal key points of a target human body and displaying the positions of the skeletal key points; when detecting the skeletal key points of the target human body, the adopted characteristic joint points include, but are not limited to, left shoulder, right shoulder, left elbow, right elbow, left wrist, right wrist, left crotch and right crotch.
Calculating a relative position relation value of the skeleton key points, and judging whether the target human body is in a falling posture; the relative position relationship of the key points of the bones comprises, but is not limited to, the height difference of the head, the waist and the ankle in the vertical direction.
And if the target human body is in the falling posture, judging whether the time length for maintaining the falling posture of the target human body meets the preset alarm condition, and if so, sending an alarm signal.
2. The visual perception-based mine personnel falling-down detection method according to claim 1, wherein the screening of the target human body as a detection object comprises:
if the classification information of the at least one type of target object comprises the human classification information and the corresponding detection frame information, the object corresponding to the information is the target human body.
3. The visual perception-based mine personnel falling-down detection method according to claim 1, wherein the step of assigning the same ID to the same target human body comprises the steps of:
and comparing the characteristics of the target human body in the current frame with the target human bodies in the previous and next frames, and determining that the same target human body is assigned with the same ID (identity) when the characteristic similarity is greater than or equal to a specific threshold value.
4. The visual perception-based mine personnel falling-down detection method according to claim 1, wherein the calculating of the relative position relation value of the skeletal key points comprises
Calculating the confidence coefficients of key points of the left crotch and the right crotch of the target human body;
calculating the center point shoulder midpoint coordinates of the left shoulder and the right shoulder of the target human body;
calculating hip center point coordinates of center points of the left hip and the right hip of the target human body;
the difference in position in the transverse direction and/or the longitudinal direction between the shoulder midpoint and the crotch midpoint is calculated.
5. The visual perception-based mine personnel falling-down detection method according to claim 4, wherein the judging whether the target human body is in a falling-down posture comprises:
setting a standard value of the falling posture;
and comparing the relative position relation value of the bone key point with the standard value of the falling posture, and judging that the target human body is in the falling posture when the relative position relation value of the bone key point meets the standard value of the falling posture.
6. The visual perception-based mine personnel falling ground detection method according to claim 5, wherein the setting of the standard value of the falling ground posture comprises:
the confidence coefficients of the key points of the left crotch and the right crotch of the target human body are both greater than 0.5;
center point shoulder midpoint coordinates (x1, y1) of the left and right shoulders of the target human body;
center point crotch point coordinates (x2, y2) of the left crotch and right crotch of the target human body;
a flip posture is considered to occur when y1 < y2, or | x2-x1| > 2 | y2-y1 |.
7. The visual perception-based mine personnel falling-down detection method according to claim 1, wherein the step of judging whether the falling-down posture maintaining duration of the target human body meets an alarm preset comprises the following steps:
acquiring position information of a target human body;
intercepting a human body frame of the same target human body in the video frame image to obtain a corresponding video sequence;
calculating the duration of the target human body for maintaining the falling posture according to the video sequence;
comparing the time length of the target human body inverting alarm posture with a preset time length standard value;
and when the time length of the target human body maintaining the falling posture is more than or equal to the preset time length standard value, the alarm preset is met.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110065517.9A CN112800900A (en) | 2021-01-18 | 2021-01-18 | Mine personnel land falling detection method based on visual perception |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110065517.9A CN112800900A (en) | 2021-01-18 | 2021-01-18 | Mine personnel land falling detection method based on visual perception |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112800900A true CN112800900A (en) | 2021-05-14 |
Family
ID=75811022
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110065517.9A Pending CN112800900A (en) | 2021-01-18 | 2021-01-18 | Mine personnel land falling detection method based on visual perception |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112800900A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115985050A (en) * | 2022-12-28 | 2023-04-18 | 浪潮数字粮储科技有限公司 | Automatic alarm method, equipment and medium based on abnormal falling of personnel in grain depot |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108805093A (en) * | 2018-06-19 | 2018-11-13 | 华南理工大学 | Escalator passenger based on deep learning falls down detection algorithm |
| CN111274954A (en) * | 2020-01-20 | 2020-06-12 | 河北工业大学 | Embedded platform real-time falling detection method based on improved attitude estimation algorithm |
| CN111753587A (en) * | 2019-03-28 | 2020-10-09 | 杭州海康威视数字技术股份有限公司 | Method and device for detecting falling to ground |
| CN111783702A (en) * | 2020-07-20 | 2020-10-16 | 杭州叙简科技股份有限公司 | Efficient pedestrian tumble detection method based on image enhancement algorithm and human body key point positioning |
-
2021
- 2021-01-18 CN CN202110065517.9A patent/CN112800900A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108805093A (en) * | 2018-06-19 | 2018-11-13 | 华南理工大学 | Escalator passenger based on deep learning falls down detection algorithm |
| CN111753587A (en) * | 2019-03-28 | 2020-10-09 | 杭州海康威视数字技术股份有限公司 | Method and device for detecting falling to ground |
| CN111274954A (en) * | 2020-01-20 | 2020-06-12 | 河北工业大学 | Embedded platform real-time falling detection method based on improved attitude estimation algorithm |
| CN111783702A (en) * | 2020-07-20 | 2020-10-16 | 杭州叙简科技股份有限公司 | Efficient pedestrian tumble detection method based on image enhancement algorithm and human body key point positioning |
Non-Patent Citations (2)
| Title |
|---|
| LINGMEI REN ET AL.: "Research of Fall Detection and Fall Prevention Technologies: A Systematic Review", 《IEEE ACCESS》 * |
| 瞿畅 等: "基于Kinect体感传感器的老年人跌倒自动检测", 《传感技术学报》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115985050A (en) * | 2022-12-28 | 2023-04-18 | 浪潮数字粮储科技有限公司 | Automatic alarm method, equipment and medium based on abnormal falling of personnel in grain depot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112800901A (en) | Mine personnel safety detection method based on visual perception | |
| CN111144263B (en) | Construction worker high-falling accident early warning method and device | |
| CN106981163B (en) | A kind of personnel's invasion abnormal event alarming method | |
| US11917333B2 (en) | Systems and methods for personnel location at a drilling site | |
| US20030095782A1 (en) | System and method for detection and analysis of video recordings | |
| CN111445524A (en) | Scene understanding-based construction site worker unsafe behavior identification method | |
| KR101541272B1 (en) | Apparatus and method for detecting terrorism using irregular motion of peoples | |
| CN111079722B (en) | Hoisting process personnel safety monitoring method and system | |
| CN114155601A (en) | Vision-based method and system for detecting dangerous behaviors of operating personnel | |
| CN112804498A (en) | System and method for monitoring wearing condition of safety protection harness | |
| KR100916960B1 (en) | Monitoring system in railway station | |
| CN110231605B (en) | Human behavior recognition method and device, computer equipment and storage medium | |
| CN113223046B (en) | Method and system for identifying prisoner behaviors | |
| CN110956769B (en) | Monitoring method of perimeter anti-intrusion system based on target position | |
| KR20200017594A (en) | Method for Recognizing and Tracking Large-scale Object using Deep learning and Multi-Agent | |
| CN113506416A (en) | Engineering abnormity early warning method and system based on intelligent visual analysis | |
| CN106801617A (en) | Coal mine down-hole personnel injures early warning system | |
| CN112184773A (en) | Helmet wearing detection method and system based on deep learning | |
| CN109800715B (en) | Park entrance and exit monitoring method and system based on Internet of things | |
| CN113822250A (en) | Ship driving abnormal behavior detection method | |
| CN112541439A (en) | Intelligent video monitoring method and system for power plant | |
| CN116704411A (en) | Security control method, system and storage medium based on Internet of things | |
| Zhu et al. | Automated monitoring of hardhats wearing for onsite safety enhancement | |
| CN112800900A (en) | Mine personnel land falling detection method based on visual perception | |
| CN115797856A (en) | Intelligent construction scene safety monitoring method based on machine vision |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210514 |
|
| WD01 | Invention patent application deemed withdrawn after publication |