CN113137983B - Self-learning well lid posture monitoring method and monitoring system - Google Patents
Self-learning well lid posture monitoring method and monitoring system Download PDFInfo
- Publication number
- CN113137983B CN113137983B CN202110485720.1A CN202110485720A CN113137983B CN 113137983 B CN113137983 B CN 113137983B CN 202110485720 A CN202110485720 A CN 202110485720A CN 113137983 B CN113137983 B CN 113137983B
- Authority
- CN
- China
- Prior art keywords
- data
- well lid
- gesture
- gesture data
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
-
- 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Alarm Systems (AREA)
Abstract
The invention provides a self-learning well lid posture monitoring method and a self-learning well lid posture monitoring system, wherein the well lid posture monitoring method comprises the following steps: acquiring attitude data of the well lid body in real time; comparing the gesture data with normal gesture data in a gesture database, and judging whether the gesture data are normal or not according to the occurrence frequency, occurrence time and duration of the gesture data; when the gesture data are normal, updating the gesture data into the gesture database; and when the gesture data is abnormal, reporting alarm information of the abnormal well lid to a server. The well lid posture monitoring method can autonomously distinguish whether the data of the well lid body is abnormal or not, and can autonomously update the posture database, so that the sensitivity of monitoring is dynamically adjusted, and the intelligent degree is higher.
Description
Technical Field
The invention relates to the technical field of well lid monitoring, in particular to a self-learning well lid posture monitoring method and a self-learning well lid posture monitoring system.
Background
With the continuous acceleration of urban construction pace, the number and scale of urban road reconstruction, new construction, extension and overhaul projects are continuously enlarged, and with the continuous increase of underground pipe network projects and ground inspection wells corresponding to each road project, the number of various well covers is correspondingly increased, various rain sewage wells, water collecting wells, various rain sewage pipeline inspection wells on district pipe streets and various inspection wells of departments such as communication, electric power, water supply, heating power, gas and the like on primary and secondary roads are continuously increased, and the number of the inspection wells is very large.
The well cover is not timely repaired due to the fact that an external well cover lacks an effective real-time monitoring management means, lawless persons are provided with a machine capable of taking the well cover, illegal actions such as moving and stealing the well cover occur, meanwhile, damaged well covers cannot be timely recovered due to the fact that the damaged well covers cannot be timely obtained, normal operation of related equipment is affected, huge direct or indirect economic loss is caused, moreover, the well heads of the lost well covers can cause great harm to vehicles and pedestrians on a road, social stability and safety are greatly negatively affected, so that people injury and vehicle damage events occur in many areas, and great threat is caused to travel safety of people. The well lid problem has become the focus of social attention now, and well lid intelligent monitoring becomes very important.
The traditional well lid monitoring method is too simple, the monitoring sensitivity cannot be dynamically adjusted, and under the condition that high monitoring sensitivity is required, when vehicles and pedestrians pass through the well lid, false alarms of a monitoring device can be triggered; conversely, in the case where low monitoring sensitivity is required, such an action as dragging the manhole cover may occur, and no abnormal movement is detected, and the warning is missed.
Accordingly, the prior art is still in need of improvement and development.
Disclosure of Invention
The invention mainly aims to solve the technical problems that the existing well lid posture monitoring method is not intelligent enough, cannot be suitable for different monitoring scenes and cannot dynamically adjust the monitoring sensitivity.
The first aspect of the invention provides a self-learning well lid posture monitoring method, which comprises the following steps:
acquiring attitude data of the well lid body in real time;
comparing the gesture data with normal gesture data in a gesture database, and judging whether the gesture data are normal or not according to the occurrence frequency, occurrence time and duration of the gesture data;
when the gesture data are normal, updating the gesture data into the gesture database;
and when the gesture data is abnormal, reporting alarm information of the abnormal well lid to a server.
In an optional implementation manner of the first aspect of the present invention, before the collecting, in real time, attitude data of the manhole cover body includes:
collecting normal posture data of the well lid body when a vehicle or a person passes through the well lid body on a daily basis;
and uniformly dividing the measuring range of the attitude sensor into different intervals, counting the frequency value, the occurrence time and the duration of the normal attitude data in each interval, and generating an attitude database.
In an optional embodiment of the first aspect of the present invention, the dividing the range of the attitude sensor into different intervals includes:
dividing an X-direction acceleration range, a Y-direction acceleration range and a Z-direction acceleration range of the attitude sensor into 10 sections equally;
dividing the X-direction angular velocity range, the Y-direction angular velocity range and the Z-direction angular velocity range of the attitude sensor into 10 sections equally;
and dividing the inclination angle range of the attitude sensor into 10 sections.
In an alternative embodiment of the first aspect of the present invention, when the value of the X-direction acceleration, the Y-direction acceleration, the Z-direction acceleration, the X-direction angular velocity, the Y-direction angular velocity, the Z-direction angular velocity, or the inclination angle falls in 1 section thereof, the frequency value of the section increases by a fixed value Δv, and the frequency values of the remaining sections decrease by Δv/(10-1).
In an optional implementation manner of the first aspect of the present invention, the comparing the gesture data with normal gesture data in a gesture database, and determining whether the gesture data is normal according to a frequency, a time and a duration of occurrence of the gesture data includes:
calculating the frequency similarity, occurrence time similarity and duration time similarity of the gesture data and normal gesture data in a gesture database;
weighting and calculating the frequency similarity, the occurrence time similarity and the duration time similarity to obtain total similarity;
judging whether 100% -total similarity is larger than a preset threshold value;
and if the 100% -total similarity is greater than a preset threshold, judging that the gesture data is abnormal.
In an optional implementation manner of the first aspect of the present invention, after the collecting the posture data of the manhole cover body in real time, before comparing the posture data with the normal posture data in the posture database, the method includes:
and carrying out data noise reduction and data fusion on the gesture data.
In an optional implementation manner of the first aspect of the present invention, the performing data denoising and data fusion on the gesture data includes:
firstly, carrying out data denoising on the gesture data, and then carrying out data fusion on the gesture data after denoising by a quaternion method to obtain gesture information.
The invention provides a well lid posture monitoring system, which comprises a well lid and a server, wherein the well lid is in communication connection with the server through a base station, the well lid comprises a well lid body, and a posture sensor, a memory, a processor and a communication module which are arranged on the well lid body, wherein the posture sensor, the memory and the communication module are all connected with the processor, the posture sensor is used for collecting posture data of the well lid body, the memory is used for storing the posture data and storing instructions of the well lid posture monitoring method according to any one of the above, and the processor calls the instructions in the memory and executes the instructions.
In an alternative embodiment of the second aspect of the present invention, the manhole cover further comprises a GNSS sensor disposed on the manhole cover body, the GNSS sensor being connected to the processor.
In an alternative embodiment of the second aspect of the present invention, the manhole cover further comprises a temperature sensor provided on the manhole cover body, and the temperature sensor is connected to the processor.
The beneficial effects are that: the invention provides a self-learning well lid posture monitoring method and a self-learning well lid posture monitoring system, wherein the well lid posture monitoring method comprises the following steps: acquiring attitude data of the well lid body in real time; comparing the gesture data with normal gesture data in a gesture database, and judging whether the gesture data are normal or not according to the occurrence frequency, occurrence time and duration of the gesture data; when the gesture data are normal, updating the gesture data into the gesture database; and when the gesture data is abnormal, reporting alarm information of the abnormal well lid to a server. The well lid posture monitoring method can autonomously distinguish whether the data of the well lid body is abnormal or not, and can autonomously update the posture database, so that the sensitivity of monitoring is dynamically adjusted, and the intelligent degree is higher.
Drawings
FIG. 1 is a schematic diagram of an embodiment of a method for monitoring the posture of a well lid with self-learning function according to the present invention;
FIG. 2 is a schematic diagram of an embodiment of a manhole cover attitude monitoring system according to the present invention.
Detailed Description
The embodiment of the invention provides a self-learning well lid posture monitoring method and a self-learning well lid posture monitoring system.
The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
In the existing well lid monitoring method, the following defects mainly exist: monitoring is either too sensitive or monitoring is not sensitive. The traditional monitoring method is too simple, and when the monitoring sensitivity is improved, vehicles and pedestrians pass through the well cover, false alarms of the monitoring device can be triggered; conversely, if the monitoring sensitivity is reduced, then, for such an action of dragging the manhole cover, no abnormal movement is monitored, and a warning missing problem may occur.
Difficult to adapt to complicated well lid type, mounted position: in actual use, the well lid type that needs to monitor is numerous, and the material is multiple, and the mounted position is also different. Often different types of well covers and different mounting positions, different monitoring sensitivities need to be set. For example, cast iron manhole covers are usually installed on roads, and have a large bearing capacity, so that when a vehicle passes by, a large degree of vibration can occur, and lower monitoring sensitivity is required; in addition, the composite material well lid is mainly installed on a non-motor vehicle lane, the possibility of vibration is low, and the well lid is usually pulled open, so that high monitoring sensitivity is required.
The number of well covers to be monitored is huge, and if monitoring sensitivity is configured manually, the requirement on the professional of operators is met, and the workload is also great. The monitoring equipment needs to have self-learning and self-adapting capability, and automatically controls alarm conditions and alarm sensitivity according to the condition of each alarm and the result of manual error correction, so as to reduce the problems of false alarm and missing alarm. Existing monitoring technologies lack such characteristics, and improvements are needed.
In order to solve the above technical problems, referring to fig. 1, the first aspect of the present invention provides a method for monitoring the posture of a well lid in a self-learning manner, which includes:
s100, acquiring attitude data of a well lid body in real time;
in this embodiment, the attitude data is acquired by sensors including, but not limited to, GNSS sensors, attitude sensors, and temperature sensors;
s200, comparing the gesture data with normal gesture data in a gesture database, and judging whether the gesture data are normal or not according to the occurrence frequency, the occurrence time and the duration time of the gesture data;
in this embodiment, when a vehicle and a person pass through the manhole cover, the manhole cover will vibrate, the sensor can obtain the posture data of the manhole cover body, and then through comparison of historical data, whether the posture data is normal data can be judged, and in the judging process, the frequency, the time and the duration of occurrence of the posture data are mainly compared, if the posture data belong to a section with a higher frequency value, the occurrence frequency is relatively high, and the posture data can be considered as a normal state, and the occurrence time and the duration of the posture data can be comprehensively judged in the judging process, so that the judging accuracy is improved;
s300, when the gesture data are normal, updating the gesture data into the gesture database; in this embodiment, when the gesture data is judged to be normal data, the normal gesture data is stored in the gesture database, so that when the same event occurs again, the gesture data can be recognized more quickly, and the judging efficiency is improved;
and S400, reporting alarm information of well lid abnormality to a server when the gesture data are abnormal. In this embodiment, when the gesture data is obvious and does not coincide with the history data, an alarm message is sent to the server to prompt the occurrence of an abnormality in the manhole cover.
Specifically, the invention relates to a multi-vector self-learning well lid posture monitoring method, which can be used for monitoring the well lid state on a well lid state monitor, collecting posture, acceleration, geographic coordinates and temperature data of the well lid, converting the data by a processor, processing the data through a multi-vector self-learning posture algorithm to obtain well lid state information, transmitting the information to a remote monitoring platform by utilizing a narrow-band honeycomb Internet of things technology, and realizing edge calculation and intelligent well lid alarm.
In an optional implementation manner of the first aspect of the present invention, before the collecting, in real time, attitude data of the manhole cover body includes:
collecting normal posture data of the well lid body when a vehicle or a person passes through the well lid body on a daily basis;
and uniformly dividing the measuring range of the attitude sensor into different intervals, counting the frequency value, the occurrence time and the duration of the normal attitude data in each interval, and generating an attitude database.
In this embodiment, before the well lid posture monitoring method is used, a large amount of daily well lid body posture data are collected for subsequent well lid posture recognition and judgment, the occurrence frequency, occurrence time and duration of each occurrence of the posture data are counted, the positive abnormality of the posture data is judged through a plurality of conditions, and the judgment is more accurate.
In an optional embodiment of the first aspect of the present invention, the dividing the range of the attitude sensor into different intervals includes:
dividing an X-direction acceleration range, a Y-direction acceleration range and a Z-direction acceleration range of the attitude sensor into 10 sections equally;
dividing the X-direction angular velocity range, the Y-direction angular velocity range and the Z-direction angular velocity range of the attitude sensor into 10 sections equally;
and dividing the inclination angle range of the attitude sensor into 10 sections.
In this embodiment, an exemplary method of span division according to the present invention, as shown in the following table,
taking the X-direction acceleration range as an example, the X-direction acceleration range which can be measured by the sensor is divided into 10 sections, and initially, each section corresponds to a frequency value, the default value is 1, the maximum value is 2, and the minimum value is 0. The vehicle and the person pass through the well cover to cause the vibration of the well cover, and repeatedly, the acceleration value in the X direction falls in a plurality of sections, the frequency of falling in the sections is different, some adjustment can be carried out on the frequency value, the section with higher frequency value indicates that the occurrence number is relatively large, and the event is considered to be in a normal state when the event occurs in the section again; the interval with a low frequency value indicates a relatively small number of occurrences, and when an event occurs again in such an interval, it is regarded as an "abnormal state".
In an alternative embodiment of the first aspect of the present invention, when the value of the X-direction acceleration, the Y-direction acceleration, the Z-direction acceleration, the X-direction angular velocity, the Y-direction angular velocity, the Z-direction angular velocity, or the inclination angle falls in 1 section thereof, the frequency value of the section increases by a fixed value Δv, and the frequency values of the remaining sections decrease by Δv/(10-1).
In an optional implementation manner of the first aspect of the present invention, the comparing the gesture data with normal gesture data in a gesture database, and determining whether the gesture data is normal according to a frequency, a time and a duration of occurrence of the gesture data includes:
calculating the frequency similarity, occurrence time similarity and duration time similarity of the gesture data and normal gesture data in a gesture database;
weighting and calculating the frequency similarity, the occurrence time similarity and the duration time similarity to obtain total similarity;
judging whether 100% -total similarity is larger than a preset threshold value;
and if the 100% -total similarity is greater than a preset threshold, judging that the gesture data is abnormal.
In this embodiment, when multiple types of events in the gesture database occur simultaneously, similarity needs to be calculated, and whether an abnormal action occurs in the well lid is comprehensively determined;
frequency value similarityWherein FAn is the frequency value corresponding to the interval where the acceleration in the X direction is located, FBn is the frequency value corresponding to the interval where the acceleration in the Y direction is located, FCn is the frequency value corresponding to the interval where the acceleration in the Z direction is located, FDn is the frequency value corresponding to the interval where the angular velocity in the X direction is located, FEn is the frequency value corresponding to the interval where the angular velocity in the Y direction is located, FFn is the frequency value corresponding to the interval where the angular velocity in the Z direction is located, FGn is the frequency value corresponding to the interval where the inclination is located, n refers to the current occurrence interval, n is [1, 10]]。
And (5) calculating the occurrence time similarity:
the current occurrence time and the last occurrence time of the event are the same, the similarity is the highest, and the maximum value is 1; the similarity is the lowest, and the minimum value is 0, with the last time 12 hours later or 12 hours earlier.
The event occurrence time difference is delta t F
Δt F =|GXn This time -GXn Last time |
Wherein X represents the type of the event occurring at this time, X is { A, B, C, D, E, F, G }; n refers to the occurrence interval of the event, n is [1, 10].
Time of occurrence similarityWherein Δt is F For the occurrence time difference delta t between the currently measured gesture data and the last measured gesture data F ∈[0,24];
Duration similarity calculation:
the duration of the event is the same as the last duration, the similarity is the highest, and the maximum value is 1; the similarity is the lowest with the last duration being 500% larger or 500% smaller, and is the minimum value of 0.
The event duration difference is Δt H
Δt H =|HXn This time -HXn Last time |
Wherein X represents the type of the event occurring at this time, X is { A, B, C, D, E, F, G }; n refers to the occurrence interval of the event, n is [1, 10].
Duration similarityWherein Δt is H For the duration difference of the currently measured pose data and the last measured pose data HXn Last time For the duration of the last measured pose data.
Total similarity:
S=F S ×0.7+G S ×0.05+H S ×0.25
in the whole system, the similarity calculated by the acceleration and the angular velocity acquired by the attitude sensor is more reliable, the weight is higher, and the weight value is 0.7; similarity reliability calculated by duration is inferior, and weight is 0.25; and through the similarity calculated by the occurrence time, the reliability is very low because the occurrence time of the event is larger by accident, and the weight value is 0.05.
The total similarity indicates the similarity between the occurrence of the current time and the occurrence of the event in the past, and the lower the similarity is, the more likely the event is an abnormal event, and the more the abnormal alarm needs to be reported.
Setting a required alarm threshold P, and then:
100% -S > P, the well lid is abnormal, and the well lid abnormal alarm needs to be reported
100% -S is less than or equal to P, the well lid is normal, and an alarm does not need to be reported.
In an optional implementation manner of the first aspect of the present invention, after the collecting the posture data of the manhole cover body in real time, before comparing the posture data with the normal posture data in the posture database, the method includes:
and carrying out data noise reduction and data fusion on the gesture data.
In an optional implementation manner of the first aspect of the present invention, the performing data denoising and data fusion on the gesture data includes:
firstly, carrying out data denoising on the gesture data, and then carrying out data fusion on the gesture data after denoising by a quaternion method to obtain gesture information.
In the embodiment, data fusion and gesture calculation are performed by using quaternions, and the quaternion has obvious advantages compared with other gesture algorithms: solving the quaternion differential equation solves four differential equations. Although one equation is more than the Euler differential equation, the method has the advantages of small calculated amount, high precision and avoidance of singularities, and is one of the key points of the current research; the directional cosine method produces skew, scale, drift errors, etc. when solving for the carrier pose dynamics, however, it is important to estimate these errors when solving for the pose. Compared with the direction cosine method, the quaternion method has the advantages that not only is the skew error equal to zero, but also the derivation of the scale error is very simple, an analytical expression which is convenient for further analysis can be obtained, and the direction cosine method can only analyze and monitor the scale error under special conditions and cannot obtain a general conclusion; the euler angle method, the direction cosine method and the quaternion method are compared from different angles. The results show that the quaternion method has the best performance.
Referring to fig. 2, a second aspect of the present invention provides a manhole cover posture monitoring system, the manhole cover posture monitoring system includes a manhole cover and a server 10, the manhole cover and the server are in communication connection through a base station 20, the manhole cover includes a manhole cover body, and a posture sensor 30, a memory 40, a processor 50 and a communication module 60 disposed on the manhole cover body, the posture sensor 30, the memory 40 and the communication module 60 are all connected to the processor 50, the posture sensor 30 is used for collecting posture data of the manhole cover body, the memory 40 is used for storing the posture data and storing instructions of the manhole cover posture monitoring method according to any one of the above, and the processor 50 invokes the instructions in the memory 40 and executes the instructions.
Specifically, when the well lid posture monitoring system is operated, firstly, a sensor collects data, then a processor carries out data noise reduction and data fusion, then the processor automatically learns and distinguishes vibration and shake conditions of the well lid through the multi-vector self-learning posture algorithm, dynamically calculates and adjusts related parameters, realizes accurate identification and intelligent alarming of well lid abnormal movement, and after judging, the well lid can be networked with a server to push data such as well lid posture, acceleration, geographic coordinates, temperature and posture library to the server, so that the well lid state is truly reflected to a remote server, the posture library can also present historical state characteristics of the well lid for data analysis, and the parameters of the well lid are commands issued by the processing server to modify operation parameters of equipment, including functions such as posture library correction, posture library updating and alarm threshold modification.
Referring to fig. 2, in an alternative embodiment of the second aspect of the present invention, the manhole cover further comprises a GNSS sensor 70 disposed on the manhole cover body, and the GNSS sensor 70 is connected to the processor 50. The GNSS sensor is used for acquiring longitude and latitude information of the well lid body.
Referring to fig. 2, in an alternative embodiment of the second aspect of the present invention, the manhole cover further comprises a temperature sensor 80 disposed on the manhole cover body, and the temperature sensor 80 is connected to the processor 50. The temperature sensor user acquires the temperature of well lid, whether the temperature change condition in the detection well lid of being convenient for.
In summary, the invention provides a method and a system for monitoring the posture of a well lid by self-learning, wherein the method for monitoring the posture of the well lid comprises the following steps: acquiring attitude data of the well lid body in real time; comparing the gesture data with normal gesture data in a gesture database, and judging whether the gesture data are normal or not according to the occurrence frequency, occurrence time and duration of the gesture data; when the gesture data are normal, updating the gesture data into the gesture database; and when the gesture data is abnormal, reporting alarm information of the abnormal well lid to a server. The well lid posture monitoring method can autonomously distinguish whether the data of the well lid body is abnormal or not, and can autonomously update the posture database, so that the sensitivity of monitoring is dynamically adjusted, and the intelligent degree is higher.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (6)
1. The self-learning well lid posture monitoring method is characterized by comprising the following steps of:
acquiring attitude data of the well lid body in real time;
comparing the gesture data with normal gesture data in a gesture database, and judging whether the gesture data are normal or not according to the occurrence frequency, occurrence time and duration of the gesture data;
when the gesture data are normal, updating the gesture data into the gesture database;
reporting alarm information of well lid abnormality to a server when the gesture data is abnormal;
before the attitude data of the well lid body are collected in real time, the method comprises the following steps:
collecting normal posture data of the well lid body when a vehicle or a person passes through the well lid body on a daily basis;
dividing the measuring range of the attitude sensor into different intervals, counting the frequency value, the occurrence time and the duration of the normal attitude data in each interval, and generating an attitude database;
the step of uniformly dividing the measuring range of the attitude sensor into different intervals comprises the following steps:
dividing an X-direction acceleration range, a Y-direction acceleration range and a Z-direction acceleration range of the attitude sensor into 10 sections equally;
dividing the X-direction angular velocity range, the Y-direction angular velocity range and the Z-direction angular velocity range of the attitude sensor into 10 sections equally;
dividing the inclination angle range of the attitude sensor into 10 sections;
when the values of the X-direction acceleration, the Y-direction acceleration, the Z-direction acceleration, the X-direction angular velocity, the Y-direction angular velocity, the Z-direction angular velocity or the inclination angle fall in 1 specific section in 10 sections corresponding to each other, the frequency value of the specific section is increased by a fixed value DeltaV, and the frequency value of the rest sections is reduced by DeltaV/(10-1);
comparing the gesture data with normal gesture data in a gesture database, and judging whether the gesture data is normal according to the occurrence frequency, occurrence time and duration of the gesture data comprises:
calculating the frequency similarity, occurrence time similarity and duration time similarity of the gesture data and normal gesture data in a gesture database;
weighting and calculating the frequency similarity, the occurrence time similarity and the duration time similarity to obtain total similarity;
judging whether 100% -total similarity is larger than a preset threshold value;
if the 100% -total similarity is greater than a preset threshold, judging that the gesture data is abnormal;
the saidFrequency similarityWherein FAn is the frequency value corresponding to the interval where the acceleration in the X direction is located, FBn is the frequency value corresponding to the interval where the acceleration in the Y direction is located, FCn is the frequency value corresponding to the interval where the acceleration in the Z direction is located, FDn is the frequency value corresponding to the interval where the angular velocity in the X direction is located, FEn is the frequency value corresponding to the interval where the angular velocity in the Y direction is located, FFn is the frequency value corresponding to the interval where the angular velocity in the Z direction is located, FGn is the frequency value corresponding to the interval where the inclination is located, n refers to the current occurrence interval, n is [1, 10]];
The occurrence time similarityWherein Δt is F For the occurrence time difference delta t between the currently measured gesture data and the last measured gesture data F ∈[0,24];
The duration similarityWherein Δt is H For the duration difference of the currently measured posture data and the last measured posture data +.>A duration of the gesture data measured last time;
the calculation formula of the total similarity is as follows:the total similarity represents the similarity of the current occurrence event and the past occurrence event.
2. The well lid posture monitoring method according to claim 1, wherein after the posture data of the well lid body is collected in real time, before the comparing the posture data with the normal posture data in the posture database, the method comprises:
and carrying out data noise reduction and data fusion on the gesture data.
3. The well lid pose monitoring method according to claim 2, wherein the performing data denoising and data fusion on the pose data comprises:
firstly, carrying out data denoising on the gesture data, and then carrying out data fusion on the gesture data after denoising by a quaternion method to obtain gesture information.
4. The utility model provides a well lid gesture monitoring system, its characterized in that, well lid gesture monitoring system includes well lid and server, the well lid with the server passes through the basic station communication and connects, the well lid includes the well lid body and sets up attitude sensor, memory, treater and the communication module on the well lid body, attitude sensor, memory and the communication module all with the treater links to each other, attitude sensor is used for gathering the gesture data of well lid body, the memory is used for storing the gesture data and the instruction of the well lid gesture monitoring method of storage in any one of claims 1-3, the treater is called the instruction in the memory and is carried out.
5. The well lid attitude monitoring system of claim 4, wherein the well lid further comprises a GNSS sensor disposed on the well lid body, the GNSS sensor being coupled to the processor.
6. The well lid posture monitoring system of claim 4, wherein the well lid further comprises a temperature sensor disposed on the well lid body, the temperature sensor being coupled to the processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110485720.1A CN113137983B (en) | 2021-04-30 | 2021-04-30 | Self-learning well lid posture monitoring method and monitoring system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110485720.1A CN113137983B (en) | 2021-04-30 | 2021-04-30 | Self-learning well lid posture monitoring method and monitoring system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113137983A CN113137983A (en) | 2021-07-20 |
CN113137983B true CN113137983B (en) | 2023-08-22 |
Family
ID=76816817
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110485720.1A Active CN113137983B (en) | 2021-04-30 | 2021-04-30 | Self-learning well lid posture monitoring method and monitoring system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113137983B (en) |
Citations (31)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001280970A (en) * | 2000-03-28 | 2001-10-10 | Japan Aviation Electronics Industry Ltd | Attitude measuring instrument |
CN103021136A (en) * | 2012-12-24 | 2013-04-03 | 电子科技大学 | System and method for monitoring swimming pool drowning based on Internet of Things |
CN105913592A (en) * | 2016-06-30 | 2016-08-31 | 上海路辉电子科技有限公司 | Well lid, well lid state alarm monitoring method and well lid state alarm monitoring system |
CN107644229A (en) * | 2017-08-31 | 2018-01-30 | 浙江工业大学 | Indoor anomaly detection method based on multi-sensor information fusion |
CN108182397A (en) * | 2017-12-26 | 2018-06-19 | 王华锋 | A kind of multiple dimensioned face verification method of multi-pose |
CN108446733A (en) * | 2018-03-19 | 2018-08-24 | 上海玺翎智能科技有限公司 | A kind of human body behavior monitoring and intelligent identification Method based on multi-sensor data |
CN108520755A (en) * | 2018-04-16 | 2018-09-11 | 吴思 | A kind of detection method and device |
JP2018159569A (en) * | 2017-03-22 | 2018-10-11 | カシオ計算機株式会社 | Attitude calculation device, attitude instrumentation system and attitude calculation method |
CN109145739A (en) * | 2018-07-18 | 2019-01-04 | 郭渐凌 | A kind of human body gesture prediction method, apparatus and system |
CN109341776A (en) * | 2018-11-21 | 2019-02-15 | 戴天智能科技(上海)有限公司 | A kind of well lid electronic lock monitoring module system |
CN109919141A (en) * | 2019-04-09 | 2019-06-21 | 广东省智能制造研究所 | A kind of recognition methods again of the pedestrian based on skeleton pose |
RO133583A2 (en) * | 2018-02-27 | 2019-08-30 | Securifai S.R.L. | Method and algorithm based on motion gradients and convolutional neural networks for automatic detection and localization of abnormal video events |
CN110245707A (en) * | 2019-06-17 | 2019-09-17 | 吉林大学 | Human body walking posture vibration information recognition methods and system based on scorpion positioning |
CN110619276A (en) * | 2019-08-15 | 2019-12-27 | 杭州电子科技大学 | Anomaly and violence detection system and method based on unmanned aerial vehicle mobile monitoring |
CN110657839A (en) * | 2019-10-12 | 2020-01-07 | 沈阳航空航天大学 | Well lid safety monitoring system |
CN110908298A (en) * | 2019-10-17 | 2020-03-24 | 苏州卷柏电子有限公司 | Intelligent control software for intelligent well lid and related algorithm thereof |
CN110962591A (en) * | 2019-11-29 | 2020-04-07 | 浙江大华技术股份有限公司 | Fatigue driving detection method, device, computer equipment and storage medium |
CN110974641A (en) * | 2019-12-24 | 2020-04-10 | 中南民族大学 | Intelligent walking stick system integrating machine learning and Internet of things technology for blind people |
CN210533430U (en) * | 2019-10-12 | 2020-05-15 | 沈阳航空航天大学 | Well lid safety monitoring system |
CN111339905A (en) * | 2020-02-22 | 2020-06-26 | 郑州铁路职业技术学院 | CIM well lid state visual detection system based on deep learning and multi-view angle |
CN111344762A (en) * | 2017-11-21 | 2020-06-26 | 三菱电机株式会社 | Abnormality detection device and abnormality detection method |
CN211044491U (en) * | 2020-02-28 | 2020-07-17 | 四川大学 | City intelligence well lid |
CN111623869A (en) * | 2020-05-20 | 2020-09-04 | 北京必创科技股份有限公司 | Data processing method based on edge calculation and data monitoring and edge calculation device |
CN111989198A (en) * | 2018-03-29 | 2020-11-24 | 日产自动车株式会社 | Abnormality detection device and abnormality detection method |
CN111982192A (en) * | 2020-08-11 | 2020-11-24 | 国家电网有限公司技术学院分公司 | Cable fault self-learning identification method and system based on Internet of things |
CN112085003A (en) * | 2020-09-24 | 2020-12-15 | 湖北科技学院 | Automatic identification method and device for abnormal behaviors in public places and camera equipment |
CN112350974A (en) * | 2019-08-07 | 2021-02-09 | 中国移动通信集团广东有限公司 | Safety monitoring method and device of Internet of things and electronic equipment |
CN112464808A (en) * | 2020-11-26 | 2021-03-09 | 成都睿码科技有限责任公司 | Rope skipping posture and number identification method based on computer vision |
CN112492539A (en) * | 2020-11-27 | 2021-03-12 | 江西中科智慧水产业研究股份有限公司 | Wireless terminal with self-learning capability for monitoring inspection well cover in real time and monitoring method thereof |
CN112669570A (en) * | 2021-01-06 | 2021-04-16 | 常州百芝龙智慧科技有限公司 | Habit-based self-learning whole-house abnormity monitoring equipment |
CN212983949U (en) * | 2020-06-19 | 2021-04-16 | 江苏普瑞德智能科技有限公司 | Intelligent well lid monitoring terminal |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2014089113A (en) * | 2012-10-30 | 2014-05-15 | Yamaha Corp | Posture estimation device and program |
US10270788B2 (en) * | 2016-06-06 | 2019-04-23 | Netskope, Inc. | Machine learning based anomaly detection |
-
2021
- 2021-04-30 CN CN202110485720.1A patent/CN113137983B/en active Active
Patent Citations (31)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001280970A (en) * | 2000-03-28 | 2001-10-10 | Japan Aviation Electronics Industry Ltd | Attitude measuring instrument |
CN103021136A (en) * | 2012-12-24 | 2013-04-03 | 电子科技大学 | System and method for monitoring swimming pool drowning based on Internet of Things |
CN105913592A (en) * | 2016-06-30 | 2016-08-31 | 上海路辉电子科技有限公司 | Well lid, well lid state alarm monitoring method and well lid state alarm monitoring system |
JP2018159569A (en) * | 2017-03-22 | 2018-10-11 | カシオ計算機株式会社 | Attitude calculation device, attitude instrumentation system and attitude calculation method |
CN107644229A (en) * | 2017-08-31 | 2018-01-30 | 浙江工业大学 | Indoor anomaly detection method based on multi-sensor information fusion |
CN111344762A (en) * | 2017-11-21 | 2020-06-26 | 三菱电机株式会社 | Abnormality detection device and abnormality detection method |
CN108182397A (en) * | 2017-12-26 | 2018-06-19 | 王华锋 | A kind of multiple dimensioned face verification method of multi-pose |
RO133583A2 (en) * | 2018-02-27 | 2019-08-30 | Securifai S.R.L. | Method and algorithm based on motion gradients and convolutional neural networks for automatic detection and localization of abnormal video events |
CN108446733A (en) * | 2018-03-19 | 2018-08-24 | 上海玺翎智能科技有限公司 | A kind of human body behavior monitoring and intelligent identification Method based on multi-sensor data |
CN111989198A (en) * | 2018-03-29 | 2020-11-24 | 日产自动车株式会社 | Abnormality detection device and abnormality detection method |
CN108520755A (en) * | 2018-04-16 | 2018-09-11 | 吴思 | A kind of detection method and device |
CN109145739A (en) * | 2018-07-18 | 2019-01-04 | 郭渐凌 | A kind of human body gesture prediction method, apparatus and system |
CN109341776A (en) * | 2018-11-21 | 2019-02-15 | 戴天智能科技(上海)有限公司 | A kind of well lid electronic lock monitoring module system |
CN109919141A (en) * | 2019-04-09 | 2019-06-21 | 广东省智能制造研究所 | A kind of recognition methods again of the pedestrian based on skeleton pose |
CN110245707A (en) * | 2019-06-17 | 2019-09-17 | 吉林大学 | Human body walking posture vibration information recognition methods and system based on scorpion positioning |
CN112350974A (en) * | 2019-08-07 | 2021-02-09 | 中国移动通信集团广东有限公司 | Safety monitoring method and device of Internet of things and electronic equipment |
CN110619276A (en) * | 2019-08-15 | 2019-12-27 | 杭州电子科技大学 | Anomaly and violence detection system and method based on unmanned aerial vehicle mobile monitoring |
CN110657839A (en) * | 2019-10-12 | 2020-01-07 | 沈阳航空航天大学 | Well lid safety monitoring system |
CN210533430U (en) * | 2019-10-12 | 2020-05-15 | 沈阳航空航天大学 | Well lid safety monitoring system |
CN110908298A (en) * | 2019-10-17 | 2020-03-24 | 苏州卷柏电子有限公司 | Intelligent control software for intelligent well lid and related algorithm thereof |
CN110962591A (en) * | 2019-11-29 | 2020-04-07 | 浙江大华技术股份有限公司 | Fatigue driving detection method, device, computer equipment and storage medium |
CN110974641A (en) * | 2019-12-24 | 2020-04-10 | 中南民族大学 | Intelligent walking stick system integrating machine learning and Internet of things technology for blind people |
CN111339905A (en) * | 2020-02-22 | 2020-06-26 | 郑州铁路职业技术学院 | CIM well lid state visual detection system based on deep learning and multi-view angle |
CN211044491U (en) * | 2020-02-28 | 2020-07-17 | 四川大学 | City intelligence well lid |
CN111623869A (en) * | 2020-05-20 | 2020-09-04 | 北京必创科技股份有限公司 | Data processing method based on edge calculation and data monitoring and edge calculation device |
CN212983949U (en) * | 2020-06-19 | 2021-04-16 | 江苏普瑞德智能科技有限公司 | Intelligent well lid monitoring terminal |
CN111982192A (en) * | 2020-08-11 | 2020-11-24 | 国家电网有限公司技术学院分公司 | Cable fault self-learning identification method and system based on Internet of things |
CN112085003A (en) * | 2020-09-24 | 2020-12-15 | 湖北科技学院 | Automatic identification method and device for abnormal behaviors in public places and camera equipment |
CN112464808A (en) * | 2020-11-26 | 2021-03-09 | 成都睿码科技有限责任公司 | Rope skipping posture and number identification method based on computer vision |
CN112492539A (en) * | 2020-11-27 | 2021-03-12 | 江西中科智慧水产业研究股份有限公司 | Wireless terminal with self-learning capability for monitoring inspection well cover in real time and monitoring method thereof |
CN112669570A (en) * | 2021-01-06 | 2021-04-16 | 常州百芝龙智慧科技有限公司 | Habit-based self-learning whole-house abnormity monitoring equipment |
Non-Patent Citations (1)
Title |
---|
宗立波 .基于RGBD姿态识别的篮球运动员动作规范系统设计与实现.中国优秀硕士学位论文全文数据库 社会科学II辑.2021,(第3期),全文. * |
Also Published As
Publication number | Publication date |
---|---|
CN113137983A (en) | 2021-07-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220205956A1 (en) | Detection of structural anomalies in a pipeline network | |
CN107103775B (en) | Road quality detection method based on crowd-sourcing calculation | |
CN116105802B (en) | Underground facility safety monitoring and early warning method based on Internet of things | |
CN115096373A (en) | Bridge engineering health detection system based on sensor | |
CN109141528A (en) | A kind of urban track traffic civil engineering facility intelligent real-time monitoring system | |
Akhavian et al. | Productivity analysis of construction worker activities using smartphone sensors | |
CN104408578A (en) | Track-point-based quantitative assessment system and method for mechanical operation | |
CN109541022A (en) | A kind of bridge structure crack health monitoring analysis method | |
CN115165725A (en) | Data-driven marine equipment corrosion monitoring and safety early warning system | |
CN116908942A (en) | Urban waterlogging warning device and method based on rainfall forecast | |
CN116522270A (en) | Data processing system for smart sponge city | |
CN113137983B (en) | Self-learning well lid posture monitoring method and monitoring system | |
CN117037417B (en) | Intelligent identification pile fusion device | |
CN113128709A (en) | Water supply pipe network leakage management system | |
Tiwari et al. | Time-series prediction and forecasting of ambient noise levels using deep learning and machine learning techniques | |
CN117251722A (en) | Intelligent traffic management system based on big data | |
CN116758762A (en) | Control method based on big data | |
CN115131186A (en) | Urban municipal drainage safety monitoring analysis cloud platform based on artificial intelligence | |
Finogeev et al. | Proactive big data analysis for traffic accident prediction | |
CN117291430B (en) | Safety production detection method and device based on machine vision | |
KR20200144393A (en) | Rainfall measuring method using artificial intelligence | |
Dimas | Assessment on road anomalies using smartphone sensor: A review | |
CN112511979B (en) | Intelligent file bag management system for autonomous key file transportation | |
CN116560219B (en) | Self-adaptive monitoring control method and system based on transmission tower wind speed joint analysis | |
CN116704734B (en) | Monitoring and early warning method and system for preventing underground pipeline from being excavated based on Internet of things technology |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |