CN109543607A - Object abnormal state detection method, system, monitor system and storage medium - Google Patents
Object abnormal state detection method, system, monitor system and storage medium Download PDFInfo
- Publication number
- CN109543607A CN109543607A CN201811397468.3A CN201811397468A CN109543607A CN 109543607 A CN109543607 A CN 109543607A CN 201811397468 A CN201811397468 A CN 201811397468A CN 109543607 A CN109543607 A CN 109543607A
- Authority
- CN
- China
- Prior art keywords
- image
- detected
- abnormal state
- state detection
- window
- 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
Classifications
-
- 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
- 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
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
Abstract
This application discloses a kind of object abnormal state detection methods, system, monitoring system and storage medium, the disaggregated model used in the image analysis process of the object abnormal state detection method is the artificial neural network after single-frame images training, since the training sample is single-frame images, it is easy to obtain and mark, so that being readily apparent a large amount of training sample in the training process of disaggregated model, and the training of a large amount of training sample is so that the classification accuracy of the disaggregated model greatly promotes, it is advantageously implemented the accuracy of detection of the object abnormal state detection method;Also due to the procurement cost of the training sample of the training pattern is lower, the overall cost of the object abnormal state detection method is advantageously reduced.
Description
Technical field
This application involves machine learning techniques field, more specifically to a kind of object abnormal state detection method,
System, monitor system and storage medium.
Background technique
With the continuous propulsion of social senilization's process, occur more and more " empty nest " old men in society, i.e., due to
The caused old man's most of time of situations such as young man's most of time works outside or young man does not live together with old man alone one
People's the case where living of being in is more and more common.And how to find the abnormality that these old men occur in life in time becomes urgently
Problem to be solved.
Since some old man's legs and feet are inconvenient or may suffer from some diseases and it is likely to occur and falls down suddenly, and can not stand
Situation, this is breakneck for solitary old man.The existing detection of falling down for old man in family is mainly benefit
For the data of the sensor acquisition for the equipment dressed with old man come what is realized, this detection mode of falling down needs old man to dress this at the moment
A little equipment are just able to achieve, and some old men often forget about and dress these equipment, this makes existing detection method of falling down be difficult to reality
Current moment falls down detection.
Summary of the invention
In order to solve the above technical problems, this application provides a kind of object abnormal state detection method, system, monitoring systems
System and storage medium, to detect at the time of realization to object abnormality, and view-based access control model analytical technology is realized, is not necessarily to
Corresponding sensor is set on object, increases the convenience of object abnormal state detection.
To realize the above-mentioned technical purpose, the embodiment of the present application provides following technical solution:
A kind of object abnormal state detection method, comprising:
Image to be detected is obtained, described image to be detected is the single-frame images comprising object state;
The characteristic image of described image to be detected is extracted using the first artificial neural network;
Processing is carried out to the characteristic image and obtains candidate region;
The candidate region is inputted in disaggregated model, it is whether abnormal comprising object in described image to be detected to judge
State;
The disaggregated model be by training sample training after the second artificial neural network, the training sample be comprising
The single-frame images of object state.
Optionally, the disaggregated model judge in described image to be detected whether include object abnormality process packet
It includes:
The candidate window in the candidate region is extracted, the candidate window includes foreground window and backdrop window;
According to the foreground window, transformation parameter is obtained using frame regression algorithm;
The foreground window is handled according to the transformation parameter, to obtain target prospect image;
Classify to the target prospect image, judges whether the target prospect image is exception class image, if
It is then to determine in described image to be detected comprising object abnormality.
It is optionally, described to determine in described image to be detected comprising after object abnormality, further includes:
Warning message is generated, and the warning message is sent with predetermined manner.
Optionally, the warning message includes determining whether as image to be detected comprising object abnormality.
Optionally, described acquisition image to be detected includes:
Obtain video to be detected;
The video frame for extracting the video to be detected, using video frame described in each frame as the image to be detected.
A kind of object abnormal state detection system, comprising:
Image collection module, for obtaining image to be detected, described image to be detected is the single frames comprising object state
Image;
Characteristic extracting module, for extracting the characteristic image of described image to be detected using the first artificial neural network;
Processing module obtains candidate region for carrying out processing to the characteristic image;
Categorization module, for the candidate region to be inputted in disaggregated model, with judge in described image to be detected whether
Include object abnormality;
The disaggregated model be by training sample training after the second artificial neural network, the training sample be comprising
The single-frame images of object state.
Optionally, the disaggregated model judge in described image to be detected whether include object abnormality process packet
It includes:
The candidate window in the candidate region is extracted, the candidate window includes foreground window and backdrop window;
According to the foreground window, transformation parameter is obtained using frame regression algorithm;
The foreground window is handled according to the transformation parameter, to obtain target prospect image;
Classify to the target prospect image, judges whether the target prospect image is exception class image, if
It is then to determine in described image to be detected comprising object abnormality.
Optionally, further includes: alarm message module;
The categorization module is also used in determining described image to be detected comprising triggering institute after object abnormality
State alarm message module;
The alarm message module sends the warning message for generating warning message, and with predetermined manner.
Optionally, the warning message includes determining whether as image to be detected comprising object abnormality.
Optionally, described image acquisition module includes:
Video acquisition unit, for obtaining video to be detected;
Frame extraction unit, for extracting the video frame of the video to be detected, using video frame described in each frame as one
Described image to be detected.
A kind of monitor system, including image acquisition equipment and object abnormal state detection system, the object are abnormal
Condition detecting system is object abnormal state detection system as described in any one of the above embodiments;
Described image obtains equipment and is used for monitoring objective space, and sends image to be detected that monitoring generates to the mesh
Mark object abnormal state detection system.
Optionally, when the object abnormal state detection system further includes alarm message module, further includes:
Communication equipment, the warning message sent for receiving the warning message.
A kind of storage medium is stored with program code on the storage medium, and said program code is performed in realization
State described in any item object abnormal state detection methods.
It can be seen from the above technical proposal that the embodiment of the present application provide a kind of object abnormal state detection method,
System, monitoring system and storage medium, the object abnormal state detection method is based on image analysis and machine learning is real
Now to the detection of object abnormality, it is only necessary to obtain the single-frame images comprising object state and be handled i.e. it
Can, it is not necessary that the equipment such as sensor are arranged on object, avoiding can not be detected when not having sensor on object
Purpose increases the convenience of object abnormal state detection.
In addition, the disaggregated model used in the image analysis process of the object abnormal state detection method is by single
Artificial neural network after the training of frame image is easy to obtain and mark since the training sample is single-frame images, so that classification
A large amount of training sample is readily apparent in the training process of model, and the training of a large amount of training sample is so that the classification
The classification accuracy of model greatly promotes, and is advantageously implemented the accuracy of detection of the object abnormal state detection method;Together
Sample advantageously reduces the object abnormal state detection side since the procurement cost of the training sample of the training pattern is lower
The overall cost of method.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of application for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of flow diagram for object abnormal state detection method that one embodiment of the application provides;
Fig. 2-Fig. 6 is the schematic diagram for some object states that one embodiment of the application provides;
Fig. 7 is a kind of process signal for object abnormal state detection method that another embodiment of the application provides
Figure;
Fig. 8 is a kind of process signal for object abnormal state detection method that another embodiment of the application provides
Figure;
Fig. 9 is a kind of process signal for object abnormal state detection method that the further embodiment of the application provides
Figure.
Specific embodiment
As described in background, sensor-based object method for detecting abnormality is difficult to realize the moment in the prior art
Detection.
In addition, inventor proposes a kind of object method for detecting abnormality of view-based access control model the study found that having in document, it should
Method mainly passes through the video that analysis includes object state generating process, to realize the process that object abnormality occurs
Identification.But include the video that object abnormality occurs acquisition cost is high, acquisition difficulty is big, it is difficult to accomplish a large amount of
Video is as training sample, and the labeling process difficulty of video is also higher, so that the instruction of disaggregated model used in this method
White silk is very difficult, and due to the negligible amounts of training sample, so that the classification accuracy of the disaggregated model is poor.
In view of this, the embodiment of the present application provides a kind of object abnormal state detection method, comprising:
Image to be detected is obtained, described image to be detected is the single-frame images comprising object state;
The characteristic image of described image to be detected is extracted using the first artificial neural network;
Processing is carried out to the characteristic image and obtains candidate region;
The candidate region is inputted in disaggregated model, it is whether abnormal comprising object in described image to be detected to judge
State;
The disaggregated model be by training sample training after the second artificial neural network, the training sample be comprising
The single-frame images of object state.
The object abnormal state detection method is based on image analysis and machine learning is realized to object exception shape
The detection of state, it is only necessary to obtain the single-frame images comprising object state and it is handled, without on object
The equipment such as sensor are set, avoids the purpose that can not be detected when not having sensor on object, increases object
The convenience of abnormal state detection.
In addition, the disaggregated model used in the image analysis process of the object abnormal state detection method is by single
Artificial neural network after the training of frame image is easy to obtain and mark since the training sample is single-frame images, so that classification
A large amount of training sample is readily apparent in the training process of model, and the training of a large amount of training sample is so that the classification
The classification accuracy of model greatly promotes, and is advantageously implemented the accuracy of detection of the object abnormal state detection method;Together
Sample advantageously reduces the object abnormal state detection side since the procurement cost of the training sample of the training pattern is lower
The overall cost of method.
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on
Embodiment in the application, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall in the protection scope of this application.
The embodiment of the present application provides a kind of object abnormal state detection method, as shown in Figure 1, comprising:
S101: obtaining image to be detected, and described image to be detected is the single-frame images comprising object state;
S102: the characteristic image of described image to be detected is extracted using the first artificial neural network;
S103: processing is carried out to the characteristic image and obtains candidate region;
Whether S104: the candidate region is inputted in disaggregated model, to judge in described image to be detected comprising target
Object abnormality;
The disaggregated model be by training sample training after the second artificial neural network, the training sample be comprising
The single-frame images of object state.
It should be noted that the object can be people or object, the abnormality of object can be the shape that falls down to the ground of people
The broken or lost condition of state or object;The normal condition of object can be the states such as the seat of people, sleeping or object standing.When described
When object abnormal state detection method is applied to the monitoring of people, in image as shown in Figure 2, Figure 3 and Figure 4 people state (lie prone,
Lie and lean to one side to drop to the ground) it may be considered object abnormality;The state of people in image as shown in Figure 5 and Figure 6 (sits and stands
It is vertical) it may be considered the normal condition of object.Of course, Fig. 3-Fig. 6 be only object be people when several abnormalities and
The explanation of normal condition, the application to this and without limitation, specifically depending on actual conditions.
In the present embodiment, since the training sample of the training disaggregated model is as Fig. 2-is shown in fig. 6 by label
Single-frame images, dropped significantly compared to the acquisition of the video comprising object abnormality generating process and the difficulty of labeling process
It is low, to reduce the training cost of training pattern, the training difficulty of training pattern is also reduced, so that the training pattern can
It is trained with obtaining a large amount of training sample easily, help to obtain the high training pattern of classification accuracy, to promote institute
Object abnormal state detection method is stated for the accuracy of detection of object abnormality.
Optionally, the artificial neural network can be convolutional neural networks (Convoltional Neural
Networks,CNN).Of course, in the other embodiments of the application, the artificial neural network can also be circulation nerve
Network (Recurrent Neural Network, RNN).The application to this and without limitation, specifically depending on actual conditions.Separately
Outside, for first artificial neural network is with function, and it can be described as Area generation network (Region Proposal
Network)。
On the basis of the above embodiments, in one embodiment of the application, as shown in fig. 7, the disaggregated model is sentenced
Whether the process comprising object abnormality includes: in described image to be detected of breaking
S201: obtaining image to be detected, and described image to be detected is the single-frame images comprising object state;
S202: the characteristic image of described image to be detected is extracted using the first artificial neural network;
S203: processing is carried out to the characteristic image and obtains candidate region;
S204: the candidate region is inputted in disaggregated model;
S205: the disaggregated model extracts the candidate window in the candidate region, and the candidate window includes prospect window
Mouth and backdrop window;
S206: according to the foreground window, transformation parameter is obtained using frame regression algorithm;
S207: the foreground window is handled according to the transformation parameter, to obtain target prospect image;
S208: classifying to the target prospect image, judges whether the target prospect image is exception class image,
If it is, determining in described image to be detected comprising object abnormality.
Disaggregated model shown in Fig. 7 judge in described image to be detected whether the process comprising object abnormality be with
It is illustrated for two stage disaggregated model, which can be fast-R-CNN or faster-R-
CNN etc..In some embodiments of the present application, the disaggregated model can also be triphasic disaggregated model, and the application is to institute
The specific type of disaggregated model is stated without limitation, specifically depending on actual conditions.
In the embodiment shown in fig. 7, the disaggregated model has also carried out frame recurrence processing to image to be detected, to obtain
It obtains the target prospect image more close to object region to be analyzed, further increases disaggregated model to figure to be detected
The classification accuracy of picture realizes the purpose for the object state in image to be detected that accurately identifies.
It should be noted that before the foreground window (foreground anchor) mainly contains in candidate window
Scape information, such as when object is people, foreground information is the information of people in candidate window.So correspondingly, the background
Window (background anchor) mainly contains the background information in candidate window.
On the basis of the above embodiments, in another embodiment of the application, as shown in figure 8, the object is different
Often condition detection method includes:
S301: obtaining image to be detected, and described image to be detected is the single-frame images comprising object state;
S302: the characteristic image of described image to be detected is extracted using the first artificial neural network;
S303: processing is carried out to the characteristic image and obtains candidate region;
S304: the candidate region is inputted in disaggregated model;
S305: extracting the candidate window in the candidate region, and the candidate window includes foreground window and backdrop window;
S306: according to the foreground window, transformation parameter is obtained using frame regression algorithm;
S307: the foreground window is handled according to the transformation parameter, to obtain target prospect image;
S308: classifying to the target prospect image, judges whether the target prospect image is exception class image,
If it is, determining to generate warning message, and send comprising object abnormality in described image to be detected with predetermined manner
The warning message;If it is not, then return step S301.
In the present embodiment, it when detecting in described image to be detected comprising object abnormality, generates and with pre-
If mode sends the warning message, there is the purpose of abnormality to play warning related personnel.It is described to be checked when detecting
When not including object abnormality in altimetric image, then the step of obtaining image to be detected is returned to, to next frame figure to be detected
It is detected.
The predetermined manner can be the modes such as short message, mail or program information push.The application is to this and without limitation.
Optionally, the warning message includes determining whether as image to be detected comprising object abnormality, so as to receive
The related personnel of the warning message can further determine the abnormal conditions of appearance.
On the basis of the above embodiments, in another embodiment of the application, as shown in figure 9, the object is different
Often condition detection method includes:
S401: video to be detected is obtained;
S402: extracting the video frame of the video to be detected, described to be detected using video frame described in each frame as one
Image;
S403: the characteristic image of described image to be detected is extracted using the first artificial neural network;
S404: processing is carried out to the characteristic image and obtains candidate region;
S405: the candidate region is inputted in disaggregated model;
S406: extracting the candidate window in the candidate region, and the candidate window includes foreground window and backdrop window;
S407: according to the foreground window, transformation parameter is obtained using frame regression algorithm;
S408: the foreground window is handled according to the transformation parameter, to obtain target prospect image;
S409: classifying to the target prospect image, judges whether the target prospect image is exception class image,
If it is, determining to generate warning message, and send comprising object abnormality in described image to be detected with predetermined manner
The warning message.
In the present embodiment, a kind of process for specifically obtaining image to be detected is proposed, it can pass through picture pick-up device
In the video to be detected obtained, obtained by way of extracting video frame.
Object abnormal state detection system provided by the embodiments of the present application is illustrated below, target described below
Object abnormal state detection system can correspond to each other reference with object described above object abnormal state detection method.
Correspondingly, the embodiment of the present application provides a kind of object abnormal state detection system, comprising:
Image collection module, for obtaining image to be detected, described image to be detected is the single frames comprising object state
Image;
Characteristic extracting module, for extracting the characteristic image of described image to be detected using the first artificial neural network;
Processing module obtains candidate region for carrying out processing to the characteristic image;
Categorization module, for the candidate region to be inputted in disaggregated model, with judge in described image to be detected whether
Include object abnormality;
The disaggregated model be by training sample training after the second artificial neural network, the training sample be comprising
The single-frame images of object state.
Optionally, the disaggregated model judge in described image to be detected whether include object abnormality process packet
It includes:
The candidate window in the candidate region is extracted, the candidate window includes foreground window and backdrop window;
According to the foreground window, transformation parameter is obtained using frame regression algorithm;
The foreground window is handled according to the transformation parameter, to obtain target prospect image;
Classify to the target prospect image, judges whether the target prospect image is exception class image, if
It is then to determine in described image to be detected comprising object abnormality.
Optionally, further includes: alarm message module;
The categorization module is also used in determining described image to be detected comprising triggering institute after object abnormality
State alarm message module;
The alarm message module sends the warning message for generating warning message, and with predetermined manner.
Optionally, the warning message includes determining whether as image to be detected comprising object abnormality.
Optionally, described image acquisition module includes:
Video acquisition unit, for obtaining video to be detected;
Frame extraction unit, for extracting the video frame of the video to be detected, using video frame described in each frame as one
Described image to be detected.
Correspondingly, the embodiment of the present application also provides a kind of monitor system, including image acquisition equipment and object exception
Condition detecting system, the object abnormal state detection system are the object abnormality as described in above-mentioned any embodiment
Detection system;
Described image obtains equipment and is used for monitoring objective space, and sends image to be detected that monitoring generates to the mesh
Mark object abnormal state detection system.
It is different to play detection object for the monitoring of the special populations such as old man, child that the monitor system can play the role of
The purpose of normal state.
Optionally, when the object abnormal state detection system further includes alarm message module, the monitor system
Further include:
Communication equipment, the warning message sent for receiving the warning message.
Correspondingly, it is stored with program code on the storage medium the embodiment of the present application also provides a kind of storage medium,
Said program code, which is performed, realizes object abnormal state detection method described in any of the above-described embodiment.
In conclusion the embodiment of the present application provide a kind of object abnormal state detection method, system, monitoring system and
Storage medium, the object abnormal state detection method is based on image analysis and machine learning is realized to object exception shape
The detection of state, it is only necessary to obtain the single-frame images comprising object state and it is handled, without on object
The equipment such as sensor are set, avoids the purpose that can not be detected when not having sensor on object, increases object
The convenience of abnormal state detection.
In addition, the disaggregated model used in the image analysis process of the object abnormal state detection method is by single
Artificial neural network after the training of frame image is easy to obtain and mark since the training sample is single-frame images, so that classification
A large amount of training sample is readily apparent in the training process of model, and the training of a large amount of training sample is so that the classification
The classification accuracy of model greatly promotes, and is advantageously implemented the accuracy of detection of the object abnormal state detection method;Together
Sample advantageously reduces the object abnormal state detection side since the procurement cost of the training sample of the training pattern is lower
The overall cost of method.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with other
The difference of embodiment, the same or similar parts in each embodiment may refer to each other.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use the application.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the application.Therefore, the application
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (13)
1. a kind of object abnormal state detection method characterized by comprising
Image to be detected is obtained, described image to be detected is the single-frame images comprising object state;
The characteristic image of described image to be detected is extracted using the first artificial neural network;
Processing is carried out to the characteristic image and obtains candidate region;
The candidate region is inputted in disaggregated model, whether to judge in described image to be detected comprising object exception shape
State;
The disaggregated model is the second artificial neural network after training sample training, and the training sample is to include target
The single-frame images of object state.
2. the method according to claim 1, wherein the disaggregated model judge in described image to be detected whether
Process comprising object abnormality includes:
The candidate window in the candidate region is extracted, the candidate window includes foreground window and backdrop window;
According to the foreground window, transformation parameter is obtained using frame regression algorithm;
The foreground window is handled according to the transformation parameter, to obtain target prospect image;
Classify to the target prospect image, judges whether the target prospect image is exception class image, if it is,
Determine in described image to be detected comprising object abnormality.
3. according to the method described in claim 2, it is characterized in that, described determine to include that object is different in described image to be detected
After normal state, further includes:
Warning message is generated, and the warning message is sent with predetermined manner.
4. according to the method described in claim 3, it is characterized in that, the warning message includes determining whether as comprising object exception
Image to be detected of state.
5. the method according to claim 1, wherein described acquisition image to be detected includes:
Obtain video to be detected;
The video frame for extracting the video to be detected, using video frame described in each frame as the image to be detected.
6. a kind of object abnormal state detection system characterized by comprising
Image collection module, for obtaining image to be detected, described image to be detected is the single-frame images comprising object state;
Characteristic extracting module, for extracting the characteristic image of described image to be detected using the first artificial neural network;
Processing module obtains candidate region for carrying out processing to the characteristic image;
Categorization module, for the candidate region to be inputted in disaggregated model, with judge in described image to be detected whether include
Object abnormality;
The disaggregated model is the second artificial neural network after training sample training, and the training sample is to include target
The single-frame images of object state.
7. system according to claim 6, which is characterized in that the disaggregated model judge in described image to be detected whether
Process comprising object abnormality includes:
The candidate window in the candidate region is extracted, the candidate window includes foreground window and backdrop window;
According to the foreground window, transformation parameter is obtained using frame regression algorithm;
The foreground window is handled according to the transformation parameter, to obtain target prospect image;
Classify to the target prospect image, judges whether the target prospect image is exception class image, if it is,
Determine in described image to be detected comprising object abnormality.
8. system according to claim 7, which is characterized in that further include: alarm message module;
The categorization module is also used in determining described image to be detected comprising triggering the report after object abnormality
Alert information module;
The alarm message module sends the warning message for generating warning message, and with predetermined manner.
9. system according to claim 8, which is characterized in that the warning message includes determining whether as comprising object exception
Image to be detected of state.
10. system according to claim 6, which is characterized in that described image obtains module and includes:
Video acquisition unit, for obtaining video to be detected;
Frame extraction unit, for extracting the video frame of the video to be detected, using video frame described in each frame as described in one
Image to be detected.
11. a kind of monitor system, which is characterized in that described including image acquisition equipment and object abnormal state detection system
Object abnormal state detection system is such as the described in any item object abnormal state detection systems of claim 6-10;
Described image obtains equipment and is used for monitoring objective space, and sends image to be detected that monitoring generates to the object
Abnormal state detection system.
12. monitor system according to claim 11, which is characterized in that when the object abnormal state detection system also
When including alarm message module, further includes:
Communication equipment, the warning message sent for receiving the warning message.
13. a kind of storage medium, which is characterized in that be stored with program code on the storage medium, said program code is held
Claim 1-5 described in any item object abnormal state detection methods are realized when row.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811397468.3A CN109543607A (en) | 2018-11-22 | 2018-11-22 | Object abnormal state detection method, system, monitor system and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811397468.3A CN109543607A (en) | 2018-11-22 | 2018-11-22 | Object abnormal state detection method, system, monitor system and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109543607A true CN109543607A (en) | 2019-03-29 |
Family
ID=65849220
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811397468.3A Pending CN109543607A (en) | 2018-11-22 | 2018-11-22 | Object abnormal state detection method, system, monitor system and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109543607A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110094782A (en) * | 2019-05-20 | 2019-08-06 | 珠海格力电器股份有限公司 | A kind of control method and intelligent range hood of intelligent range hood |
CN110113226A (en) * | 2019-04-16 | 2019-08-09 | 新华三信息安全技术有限公司 | A kind of method and device of detection device exception |
CN110991222A (en) * | 2019-10-16 | 2020-04-10 | 北京海益同展信息科技有限公司 | Object state monitoring and sow oestrus monitoring method, device and system |
CN111460991A (en) * | 2020-03-31 | 2020-07-28 | 科大讯飞股份有限公司 | Anomaly detection method, related device and readable storage medium |
CN112329517A (en) * | 2020-09-17 | 2021-02-05 | 中国南方电网有限责任公司超高压输电公司南宁监控中心 | Transformer substation disconnecting link confirmation video image analysis method and system |
CN112669316A (en) * | 2021-01-29 | 2021-04-16 | 南方电网调峰调频发电有限公司 | Power production abnormity monitoring method and device, computer equipment and storage medium |
WO2021104007A1 (en) * | 2019-11-26 | 2021-06-03 | 京东数科海益信息科技有限公司 | Method and device for animal state monitoring, electronic device, and storage medium |
CN113393709A (en) * | 2021-05-19 | 2021-09-14 | 深圳市安星数字***有限公司 | Remote early warning system and method |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107945458A (en) * | 2017-12-11 | 2018-04-20 | 北京奇虎科技有限公司 | Fall detection method, device and computing device under camera scene |
CN108171181A (en) * | 2017-12-31 | 2018-06-15 | 南昌大学 | A kind of human fall detection method suitable in household |
-
2018
- 2018-11-22 CN CN201811397468.3A patent/CN109543607A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107945458A (en) * | 2017-12-11 | 2018-04-20 | 北京奇虎科技有限公司 | Fall detection method, device and computing device under camera scene |
CN108171181A (en) * | 2017-12-31 | 2018-06-15 | 南昌大学 | A kind of human fall detection method suitable in household |
Non-Patent Citations (1)
Title |
---|
黎德源: "基于机器视觉的扶梯安全检测方法的研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110113226A (en) * | 2019-04-16 | 2019-08-09 | 新华三信息安全技术有限公司 | A kind of method and device of detection device exception |
CN110113226B (en) * | 2019-04-16 | 2021-03-12 | 新华三信息安全技术有限公司 | Method and device for detecting equipment abnormity |
CN110094782A (en) * | 2019-05-20 | 2019-08-06 | 珠海格力电器股份有限公司 | A kind of control method and intelligent range hood of intelligent range hood |
CN110991222A (en) * | 2019-10-16 | 2020-04-10 | 北京海益同展信息科技有限公司 | Object state monitoring and sow oestrus monitoring method, device and system |
CN110991222B (en) * | 2019-10-16 | 2023-12-08 | 京东科技信息技术有限公司 | Object state monitoring and sow oestrus monitoring method, device and system |
WO2021104007A1 (en) * | 2019-11-26 | 2021-06-03 | 京东数科海益信息科技有限公司 | Method and device for animal state monitoring, electronic device, and storage medium |
CN111460991A (en) * | 2020-03-31 | 2020-07-28 | 科大讯飞股份有限公司 | Anomaly detection method, related device and readable storage medium |
CN112329517A (en) * | 2020-09-17 | 2021-02-05 | 中国南方电网有限责任公司超高压输电公司南宁监控中心 | Transformer substation disconnecting link confirmation video image analysis method and system |
CN112329517B (en) * | 2020-09-17 | 2022-11-29 | 中国南方电网有限责任公司超高压输电公司南宁监控中心 | Transformer substation disconnecting link confirmation video image analysis method and system |
CN112669316A (en) * | 2021-01-29 | 2021-04-16 | 南方电网调峰调频发电有限公司 | Power production abnormity monitoring method and device, computer equipment and storage medium |
CN113393709A (en) * | 2021-05-19 | 2021-09-14 | 深圳市安星数字***有限公司 | Remote early warning system and method |
CN113393709B (en) * | 2021-05-19 | 2022-08-16 | 深圳市安星数字***有限公司 | Remote early warning system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109543607A (en) | Object abnormal state detection method, system, monitor system and storage medium | |
CN108319926A (en) | A kind of the safety cap wearing detecting system and detection method of building-site | |
US11935378B2 (en) | Intrusion detection methods and devices | |
US20210124914A1 (en) | Training method of network, monitoring method, system, storage medium and computer device | |
CN105940434A (en) | Information processing device, information processing method, and program | |
CN110956118B (en) | Target object detection method and device, storage medium and electronic device | |
CN111241913A (en) | Method, device and system for detecting falling of personnel | |
CN110830772A (en) | Kitchen video analysis resource scheduling method, device and system | |
CN107330414A (en) | Act of violence monitoring method | |
CN112001230A (en) | Sleeping behavior monitoring method and device, computer equipment and readable storage medium | |
CN110674753A (en) | Theft early warning method, terminal device and storage medium | |
CN111191507A (en) | Safety early warning analysis method and system for smart community | |
CN110782627A (en) | Anti-lost system based on multi-person face recognition | |
CN102457705A (en) | Method and system for detecting and monitoring fight behavior | |
Pramerdorfer et al. | Fall detection based on depth-data in practice | |
CN114613101A (en) | Intelligent hotel security system based on big data | |
JP2024040203A (en) | Person detection device, person tracking device, person tracking system, person detection method, person tracking method, person detection program, and person tracking program | |
CN105718886A (en) | Moving personnel safety abnormity tumbling detection method | |
CN113314230A (en) | Intelligent epidemic prevention method, device, equipment and storage medium based on big data | |
CN111753587B (en) | Ground falling detection method and device | |
CN112070185A (en) | Re-ID-based non-contact fever person tracking system and tracking method thereof | |
CN111191499A (en) | Fall detection method and device based on minimum center line | |
CN116959099A (en) | Abnormal behavior identification method based on space-time diagram convolutional neural network | |
CN113660455B (en) | Method, system and terminal for fall detection based on DVS data | |
WO2022262013A1 (en) | Method and system for collecting and monitoring vehicle state information |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190329 |