CN115984767A - Abnormity early warning method and system based on real-time analysis of monitoring picture - Google Patents
Abnormity early warning method and system based on real-time analysis of monitoring picture Download PDFInfo
- Publication number
- CN115984767A CN115984767A CN202211593792.9A CN202211593792A CN115984767A CN 115984767 A CN115984767 A CN 115984767A CN 202211593792 A CN202211593792 A CN 202211593792A CN 115984767 A CN115984767 A CN 115984767A
- Authority
- CN
- China
- Prior art keywords
- information
- real
- scene
- personnel
- target
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 113
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000010223 real-time analysis Methods 0.000 title claims abstract description 15
- 238000011156 evaluation Methods 0.000 claims abstract description 132
- 206010000117 Abnormal behaviour Diseases 0.000 claims abstract description 62
- 230000002159 abnormal effect Effects 0.000 claims abstract description 39
- 238000012806 monitoring device Methods 0.000 claims abstract description 18
- 238000007781 pre-processing Methods 0.000 claims abstract description 14
- 238000013210 evaluation model Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000011897 real-time detection Methods 0.000 claims abstract description 7
- 230000036541 health Effects 0.000 claims description 54
- 238000012502 risk assessment Methods 0.000 claims description 28
- 238000012216 screening Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 230000036544 posture Effects 0.000 description 32
- 239000000523 sample Substances 0.000 description 19
- 230000009471 action Effects 0.000 description 7
- 230000006378 damage Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Landscapes
- Alarm Systems (AREA)
Abstract
The application relates to the technical field of image data processing, and provides an abnormity early warning method and system based on real-time analysis of a monitoring picture, wherein the method comprises the following steps: monitoring the target personnel in real time through a monitoring device; preprocessing the real-time detection image information; based on the real-time monitoring picture information, identifying abnormal behaviors of the target personnel; carrying out scene recognition on the target personnel based on the real-time monitoring picture information; inputting the target abnormal behavior recognition result and the scene recognition result into a personnel safety evaluation model to obtain a target safety evaluation coefficient; if the target safety evaluation coefficient is larger than the preset safety evaluation coefficient, acquiring an abnormal early warning signal; and carrying out real-time safety early warning on the target personnel based on the abnormal early warning signal. By adopting the method, the technical problem that the safety monitoring accuracy of the users such as the old, children and the like is not enough to carry out timely and effective safety early warning on the users can be solved.
Description
Technical Field
The application relates to the technical field of image data processing, in particular to an abnormity early warning method and system based on real-time analysis of a monitoring picture.
Background
Only old man and child can have certain potential safety hazard when being at home, and current family supervisory equipment can only watch the personnel in the picture usually to dangerous action in the picture carries out intelligent analysis and safety precaution, so that when the unusual condition appears in old man and child at home, can not in time take measures to save and cure.
In conclusion, the technical problem that the safety monitoring accuracy of users such as old people and children is not high enough to perform timely and effective safety early warning on the users exists in the prior art.
Disclosure of Invention
Therefore, it is necessary to provide an anomaly early warning method and system based on real-time analysis of a monitoring picture in order to solve the above technical problems.
An abnormity early warning method based on real-time analysis of a monitoring picture is applied to an abnormity early warning system, the system is in communication connection with a monitoring device, and the method comprises the following steps: monitoring the target person in real time through the monitoring device to obtain real-time monitoring image information; preprocessing the real-time detection image information to obtain real-time monitoring picture information; based on the real-time monitoring picture information, carrying out abnormal behavior recognition on the target person to obtain a target abnormal behavior recognition result; based on the real-time monitoring picture information, carrying out scene recognition on the target person to obtain a scene recognition result; inputting the target abnormal behavior recognition result and the scene recognition result into a personnel safety evaluation model to obtain a target safety evaluation coefficient; obtaining a preset safety evaluation coefficient; judging whether the target safety evaluation coefficient is larger than the preset safety evaluation coefficient or not, and if the target safety evaluation coefficient is larger than the preset safety evaluation coefficient, obtaining an abnormal early warning signal; and carrying out real-time safety early warning on the target personnel based on the abnormal early warning signal.
In one embodiment, the preprocessing the real-time monitoring image information to obtain real-time monitoring picture information further includes: carrying out grid division on the real-time monitoring image information to obtain monitoring division image information; based on a median filtering algorithm, carrying out denoising processing on the monitoring division image information to obtain denoising monitoring image information; and performing image enhancement on the denoising monitoring image information by generating a countermeasure network to obtain the real-time monitoring image information.
In one embodiment, the identifying the abnormal behavior of the target person based on the real-time monitoring picture information to obtain a target abnormal behavior identification result further includes: carrying out motion attitude identification on the target personnel based on the real-time monitoring picture information to obtain personnel motion attitude information; acquiring health information of the target person to obtain person health information; evaluating based on the personnel health information to obtain a personnel health index; acquiring personnel health characteristic information based on the personnel health information and the personnel health index; constructing an abnormal behavior identification unit based on the personnel health characteristic information; and inputting the personnel motion posture information into the abnormal behavior recognition unit to obtain the target abnormal behavior recognition result.
In one embodiment, the constructing an abnormal behavior recognition unit based on the personnel health feature information further includes: screening personnel based on the personnel health characteristic information to obtain a plurality of sample personnel; acquiring information based on the plurality of sample persons to obtain a plurality of sample motion attitude information; performing posture risk assessment based on the motion posture information of the samples to obtain a plurality of posture risk assessment coefficients; screening the plurality of attitude risk evaluation coefficients based on an attitude risk evaluation coefficient threshold value to obtain a plurality of attitude risk evaluation abnormal coefficients meeting the attitude risk evaluation coefficient threshold value; matching the plurality of sample motion attitude information based on the plurality of attitude risk assessment abnormal coefficients to obtain a plurality of abnormal motion attitude information; embedding the plurality of abnormal motion posture information into the abnormal behavior recognition unit.
In one embodiment, the performing scene recognition on the target person based on the real-time monitoring picture information to obtain a scene recognition result further includes: based on the real-time monitoring picture information, obtaining real-time scene information of the target person; identifying scene space environment based on the real-time scene information to obtain scene space environment information; carrying out article distribution identification based on the real-time scene information to obtain scene article distribution information; and obtaining the scene identification result based on the scene space environment information and the scene article distribution information.
In one embodiment, further comprising: the scene article distribution information comprises a plurality of scene article information and a plurality of scene article position information; carrying out risk assessment based on the information of the plurality of scene articles to obtain a plurality of article risk assessment coefficients; performing touch probability analysis based on the position information of the scene objects to obtain a plurality of object touch probability indexes; marking the distribution information of the scene articles based on the risk evaluation coefficients of the articles and the touch probability indexes of the articles to obtain marking characteristic information of the plurality of scene articles; adding the scene item mark feature information to the scene recognition result.
In one embodiment, the obtaining the target security assessment coefficient further includes: the personnel safety assessment model comprises an input layer, a scene safety assessment layer, a personnel behavior safety assessment layer and an output layer; inputting the scene recognition result into the scene safety evaluation layer to obtain a scene safety evaluation index; inputting the target abnormal behavior identification result into the personnel behavior safety evaluation layer to obtain a personnel behavior safety evaluation index; obtaining a preset weight distribution condition, wherein the preset weight distribution condition comprises a preset scene weight coefficient and a preset behavior weight coefficient; and performing weighted calculation on the scene safety evaluation index and the personnel behavior safety evaluation index based on the preset weight distribution condition to obtain the target safety evaluation coefficient.
An abnormity early warning system based on real-time analysis of a monitoring picture, the system is in communication connection with a monitoring device, and the system comprises:
the real-time monitoring module is used for monitoring a target person in real time through the monitoring device to obtain real-time monitoring image information;
the image preprocessing module is used for preprocessing the real-time detection image information to obtain real-time monitoring picture information;
the abnormal behavior recognition module is used for recognizing the abnormal behavior of the target person based on the real-time monitoring picture information to obtain a target abnormal behavior recognition result;
the scene recognition module is used for carrying out scene recognition on the target personnel based on the real-time monitoring picture information to obtain a scene recognition result;
the personnel safety evaluation module is used for inputting the target abnormal behavior identification result and the scene identification result into a personnel safety evaluation model to obtain a target safety evaluation coefficient;
the system comprises a preset evaluation coefficient module, a safety evaluation module and a safety evaluation module, wherein the preset evaluation coefficient module is used for obtaining a preset safety evaluation coefficient;
the safety evaluation judging module is used for judging whether the target safety evaluation coefficient is larger than the preset safety evaluation coefficient or not, and if the target safety evaluation coefficient is larger than the preset safety evaluation coefficient, obtaining an abnormal early warning signal;
and the real-time safety early warning module is used for carrying out real-time safety early warning on the target personnel based on the abnormal early warning signal.
The abnormity early warning method and the abnormity early warning system based on real-time analysis of the monitoring picture solve the technical problem that the safety monitoring accuracy of users such as old people and children is not enough to carry out timely and effective safety early warning on the users. The method comprises the steps of obtaining real-time monitoring image information through a monitoring device, simultaneously carrying out denoising processing on the real-time monitoring image information to obtain monitoring picture information, carrying out abnormal behavior and scene recognition on the monitoring picture information, evaluating a recognition result through a personnel safety evaluation model, judging an evaluation result, generating early warning information according to a judgment result, finding a dangerous signal in time, and taking measures to timely treat a dangerous accident.
The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.
Drawings
Fig. 1 is a schematic flow chart of an anomaly early warning method based on real-time analysis of a monitored picture according to the present application;
fig. 2 is a schematic structural diagram of an anomaly early warning system based on real-time analysis of a monitoring picture according to the present application.
Description of reference numerals: the system comprises a real-time monitoring module 1, an image preprocessing module 2, an abnormal behavior identification module 3, a scene identification module 4, a personnel safety evaluation module 5, a preset evaluation coefficient module 6, a safety evaluation judgment module 7 and a real-time safety early warning module 8.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
As shown in fig. 1, the present application provides an anomaly early warning method based on real-time analysis of a monitoring picture, where the method is applied to an anomaly early warning system, the system is in communication connection with a monitoring device, and the method includes:
step S100: monitoring the target personnel in real time through the monitoring device to obtain real-time monitoring image information;
specifically, the communication connection between the system and the monitoring device means that the monitoring device sends monitored real-time image information to the abnormality early warning system through a signal transmission module, and the monitoring device is used for monitoring the activity state and the scene of target personnel in real time and comprises equipment with an image acquisition function, such as a household monitoring probe, a computer camera and the like. Firstly, a target person is monitored in real time through an installed monitoring device to obtain real-time monitoring image information, the target person mainly comprises special people such as old people, children and disabled people, and original data are provided for risk analysis of the target person in the next step through obtaining the real-time monitoring image information.
Step S200: preprocessing the real-time detection image information to obtain real-time monitoring picture information;
in one embodiment, step S200 of the present application further includes:
step S210: carrying out grid division on the real-time monitoring image information to obtain monitoring division image information;
step S220: based on a median filtering algorithm, carrying out denoising processing on the monitoring division image information to obtain denoising monitoring image information;
step S230: and carrying out image enhancement on the de-noising monitoring image information by generating a countermeasure network to obtain the real-time monitoring image information.
Specifically, the real-time monitoring image information is first subjected to grid division, where the grid division refers to dividing the real-time monitoring image information into many small cells according to a grid, and the obtained detection division image information is a plurality of obtained grid cells. And then denoising the monitoring division image information according to a median filtering algorithm, wherein the median filtering algorithm is a nonlinear signal processing technology which is based on a sorting statistical theory and can effectively inhibit noise, and is mainly used for replacing the value of one point in the digital image with the median of each point value in a neighborhood of the point and enabling the surrounding pixel value to be close to a real value, so that an isolated noise point is eliminated, and the denoising detection image information is obtained. And carrying out image enhancement on the de-noised monitoring image information by generating a countermeasure network, wherein the generated countermeasure network is an important generation model in the field of deep learning, namely a generator and a discriminator are trained at the same time and compete in a minimized maximum algorithm to obtain the real-time monitoring picture information. Through the steps, the obtained real-time monitoring picture information can be clearer and can be easily identified.
Step S300: based on the real-time monitoring picture information, carrying out abnormal behavior identification on the target person to obtain a target abnormal behavior identification result;
in one embodiment, step S300 of the present application further includes:
step S310: performing motion posture recognition on the target personnel based on the real-time monitoring picture information to obtain personnel motion posture information;
step S320: acquiring health information of the target person to obtain person health information;
step S330: evaluating based on the personnel health information to obtain a personnel health index;
step S340: acquiring personnel health characteristic information based on the personnel health information and the personnel health index;
step S350: constructing an abnormal behavior identification unit based on the personnel health characteristic information;
in one embodiment, step S350 of the present application further includes:
step S351: screening personnel based on the personnel health characteristic information to obtain a plurality of sample personnel;
step S352: acquiring information based on the plurality of sample persons to obtain a plurality of sample motion attitude information;
step S353: performing posture risk assessment based on the motion posture information of the samples to obtain a plurality of posture risk assessment coefficients;
step S354: screening the plurality of attitude risk evaluation coefficients based on an attitude risk evaluation coefficient threshold value to obtain a plurality of attitude risk evaluation abnormal coefficients meeting the attitude risk evaluation coefficient threshold value;
step S355: matching the plurality of sample motion attitude information based on the plurality of attitude risk assessment abnormal coefficients to obtain a plurality of abnormal motion attitude information;
step S356: embedding the plurality of abnormal motion posture information into the abnormal behavior recognition unit.
Specifically, personnel screening is performed based on big data according to the health characteristic information of the target personnel, and a plurality of sample personnel are obtained, wherein the sample personnel are the personnel with the same and similar health characteristic information as the target personnel. And acquiring the motion postures of the plurality of sample personnel to obtain the motion posture information of the plurality of samples. And performing risk evaluation on the plurality of sample posture information, judging whether the motion posture damages the sample personnel, if so, obtaining a plurality of posture risk evaluation coefficients, wherein the posture risk evaluation coefficient threshold is a posture risk evaluation coefficient risk value, and is used for judging the posture risk evaluation coefficient, and listing a coefficient with the posture risk evaluation coefficient being greater than a preset posture risk evaluation risk value as a posture risk evaluation abnormal coefficient to obtain a plurality of posture risk evaluation abnormal coefficients. And further obtaining a plurality of abnormal motion posture information corresponding to the posture risk assessment abnormal coefficient, and finally embedding the plurality of abnormal motion posture information into the abnormal behavior recognition unit as storage information. And by obtaining the abnormal behavior recognition unit, a basis is provided for analyzing the abnormal behavior of the target person in the next step.
Step S360: and inputting the personnel motion posture information into the abnormal behavior recognition unit to obtain the target abnormal behavior recognition result.
Specifically, motion information of a target person is obtained according to the real-time monitoring picture information, motion posture recognition is performed, and the person motion posture information is obtained, for example: standing, walking, running, pouring water, etc. The method comprises the steps of collecting health information of a target person, wherein the health information comprises various physical indexes such as age, blood pressure and health condition of the target person. And evaluating the health condition of the target person according to the health characteristic information of the person to obtain a person health index, wherein the larger the person health index is, the healthier the target person is. And comprehensively analyzing the personnel health information and the personnel health index to obtain personnel health characteristic information, and further obtaining an abnormal behavior identification unit according to the personnel health characteristic information, wherein the abnormal behavior identification unit, namely the behavior in the unit, is a dangerous action for a target person and can cause injury. And then inputting the real-time motion attitude information of the target person into the abnormal behavior recognition unit for recognition, judging whether the real-time motion attitude of the target person conforms to dangerous actions in the abnormal behavior recognition unit or not, obtaining a target abnormal behavior recognition result, and providing data support for the next step of safety evaluation on the target person by obtaining the target abnormal behavior recognition result.
Step S400: based on the real-time monitoring picture information, carrying out scene recognition on the target person to obtain a scene recognition result;
in one embodiment, step S400 of the present application further includes:
step S410: based on the real-time monitoring picture information, obtaining real-time scene information of the target person;
step S420: identifying scene space environment based on the real-time scene information to obtain scene space environment information;
step S430: carrying out article distribution identification based on the real-time scene information to obtain scene article distribution information;
step S440: and obtaining the scene identification result based on the scene space environment information and the scene article distribution information.
In one embodiment, step S440 of the present application further includes:
step S441: the scene article distribution information comprises a plurality of scene article information and a plurality of scene article position information;
step S442: performing risk assessment based on the scene article information to obtain a plurality of article risk assessment coefficients;
step S443: performing touch probability analysis based on the position information of the scene articles to obtain a plurality of article touch probability indexes;
step S444: marking the distribution information of the scene articles based on the risk evaluation coefficients of the articles and the touch probability indexes of the articles to obtain marking characteristic information of the plurality of scene articles;
step S445: adding the scene item mark feature information to the scene recognition result.
Specifically, scene information of the target person is obtained according to the real-time monitoring picture information, and the scene information comprises the current space environment and dangerous goods in the environment. And identifying the space environment of the target person to obtain scene space environment information. For example: kitchens, balconies, living rooms, toilets, etc. And then identifying the distribution of the articles in the scene to obtain article distribution information, wherein the article distribution information comprises the types of the articles in the scene and the positions of the articles. And then carrying out risk assessment on the plurality of articles to obtain risk assessment coefficients of the articles, wherein the higher the risk of the articles is, the higher the risk assessment coefficients are. And analyzing the probability of the target person touching the object according to the position information of the object to obtain a plurality of object collision probability indexes, wherein the collision probability is larger when the distance between the object and the target person is smaller and the collision probability is larger. Marking the articles according to the article risk evaluation coefficients and the article touch probability indexes to obtain a scene identification result, wherein the scene identification result comprises scene space environment information, scene article distribution information and scene article marking characteristic information. And by obtaining the scene recognition result, data support is provided for the next step of safety evaluation on the target personnel.
Step S500: inputting the target abnormal behavior recognition result and the scene recognition result into a personnel safety evaluation model to obtain a target safety evaluation coefficient;
in one embodiment, step S500 of the present application further includes:
step S510: the personnel safety assessment model comprises an input layer, a scene safety assessment layer, a personnel behavior safety assessment layer and an output layer;
step S520: inputting the scene recognition result into the scene safety evaluation layer to obtain a scene safety evaluation index;
step S530: inputting the target abnormal behavior identification result into the personnel behavior safety evaluation layer to obtain a personnel behavior safety evaluation index;
step S540: obtaining a preset weight distribution condition, wherein the preset weight distribution condition comprises a preset scene weight coefficient and a preset behavior weight coefficient;
step S550: and performing weighted calculation on the scene safety evaluation index and the personnel behavior safety evaluation index based on the preset weight distribution condition to obtain the target safety evaluation coefficient.
Specifically, the personnel safety assessment model is a mathematical model, assesses the safety condition of a target person by generating a safety assessment coefficient, and comprises an input layer, a scene safety assessment layer, a personnel behavior safety assessment layer and an output layer. And the input layer is used for inputting the target abnormal behavior recognition result and the scene recognition result into a personnel safety evaluation model. The scene safety evaluation layer is used for evaluating the scene recognition result and outputting a scene safety evaluation index, and the smaller the scene safety evaluation index is, the lower the risk coefficient is. The personnel behavior safety evaluation layer is used for evaluating the target abnormal behavior recognition result to obtain a personnel behavior safety evaluation index, and the smaller the personnel behavior safety evaluation index is, the lower the risk coefficient of the action of the target personnel behavior is. And carrying out weight distribution on the scene safety evaluation index and the personnel behavior safety evaluation index, then carrying out weighted calculation, and outputting a target safety evaluation coefficient. And a basis is provided for safety analysis of target personnel by obtaining the target safety evaluation coefficient.
Step S600: obtaining a preset safety evaluation coefficient;
step S700: judging whether the target safety evaluation coefficient is larger than the preset safety evaluation coefficient or not, and if the target safety evaluation coefficient is larger than the preset safety evaluation coefficient, acquiring an abnormal early warning signal;
step S800: and carrying out real-time safety early warning on the target personnel based on the abnormal early warning signal.
Specifically, the preset safety evaluation coefficient is used for judging a target safety evaluation coefficient, and when the target safety evaluation coefficient is greater than the preset safety evaluation coefficient, it indicates that the action is a dangerous action, and the system generates an abnormal early warning signal and sends the early warning signal to a guardian of a target person, for example: parents of children or children of old people and the like contact target personnel in time in various ways to perform safety early warning after the guardian receives the early warning information, so that safety accidents are avoided, and the technical problem that the safety monitoring accuracy of users such as the old people and the children is not high enough to perform timely and effective safety early warning on the users can be solved.
In one embodiment, as shown in fig. 2, an anomaly early warning system based on real-time analysis of a monitoring picture is provided, and the system is in communication connection with a monitoring device, and includes: real-time supervision module 1, image preprocessing module 2, unusual behavior identification module 3, scene identification module 4, personnel's safety assessment module 5, predetermine evaluation coefficient module 6, safety assessment judging module 7, real-time safety precaution module 8, wherein:
the real-time monitoring module 1 is used for monitoring a target person in real time through the monitoring device to obtain real-time monitoring image information;
the image preprocessing module 2 is used for preprocessing the real-time detection image information to obtain real-time monitoring picture information;
the abnormal behavior recognition module 3 is used for recognizing the abnormal behavior of the target person based on the real-time monitoring picture information to obtain a target abnormal behavior recognition result;
the scene recognition module 4 is used for carrying out scene recognition on the target personnel based on the real-time monitoring picture information to obtain a scene recognition result;
the personnel safety evaluation module 5 is used for inputting the target abnormal behavior recognition result and the scene recognition result into a personnel safety evaluation model to obtain a target safety evaluation coefficient;
the preset evaluation coefficient module 6 is used for obtaining a preset safety evaluation coefficient;
the safety evaluation judging module 7 is used for judging whether the target safety evaluation coefficient is larger than the preset safety evaluation coefficient or not, and if the target safety evaluation coefficient is larger than the preset safety evaluation coefficient, obtaining an abnormal early warning signal;
and the real-time safety early warning module 8 is used for carrying out real-time safety early warning on the target personnel based on the abnormal early warning signal.
In one embodiment, the system further comprises:
the grid division module is used for carrying out grid division on the real-time monitoring image information to obtain monitoring division image information;
the image denoising module is used for denoising the monitoring division image information based on a median filtering algorithm to obtain denoising monitoring image information;
and the image enhancement module is used for carrying out image enhancement on the de-noising monitoring image information by generating a countermeasure network to obtain the real-time monitoring image information.
In one embodiment, the system further comprises:
the motion gesture recognition module is used for recognizing the motion gesture of the target person based on the real-time monitoring picture information to obtain person motion gesture information;
the health information acquisition module is used for acquiring the health information of the target personnel and acquiring personnel health information;
the health information evaluation module is used for evaluating based on the personnel health information to obtain a personnel health index;
the health characteristic information obtaining module is used for obtaining personnel health characteristic information based on the personnel health information and the personnel health index;
the abnormal unit construction module is used for constructing an abnormal behavior identification unit based on the personnel health characteristic information;
and the abnormal result obtaining module is used for inputting the personnel motion posture information into the abnormal behavior recognition unit to obtain the target abnormal behavior recognition result.
In one embodiment, the system further comprises:
the personnel screening module is used for screening personnel based on the personnel health characteristic information to obtain a plurality of sample personnel;
the information acquisition module is used for acquiring information based on the plurality of sample personnel to obtain a plurality of sample motion attitude information;
a risk assessment module for performing attitude risk assessment based on the plurality of sample motion attitude information to obtain a plurality of attitude risk assessment coefficients;
the evaluation coefficient screening module is used for screening the attitude risk evaluation coefficients based on an attitude risk evaluation coefficient threshold value to obtain a plurality of attitude risk evaluation abnormal coefficients meeting the attitude risk evaluation coefficient threshold value;
the information matching module is used for matching the plurality of sample motion attitude information based on the plurality of attitude risk assessment abnormal coefficients to obtain a plurality of abnormal motion attitude information;
an information embedding module to embed the plurality of abnormal motion posture information into the abnormal behavior recognition unit.
In one embodiment, the system further comprises:
the real-time scene information acquisition module is used for acquiring real-time scene information of a target person based on the real-time monitoring picture information;
the spatial environment recognition module is used for recognizing scene spatial environment based on the real-time scene information to obtain scene spatial environment information;
the article distribution identification module is used for identifying article distribution based on the real-time scene information to obtain scene article distribution information;
a scene recognition result obtaining module, configured to obtain the scene recognition result based on the scene space environment information and the scene article distribution information.
In one embodiment, the system further comprises:
the article distribution information module is used for indicating that the scene article distribution information comprises a plurality of pieces of scene article information and a plurality of pieces of scene article position information;
the risk assessment module is used for carrying out risk assessment based on the information of the plurality of scene articles to obtain a plurality of article risk assessment coefficients;
the touch probability analysis module is used for carrying out touch probability analysis based on the position information of the scene objects to obtain a plurality of object touch probability indexes;
the information marking module is used for marking the distribution information of the scene articles based on the risk evaluation coefficients of the articles and the touch probability indexes of the articles to obtain marking characteristic information of the articles;
an information adding module for adding the scene item marking feature information to the scene recognition result.
In one embodiment, the system further comprises:
the model obtaining module is used for obtaining a personnel safety assessment model, wherein the personnel safety assessment model comprises an input layer, a scene safety assessment layer, a personnel behavior safety assessment layer and an output layer;
the scene information input module is used for inputting the scene recognition result into the scene safety evaluation layer to obtain a scene safety evaluation index;
the abnormal recognition result input module is used for inputting the target abnormal behavior recognition result into the personnel behavior safety evaluation layer to obtain a personnel behavior safety evaluation index;
the system comprises a weight obtaining module, a weight obtaining module and a weight distribution module, wherein the weight obtaining module is used for obtaining preset weight distribution conditions, and the preset weight distribution conditions comprise preset scene weight coefficients and preset behavior weight coefficients;
and the weighting calculation module is used for carrying out weighting calculation on the scene safety evaluation index and the personnel behavior safety evaluation index based on the preset weight distribution condition to obtain the target safety evaluation coefficient.
In summary, the present application provides an anomaly early warning method and system based on real-time analysis of monitoring images, which has the following technical effects:
1. the safety monitoring system solves the technical problem that the safety monitoring accuracy of users such as old people and children is not enough to carry out timely and effective safety early warning on the users, can timely find out dangerous signals by generating early warning information, avoids dangerous accidents, and timely takes measures to treat the dangerous accidents.
2. The method comprises the steps of establishing a personnel safety assessment model, carrying out double recognition and assessment on the posture action and the scene environment of the target person, obtaining a target safety assessment coefficient based on a scene safety assessment index and a personnel behavior safety assessment index, and improving the accuracy of risk prediction on the target person.
All possible combinations of the technical features in the above embodiments may not be described for the sake of brevity, but should be considered as being within the scope of the present disclosure as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is specific and detailed, but not to be understood as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (8)
1. An abnormity early warning method based on real-time analysis of a monitoring picture is characterized in that the method is applied to an abnormity early warning system, the system is in communication connection with a monitoring device, and the method comprises the following steps:
monitoring the target personnel in real time through the monitoring device to obtain real-time monitoring image information;
preprocessing the real-time detection image information to obtain real-time monitoring picture information;
based on the real-time monitoring picture information, carrying out abnormal behavior identification on the target person to obtain a target abnormal behavior identification result;
based on the real-time monitoring picture information, carrying out scene recognition on the target person to obtain a scene recognition result;
inputting the target abnormal behavior recognition result and the scene recognition result into a personnel safety evaluation model to obtain a target safety evaluation coefficient;
obtaining a preset safety evaluation coefficient;
judging whether the target safety evaluation coefficient is larger than the preset safety evaluation coefficient or not, and if the target safety evaluation coefficient is larger than the preset safety evaluation coefficient, acquiring an abnormal early warning signal;
and carrying out real-time safety early warning on the target personnel based on the abnormal early warning signal.
2. The method of claim 1, wherein said preprocessing said real-time monitoring image information to obtain real-time monitoring picture information, further comprises:
carrying out grid division on the real-time monitoring image information to obtain monitoring division image information;
based on a median filtering algorithm, carrying out denoising processing on the monitoring division image information to obtain denoising monitoring image information;
and carrying out image enhancement on the de-noising monitoring image information by generating a countermeasure network to obtain the real-time monitoring image information.
3. The method of claim 1, wherein the identifying abnormal behavior of the target person based on the real-time monitoring picture information to obtain a target abnormal behavior identification result further comprises:
performing motion posture recognition on the target personnel based on the real-time monitoring picture information to obtain personnel motion posture information;
acquiring health information of the target person to obtain person health information;
evaluating based on the personnel health information to obtain a personnel health index;
acquiring personnel health characteristic information based on the personnel health information and the personnel health index;
constructing an abnormal behavior identification unit based on the personnel health characteristic information;
and inputting the personnel motion posture information into the abnormal behavior recognition unit to obtain the target abnormal behavior recognition result.
4. The method of claim 3, wherein constructing an abnormal behavior recognition unit based on the personnel health characteristic information further comprises:
screening personnel based on the personnel health characteristic information to obtain a plurality of sample personnel;
acquiring information based on the plurality of sample persons to obtain a plurality of sample motion attitude information;
performing attitude risk assessment based on the motion attitude information of the plurality of samples to obtain a plurality of attitude risk assessment coefficients;
screening the plurality of attitude risk evaluation coefficients based on an attitude risk evaluation coefficient threshold value to obtain a plurality of attitude risk evaluation abnormal coefficients meeting the attitude risk evaluation coefficient threshold value;
matching the plurality of sample motion attitude information based on the plurality of attitude risk assessment abnormal coefficients to obtain a plurality of abnormal motion attitude information;
and embedding the plurality of abnormal motion posture information into the abnormal behavior recognition unit.
5. The method of claim 1, wherein the performing scene recognition on the target person based on the real-time monitoring picture information to obtain a scene recognition result further comprises:
based on the real-time monitoring picture information, obtaining real-time scene information of a target person;
identifying scene space environment based on the real-time scene information to obtain scene space environment information;
carrying out article distribution identification based on the real-time scene information to obtain scene article distribution information;
and obtaining the scene identification result based on the scene space environment information and the scene article distribution information.
6. The method of claim 5, further comprising:
the scene item distribution information comprises a plurality of scene item information and a plurality of scene item position information;
performing risk assessment based on the scene article information to obtain a plurality of article risk assessment coefficients;
performing touch probability analysis based on the position information of the scene objects to obtain a plurality of object touch probability indexes;
marking the scene article distribution information based on the article risk assessment coefficients and the article touch probability indexes to obtain scene article marking characteristic information;
adding the scene item mark feature information to the scene recognition result.
7. The method of claim 1, wherein the obtaining a target security assessment coefficient further comprises:
the personnel safety assessment model comprises an input layer, a scene safety assessment layer, a personnel behavior safety assessment layer and an output layer;
inputting the scene recognition result into the scene safety evaluation layer to obtain a scene safety evaluation index;
inputting the target abnormal behavior identification result into the personnel behavior safety evaluation layer to obtain a personnel behavior safety evaluation index;
obtaining a preset weight distribution condition, wherein the preset weight distribution condition comprises a preset scene weight coefficient and a preset behavior weight coefficient;
and performing weighted calculation on the scene safety evaluation index and the personnel behavior safety evaluation index based on the preset weight distribution condition to obtain the target safety evaluation coefficient.
8. An abnormity early warning system based on real-time analysis of a monitoring picture, the system is in communication connection with a monitoring device, and the system comprises:
the real-time monitoring module is used for monitoring the target personnel in real time through the monitoring device to obtain real-time monitoring image information;
the image preprocessing module is used for preprocessing the real-time detection image information to obtain real-time monitoring picture information;
the abnormal behavior recognition module is used for recognizing the abnormal behavior of the target person based on the real-time monitoring picture information to obtain a target abnormal behavior recognition result;
the scene recognition module is used for carrying out scene recognition on the target personnel based on the real-time monitoring picture information to obtain a scene recognition result;
the personnel safety evaluation module is used for inputting the target abnormal behavior identification result and the scene identification result into a personnel safety evaluation model to obtain a target safety evaluation coefficient;
the device comprises a preset coefficient obtaining module, a safety evaluation module and a safety evaluation module, wherein the preset coefficient obtaining module is used for obtaining a preset safety evaluation coefficient;
the safety evaluation judging module is used for judging whether the target safety evaluation coefficient is larger than the preset safety evaluation coefficient or not, and if the target safety evaluation coefficient is larger than the preset safety evaluation coefficient, obtaining an abnormal early warning signal;
and the real-time safety early warning module is used for carrying out real-time safety early warning on the target personnel based on the abnormal early warning signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211593792.9A CN115984767A (en) | 2022-12-13 | 2022-12-13 | Abnormity early warning method and system based on real-time analysis of monitoring picture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211593792.9A CN115984767A (en) | 2022-12-13 | 2022-12-13 | Abnormity early warning method and system based on real-time analysis of monitoring picture |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115984767A true CN115984767A (en) | 2023-04-18 |
Family
ID=85975020
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211593792.9A Pending CN115984767A (en) | 2022-12-13 | 2022-12-13 | Abnormity early warning method and system based on real-time analysis of monitoring picture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115984767A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116743970A (en) * | 2023-08-14 | 2023-09-12 | 安徽塔联智能科技有限责任公司 | Intelligent management platform with video AI early warning analysis |
| CN116777713A (en) * | 2023-08-24 | 2023-09-19 | 湖南天成新宇网络科技有限公司 | Intelligent park safety management method and system based on data analysis |
| CN117082338A (en) * | 2023-05-26 | 2023-11-17 | 哈尔滨市法智科技开发有限公司 | Video analysis monitoring system based on big data |
| CN117315592A (en) * | 2023-11-27 | 2023-12-29 | 四川省医学科学院·四川省人民医院 | Identification early warning system based on robot end real-time monitoring camera shooting |
| CN117334005A (en) * | 2023-09-22 | 2024-01-02 | 江苏前来信息科技有限公司 | Man-machine interaction safety early warning device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170140631A1 (en) * | 2015-09-17 | 2017-05-18 | Luvozo Pbc | Automated environment hazard detection |
| CN106777954A (en) * | 2016-12-09 | 2017-05-31 | 电子科技大学 | The intelligent guarding system and method for a kind of Empty nest elderly health |
| CN108876822A (en) * | 2018-07-09 | 2018-11-23 | 山东大学 | A kind of behavior risk assessment method and household safety-protection nursing system |
| US20190318165A1 (en) * | 2018-04-16 | 2019-10-17 | Peerwell, Inc. | Hazard recognition |
| US20190388011A1 (en) * | 2018-06-25 | 2019-12-26 | Careview Communications, Inc. | Smart monitoring safety and quality of life system using sensors |
| CN111932828A (en) * | 2019-11-05 | 2020-11-13 | 上海中侨健康智能科技有限公司 | Intelligent old-age care monitoring and early warning system based on digital twin technology |
-
2022
- 2022-12-13 CN CN202211593792.9A patent/CN115984767A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170140631A1 (en) * | 2015-09-17 | 2017-05-18 | Luvozo Pbc | Automated environment hazard detection |
| CN106777954A (en) * | 2016-12-09 | 2017-05-31 | 电子科技大学 | The intelligent guarding system and method for a kind of Empty nest elderly health |
| US20190318165A1 (en) * | 2018-04-16 | 2019-10-17 | Peerwell, Inc. | Hazard recognition |
| US20190388011A1 (en) * | 2018-06-25 | 2019-12-26 | Careview Communications, Inc. | Smart monitoring safety and quality of life system using sensors |
| CN108876822A (en) * | 2018-07-09 | 2018-11-23 | 山东大学 | A kind of behavior risk assessment method and household safety-protection nursing system |
| CN111932828A (en) * | 2019-11-05 | 2020-11-13 | 上海中侨健康智能科技有限公司 | Intelligent old-age care monitoring and early warning system based on digital twin technology |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117082338A (en) * | 2023-05-26 | 2023-11-17 | 哈尔滨市法智科技开发有限公司 | Video analysis monitoring system based on big data |
| CN116743970A (en) * | 2023-08-14 | 2023-09-12 | 安徽塔联智能科技有限责任公司 | Intelligent management platform with video AI early warning analysis |
| CN116743970B (en) * | 2023-08-14 | 2023-11-21 | 安徽塔联智能科技有限责任公司 | Intelligent management platform with video AI early warning analysis |
| CN116777713A (en) * | 2023-08-24 | 2023-09-19 | 湖南天成新宇网络科技有限公司 | Intelligent park safety management method and system based on data analysis |
| CN116777713B (en) * | 2023-08-24 | 2023-11-03 | 湖南天成新宇网络科技有限公司 | Intelligent park safety management method and system based on data analysis |
| CN117334005A (en) * | 2023-09-22 | 2024-01-02 | 江苏前来信息科技有限公司 | Man-machine interaction safety early warning device |
| CN117315592A (en) * | 2023-11-27 | 2023-12-29 | 四川省医学科学院·四川省人民医院 | Identification early warning system based on robot end real-time monitoring camera shooting |
| CN117315592B (en) * | 2023-11-27 | 2024-01-30 | 四川省医学科学院·四川省人民医院 | Identification early warning system based on robot end real-time monitoring camera shooting |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115984767A (en) | Abnormity early warning method and system based on real-time analysis of monitoring picture | |
| Harrou et al. | An integrated vision-based approach for efficient human fall detection in a home environment | |
| JP6905850B2 (en) | Image processing system, imaging device, learning model creation method, information processing device | |
| JP5143212B2 (en) | Method and apparatus for modeling behavior using probability distribution functions | |
| CN109714324B (en) | User network abnormal behavior discovery method and system based on machine learning algorithm | |
| US20220028230A1 (en) | Methods And System For Monitoring An Environment | |
| CN112364696A (en) | Method and system for improving family safety by using family monitoring video | |
| CN112478975A (en) | Elevator door fault detection method based on audio features | |
| Hemmatpour et al. | Nonlinear Predictive Threshold Model for Real‐Time Abnormal Gait Detection | |
| CN115690653A (en) | Monitoring and early warning for realizing abnormal nursing behaviors of nursing staff based on AI behavior recognition | |
| CN111951505B (en) | Fence vibration intrusion positioning and mode identification method based on distributed optical fiber system | |
| CN116977129A (en) | Intelligent property management system and method thereof | |
| CN116701866A (en) | Park event linkage processing method based on Internet of things equipment | |
| CN108537105B (en) | Dangerous behavior identification method in home environment | |
| CN116485135A (en) | Wisdom commercial building big data management platform | |
| KR20170004270A (en) | Apparatus and system for detecting falldown | |
| CN115171335A (en) | Image and voice fused indoor safety protection method and device for elderly people living alone | |
| CN113743293A (en) | Fall behavior detection method and device, electronic equipment and storage medium | |
| CN116822929A (en) | Alarm method, alarm device, electronic equipment and storage medium | |
| JP2021099818A (en) | Information processing device, information processing method, and program | |
| CN117994863B (en) | Human behavior recognition method and recognition system thereof | |
| CN118132953A (en) | Life detection system suitable for aging space | |
| WO2023275968A1 (en) | Abnormality determination device, abnormality determination method, and abnormality determination program | |
| CN117541054A (en) | Community security monitoring method and system based on intelligent property | |
| CN117633604A (en) | Audio and video intelligent processing method and device, storage medium and electronic equipment |
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 |