Emergency treatment anti-falling monitoring method and system for emergency treatment monitoring

Abstract

According to the emergency treatment anti-falling monitoring method and system for emergency treatment monitoring provided by the invention, the first emergency treatment monitoring data is subjected to early warning processing according to the target early warning parameter corresponding to each falling hidden danger element, and the falling hidden danger prompt data is obtained. In the above process, the first emergency treatment monitoring data, the second emergency treatment monitoring data and the analysis result features can be processed to obtain the target early warning parameter corresponding to each falling hidden danger element. And performing early warning processing on the first emergency treatment monitoring data according to the target early warning parameter corresponding to each falling hidden danger element to obtain falling hidden danger prompt data. And hidden danger points can be prompted in advance, so that the risk of falling injury of the patient and family members is reduced.

Description

Technical Field

[0001] The present application relates to the field of information monitoring technology, and in particular to an emergency fall prevention monitoring method and system for emergency monitoring. Background Art

[0002] The flow of people in the hospital is relatively large, and there are not only normal people walking, but also patients. If a fall occurs, normal people may be injured and the patient's injury may worsen. Therefore, some hidden dangers of falling need to be reminded. However, how to improve the accuracy of the reminder of hidden dangers of falling is a problem that is difficult to solve at present. Summary of the invention

[0003] In order to improve the technical problems existing in the related technologies, the present application provides an emergency fall prevention monitoring method and system for emergency monitoring.

[0004] In a first aspect, a method for emergency fall prevention monitoring is provided, comprising: Acquire first emergency monitoring data that needs to be processed, wherein the first emergency monitoring data includes at least one fall risk factor; Analyze and process the first emergency monitoring data to obtain second emergency monitoring data and analysis result characteristics, wherein the second emergency monitoring data is used to represent description details of each attribute in the first emergency monitoring data; Processing the first emergency monitoring data, the second emergency monitoring data, and the analysis result characteristics to obtain a target warning parameter corresponding to each fall risk factor; According to the target warning parameters corresponding to each fall risk factor, the first emergency monitoring data is subjected to warning processing to obtain fall risk prompt data.

[0005] In the present application, the first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics are processed to obtain target warning parameters corresponding to each fall risk factor, including: Determining an artificial intelligence recognition thread, wherein the artificial intelligence recognition thread includes at least one derivation unit, at least one compression unit, and at least one decision unit; Processing the first emergency monitoring data, the second emergency monitoring data and the analysis result feature through the at least one derivation unit and the at least one decision unit to obtain at least one transition emergency monitoring data; The at least one compression unit and the at least one decision unit are used to process the at least one transition emergency monitoring data and the analysis result characteristics to obtain target warning parameters corresponding to each fall risk factor.

[0006] In the present application, the first emergency monitoring data, the second emergency monitoring data and the analysis result feature are processed by the at least one derivation unit and the at least one decision unit to obtain at least one transitional emergency monitoring data, including: Integrate the first emergency monitoring data and the second emergency monitoring data to obtain third emergency monitoring data; Processing the third emergency monitoring data through a first derivative unit to obtain a first emergency monitoring data mining result; Processing the first emergency monitoring data mining result and the analysis result feature by a first decision unit to obtain first transition emergency monitoring data; Processing the a-1th transition emergency monitoring data through the ath derivative unit to obtain the ath emergency monitoring data mining result; The ath decision unit processes the ath emergency monitoring data mining result and the analysis result feature to obtain the ath transition emergency monitoring data, where a is 2, 3, ..., X in sequence, and X is the number of the derivative units; Determining that the at least one transitional emergency monitoring data includes the X transitional emergency monitoring data.

[0007] In the present application, the derivation unit includes a feature extraction unit and an encoding unit; the a-1th transition emergency monitoring data is processed by the ath derivation unit to obtain the ath emergency monitoring data mining result, including: Processing the a-1th transition emergency monitoring data by the feature extraction unit of the ath encoding unit to obtain the ath emergency monitoring hidden danger description content; The a-th emergency monitoring hidden danger description content is downsampled by the encoding unit of the a-th encoding unit to obtain the a-th emergency monitoring data mining result.

[0008] In the present application, the decision unit includes at least one dimensionless simplification unit, an important content identification unit and a data debugging unit; the ath emergency monitoring data mining result and the analysis result feature are processed by the ath decision unit to obtain the ath transition emergency monitoring data, including: Processing the ath emergency monitoring data mining result by a first dimensionless simplification unit to obtain a first dimensionless simplification result; Processing the first dimensionless simplification result and the analysis result feature by the important content identification unit to obtain first candidate hidden danger point monitoring data; The first candidate potential risk point monitoring data and the ath emergency monitoring data mining result are fused to obtain the second candidate potential risk point monitoring data; Processing the second candidate hidden danger point monitoring data by a second dimensionless simplification unit to obtain a second dimensionless simplification result; Processing the second dimensionless simplification result by the data debugging unit to obtain third candidate hidden danger point monitoring data; The second candidate hidden danger point monitoring data and the third candidate hidden danger point monitoring data are fused to obtain the ath transition emergency monitoring data.

[0009] In the present application, the important content identification unit includes at least one feature extraction unit; the first dimensionless simplification result and the analysis result feature are processed by the important content identification unit to obtain first candidate hidden danger point monitoring data, including: Performing data simplification processing on the first dimensionless simplified result by a first feature extraction unit to obtain a first simplified result; Performing data simplification processing on the first dimensionless simplified result by a second feature extraction unit to obtain a second simplified result; Performing a first function processing on the first simplified result and the second simplified result to obtain important content identification emergency monitoring data, wherein the important content identification emergency monitoring data includes at least one fall risk factor and a matching relationship between each fall risk factor and other fall risk factors; Processing the first dimensionless simplified result by a third feature extraction unit to obtain important hidden danger description features; Performing second function processing on the important hidden danger description features and the analysis result features to obtain transition hidden danger point monitoring data; The first candidate potential risk point monitoring data is determined based on the transition potential risk point monitoring data and the important content identification emergency monitoring data.

[0010] In the present application, the first candidate risk point monitoring data is determined according to the transition risk point monitoring data and the important content identification emergency monitoring data, including: Simplifying the transition potential risk point monitoring data to obtain a third simplified result; Performing the first function processing on the third simplified result and the important content recognition result set to obtain a fourth simplified result; The fourth simplified result is repaired to obtain the first candidate hidden danger point monitoring data.

[0011] In the present application, the at least one compression unit and the at least one decision unit process the at least one transition emergency monitoring data and the analysis result characteristics to obtain the target warning parameters corresponding to each fall hazard factor, including: Processing the Xth transition emergency monitoring data and the analysis result characteristics through the X+1th decision unit to obtain the first characteristic hidden danger point monitoring data; Fusing the first characteristic hidden danger point monitoring data and the Xth transition emergency monitoring data to obtain the first hidden danger data fusion result; The b-1th hidden danger data fusion result and the analysis result feature are processed by the X+bth decision unit to obtain the bth characteristic hidden danger point monitoring data, where b is 2, 3, ..., X in sequence, and the b-1th hidden danger data fusion result is obtained by fusing the b-1th characteristic hidden danger point monitoring data and the Xb-1th transition emergency monitoring data; Processing the b-th characteristic hidden danger point monitoring data through the b-th compression unit to obtain a b-th stripe hidden danger data fusion result; According to the fusion result of the X-th stripe hidden danger data, the target warning parameters corresponding to each fall hidden danger element are determined.

[0012] In the present application, the first emergency monitoring data is analyzed and processed to obtain second emergency monitoring data and analysis result characteristics, including: Determine an analysis thread, wherein the analysis thread includes Z residual units and Z loss units, where Z is an integer greater than or equal to 1; Performing Z encoding unit processing on the first emergency monitoring data through the Z residual units to obtain key description features corresponding to each fall risk factor; Performing Z compression processing on the key description features corresponding to each fall risk factor through the Z loss units to obtain the second emergency monitoring data; Determine the characteristic emergency monitoring data output by the Zth loss unit as the analysis result feature.

[0013] In the present application, according to the target warning parameters corresponding to each fall risk factor, the first emergency monitoring data is subjected to warning processing to obtain fall risk prompt data, including: Obtaining original emergency monitoring data corresponding to the first emergency monitoring data; determining a percentage between the first emergency monitoring data and the original emergency monitoring data; Determine a set warning parameter corresponding to each fall risk factor according to the percentage and the target warning parameter corresponding to each fall risk factor; In the first emergency monitoring data, determining the attribute value of each fall risk factor; For any fall hazard factor, determine the product of the attribute value of the fall hazard factor and the set warning parameter corresponding to the fall hazard factor as the target attribute value, and obtain the target attribute value corresponding to each fall hazard factor; Determine that the fall hazard warning data includes at least one fall hazard element and a target attribute value corresponding to each fall hazard element.

[0014] In a second aspect, an emergency fall prevention monitoring system for emergency monitoring is provided, comprising a processor and a memory that communicate with each other, wherein the processor is used to read a computer program from the memory and execute it to implement the above method.

[0015] The embodiment of the present application provides an emergency monitoring method and system for emergency fall prevention, obtains the first emergency monitoring data that needs to be processed, and the first emergency monitoring data includes no less than one fall hidden danger element. The first emergency monitoring data is analyzed and processed to obtain the second emergency monitoring data and the analysis result characteristics, and the second emergency monitoring data is used to represent the description details of each attribute in the first emergency monitoring data. The first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics are processed to obtain the target warning parameters corresponding to each fall hidden danger element. According to the target warning parameters corresponding to each fall hidden danger element, the first emergency monitoring data is warned and processed to obtain the fall hidden danger prompt data. In the above process, the first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics can be processed to obtain the target warning parameters corresponding to each fall hidden danger element. And according to the target warning parameters corresponding to each fall hidden danger element, the first emergency monitoring data is warned and processed to obtain the fall hidden danger prompt data. It is possible to prompt the hidden danger point in advance, so as to reduce the risk of falling injuries for patients and their families. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A flow chart of an emergency fall prevention monitoring method for emergency monitoring provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to better understand the above technical scheme, the technical scheme of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0019] See also Figure 1 , shows an emergency fall prevention monitoring method for emergency monitoring, which may include the technical solutions described in the following steps S201-S204.

[0020] S201. Acquire first emergency monitoring data that needs to be processed.

[0021] Exemplarily, the first emergency monitoring data is obtained through monitoring. For example, the camera device may be a smart camera or a video camera. The execution subject of the embodiment of the present application may be an emergency monitoring data processing device, or an emergency monitoring data processing device disposed in the emergency monitoring data processing device.

[0022] The first emergency monitoring data includes at least one fall risk factor.

[0023] Factors that may cause risks of falling may include: water on the ground, uneven surfaces, etc.

[0024] The first emergency monitoring data to be processed may be obtained in the following manner: obtaining original emergency monitoring data and a preset data volume; and performing set processing on the original emergency monitoring data according to the preset data volume to obtain the first emergency monitoring data.

[0025] The preset data volume can be set in advance and stored in the preset storage space of the emergency monitoring data processing device. The preset data volume is smaller than the data volume of the original emergency monitoring data, so that the amount of calculation in the emergency monitoring data processing process can be reduced.

[0026] The original emergency monitoring data is obtained by the camera in the emergency monitoring data processing device, or the emergency monitoring data is obtained from other devices. There is a stroboscopic phenomenon in the original emergency monitoring data. When the original emergency monitoring data is set to a preset data volume, the resolution of the first emergency monitoring data is also adjusted to the resolution corresponding to the preset data volume.

[0027] S202: Analyze and process the first emergency monitoring data to obtain second emergency monitoring data and analysis result characteristics.

[0028] Exemplarily, the analysis result characteristics can be understood as analyzing the potential danger points and potential danger locations of the monitoring data, locations with large flow of people, etc., or caused by emergencies.

[0029] The second emergency monitoring data is used to represent the description details of each attribute in the first emergency monitoring data.

[0030] The first emergency monitoring data can be analyzed and processed in the following manner to obtain the second emergency monitoring data and analysis result characteristics: determine the analysis thread, the analysis thread includes Z residual units and Z loss units, Z is an integer greater than or equal to 1; perform Z encoding unit processing on the first emergency monitoring data through the Z residual units to obtain the key description features corresponding to each fall hazard factor; perform Z compression processing on the key description features corresponding to each fall hazard factor through the Z loss units to obtain the second emergency monitoring data; determine the characteristic emergency monitoring data output by the Zth loss unit as the analysis result feature.

[0031] S203: Process the first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics to obtain target warning parameters corresponding to each fall risk factor.

[0032] Exemplarily, the target warning parameter can be understood as being obtained according to the score of the hidden danger point, and the danger of each hidden danger corresponds to a separate warning parameter.

[0033] The first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics can be processed in the following manner to obtain the target warning parameters corresponding to each fall hazard factor: determine an artificial intelligence recognition thread, the artificial intelligence recognition thread includes no less than one derivative unit, no less than one compression unit, and no less than one decision unit; process the first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics by no less than one derivative unit and no less than one decision unit to obtain no less than one transitional emergency monitoring data; process no less than one transitional emergency monitoring data and the analysis result characteristics by no less than one compression unit and no less than one decision unit to obtain the target warning parameters corresponding to each fall hazard factor.

[0034] S204: Perform warning processing on the first emergency monitoring data according to the target warning parameters corresponding to each fall risk factor to obtain fall risk prompt data.

[0035] Exemplarily, the warning processing is based on the target warning parameter. For example, when the target warning parameter is greater than a threshold, the warning processing will be performed, and when it is less than the threshold, the warning processing will not be performed.

[0036] The first emergency monitoring data can be processed for warning according to the target warning parameters corresponding to each fall hazard factor to obtain fall hazard prompt data in the following manner: obtain the original emergency monitoring data corresponding to the first emergency monitoring data; determine the percentage between the first emergency monitoring data and the original emergency monitoring data; determine the set warning parameters corresponding to each fall hazard factor according to the percentage and the target warning parameters corresponding to each fall hazard factor; determine the attribute value of each fall hazard factor in the first emergency monitoring data; for any fall hazard factor, determine the product of the attribute value of the fall hazard factor and the set warning parameters corresponding to the fall hazard factor as the target attribute value, and obtain the target attribute value corresponding to each fall hazard factor; determine that the fall hazard prompt data includes no less than one fall hazard factor, and the target attribute value corresponding to each fall hazard factor.

[0037] The emergency anti-fall monitoring method of emergency monitoring provided in the embodiment of the present application obtains the first emergency monitoring data that needs to be processed. The first emergency monitoring data is analyzed and processed to obtain the second emergency monitoring data and the analysis result characteristics. The first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics are processed to obtain the target warning parameters corresponding to each fall hidden danger element. According to the target warning parameters corresponding to each fall hidden danger element, the first emergency monitoring data is subjected to early warning processing to obtain the fall hidden danger prompt data. In the above process, the first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics can be processed to obtain the target warning parameters corresponding to each fall hidden danger element. And according to the target warning parameters corresponding to each fall hidden danger element, the first emergency monitoring data is subjected to early warning processing to obtain the fall hidden danger prompt data. The risk of falling can be reduced as much as possible.

[0038] Another emergency fall prevention monitoring method for emergency monitoring provided by an embodiment of the present application includes: S401. Acquire first emergency monitoring data that needs to be processed.

[0039] It is understandable that the execution process of S401 can refer to S201 and will not be repeated here.

[0040] S402: Analyze and process the first emergency monitoring data to obtain second emergency monitoring data and analysis result characteristics.

[0041] For example, the emergency monitoring data processing device obtains the first emergency monitoring data to be processed as emergency monitoring data B. The emergency monitoring data processing device analyzes and processes the emergency monitoring data B through the analysis thread to obtain the second emergency monitoring data B1 and the analysis result feature B.

[0042] S403: Determine an artificial intelligence recognition thread.

[0043] Exemplarily, the artificial intelligence recognition thread may include CNN, etc.

[0044] The artificial intelligence recognition thread includes no less than one derivation unit, no less than one compression unit, and no less than one decision unit.

[0045] S404: Integrate the first emergency monitoring data and the second emergency monitoring data to obtain third emergency monitoring data.

[0046] The first emergency monitoring data and the second emergency monitoring data are integrated and processed so that the first emergency monitoring data and the second emergency monitoring data are spliced ​​in the trajectory dimension to obtain the third emergency monitoring data. The number of trajectories of the third emergency monitoring data is the sum of the number of trajectories of the first emergency monitoring data and the number of trajectories of the second emergency monitoring data.

[0047] S405. Process the third emergency monitoring data through the first derivative unit to obtain a first emergency monitoring data mining result.

[0048] The derivative unit can perform convolution processing and downsampling processing on the third emergency monitoring data, and the height, width and number of trajectories of the first emergency monitoring data mining result can be determined according to the parameters of the first derivative unit in the artificial intelligence recognition thread.

[0049] S406. Process the mining result and analysis result features of the first emergency monitoring data through the first decision unit to obtain the first transitional emergency monitoring data.

[0050] S407. Initialize a to 2.

[0051] S408. Process the a-1th transition emergency monitoring data through the ath derivative unit to obtain the ath emergency monitoring data mining result.

[0052] The a-1th transition emergency monitoring data can be processed by the a-th derivative unit to obtain the a-th emergency monitoring data mining result: the a-1th transition emergency monitoring data is processed by the feature extraction unit of the a-th coding unit to obtain the a-th emergency monitoring hidden danger description content; the a-th emergency monitoring hidden danger description content is derive-processed by the coding unit of the a-th coding unit to obtain the a-th emergency monitoring data mining result.

[0053] It can be understood that the processing process of the first derivative unit is the same as the processing process of the ath derivative unit.

[0054] S409. Process the mining result and analysis result features of the a-th emergency monitoring data through the a-th decision unit to obtain the a-th transition emergency monitoring data.

[0055] The decision unit includes no less than one dimensionless simplification unit, an important content identification unit and a data debugging unit.

[0056] The ath emergency monitoring data mining result and analysis result characteristics can be processed by the ath decision unit to obtain the ath transition emergency monitoring data in the following manner: the ath emergency monitoring data mining result is processed by the first dimensionless simplification unit to obtain the first dimensionless simplification result; the first dimensionless simplification result and analysis result characteristics are processed by the important content recognition unit to obtain the first candidate hidden danger point monitoring data; the first candidate hidden danger point monitoring data and the ath emergency monitoring data mining result are fused to obtain the second candidate hidden danger point monitoring data; the second candidate hidden danger point monitoring data is processed by the second dimensionless simplification unit to obtain the second dimensionless simplification result; the second dimensionless simplification result is processed by the data debugging unit to obtain the third candidate hidden danger point monitoring data; the second candidate hidden danger point monitoring data and the third candidate hidden danger point monitoring data are fused to obtain the ath transition emergency monitoring data.

[0057] The height, width, and number of trajectories of the a-th transition emergency monitoring data are the same as the height, width, and number of trajectories of the a-th emergency monitoring data mining result.

[0058] Below, the processing process of the decision unit is described. The processing process of the decision unit provided in the embodiment of the present application includes a decision unit. The decision unit includes two dimensionless simplification units, an important content identification unit and a data debugging unit. The emergency monitoring data mining result is processed by the first dimensionless simplification unit of the decision unit to obtain a first dimensionless simplification result. The first dimensionless simplification result and the analysis result feature are processed by the important content identification unit of the decision unit to obtain the first candidate hidden danger point monitoring data. The first candidate hidden danger point monitoring data and the ath emergency monitoring data mining result are fused to obtain the second candidate hidden danger point monitoring data. The second candidate hidden danger point monitoring data is processed by the second dimensionless simplification unit of the decision unit to obtain a second dimensionless simplification result. The second dimensionless simplification result is processed by the data debugging unit of the decision unit to obtain the third candidate hidden danger point monitoring data. The second candidate hidden danger point monitoring data and the third candidate hidden danger point monitoring data are fused to obtain transitional emergency monitoring data.

[0059] The important content identification unit includes at least one feature extraction unit. The first dimensionless simplification result and the analysis result feature can be processed by the important content identification unit in the following manner to obtain the first candidate hidden danger point monitoring data: the first dimensionless simplification result is processed by the first feature extraction unit to obtain the first simplified result; the first dimensionless simplification result is processed by the second feature extraction unit to obtain the second simplified result; the first simplified result and the second simplified result are processed by the first function to obtain the important content identification emergency monitoring data, and the important content identification emergency monitoring data includes at least one fall hidden danger element, and the matching relationship between each fall hidden danger element and the remaining fall hidden danger elements; the first dimensionless simplification result is processed by the third feature extraction unit to obtain the important hidden danger description feature; the important hidden danger description feature and the analysis result feature are processed by the second function to obtain the transition hidden danger point monitoring data; the first candidate hidden danger point monitoring data is determined according to the transition hidden danger point monitoring data and the important content identification emergency monitoring data.

[0060] The first candidate hidden danger point monitoring data can be determined based on the transition hidden danger point monitoring data and the important content identification emergency monitoring data in the following manner: simplify the transition hidden danger point monitoring data to obtain a third simplified result; perform a first function processing on the third simplified result and the important content identification result set to obtain a fourth simplified result; and repair the fourth simplified result to obtain the first candidate hidden danger point monitoring data.

[0061] S410: Determine whether a is equal to X.

[0062] If so, execute S411.

[0063] If not, update a to a+1 and execute S408.

[0064] S411. Determine that at least one transitional emergency monitoring data includes X transitional emergency monitoring data.

[0065] S412. Process at least one transition emergency monitoring data and analysis result characteristics through at least one compression unit and at least one decision unit to obtain target warning parameters corresponding to each fall hazard factor.

[0066] The target warning parameters corresponding to each fall hazard element can be obtained by processing at least one transition emergency monitoring data and analysis result characteristics through at least one compression unit and at least one decision unit as follows: the Xth transition emergency monitoring data and analysis result characteristics are processed by the X+1th decision unit to obtain the first characteristic hidden danger point monitoring data; the first characteristic hidden danger point monitoring data and the Xth transition emergency monitoring data are fused to obtain the first hidden danger data fusion result; the b-1th hidden danger data fusion result and analysis result characteristics are processed by the X+bth decision unit to obtain the bth characteristic hidden danger point monitoring data, where b is 2, 3, ..., X in sequence, and the b-1th hidden danger data fusion result is obtained by fusion processing of the b-1th characteristic hidden danger point monitoring data and the Xb-1th transition emergency monitoring data; the bth characteristic hidden danger point monitoring data is processed by the bth compression unit to obtain the bth stripe hidden danger data fusion result; and the target warning parameters corresponding to each fall hazard element are determined according to the Xth stripe hidden danger data fusion result.

[0067] It can be understood that the processing process of each decision unit in the analysis thread provided in the embodiment of the present application is the same and will not be repeated here.

[0068] S413. Perform warning processing on the first emergency monitoring data according to the target warning parameters corresponding to each fall risk factor to obtain fall risk prompt data.

[0069] Below, the process of obtaining fall hidden danger prompt data provided by the embodiment of the present application is shown, including the first emergency monitoring data and the fall hidden danger prompt data. There is a dark stripe area in the first emergency monitoring data. After the analysis thread and the artificial intelligence recognition thread process the first emergency monitoring data, the target warning parameters corresponding to each fall hidden danger element are obtained. And according to the target warning parameters corresponding to each fall hidden danger element, the first emergency monitoring data is warned and the fall hidden danger prompt data is obtained.

[0070] Before processing emergency monitoring data through the analysis thread and the artificial intelligence recognition thread, the analysis thread and the artificial intelligence recognition thread need to be trained to determine the parameters of the analysis thread and the artificial intelligence recognition thread.

[0071] Thread training can be performed in the following manner: obtaining a training data set, the training data set including at least one sample emergency monitoring data, analysis emergency monitoring data corresponding to each sample emergency monitoring data, at least one standard warning parameter corresponding to each sample emergency monitoring data, and standard emergency monitoring data corresponding to each sample emergency monitoring data; determining an original thread, the original thread including an analysis thread and an artificial intelligence recognition thread; performing the first iterative training on the original thread using the training data set to obtain the first transition thread; performing the a+1th iterative training on the ath transition thread using the training data set to obtain the a+1th transition thread, where a is successively 1, 2, 3, ... until the ath transition thread converges, and determining the ath transition thread as a target thread, the target thread including the trained analysis thread and artificial intelligence recognition thread.

[0072] The a+1th transition thread can be obtained by performing the a+1th iterative training on the a-th transition thread through the training data set in the following manner: the sample emergency monitoring data in the training data set is processed by the a-th transition thread to obtain a prediction result, which includes the predicted emergency monitoring hazard description content corresponding to the sample emergency monitoring data, at least one predicted warning parameter, and predicted emergency monitoring data; the loss value is determined according to the analyzed emergency monitoring data corresponding to the sample emergency monitoring data, at least one standard warning parameter, standard emergency monitoring data and the prediction result; the thread parameters of the a-th transition thread are updated according to the loss value to obtain the a+1-th transition thread.

[0073] The emergency anti-fall monitoring method for emergency monitoring provided in the embodiment of the present application obtains the first emergency monitoring data that needs to be processed. The first emergency monitoring data is analyzed and processed to obtain the second emergency monitoring data and the analysis result characteristics. Determine the artificial intelligence recognition thread. The first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics are processed by no less than one derivative unit and no less than one decision unit to obtain no less than one transition emergency monitoring data. No less than one transition emergency monitoring data and the analysis result characteristics are processed by no less than one compression unit and no less than one decision unit to obtain the target warning parameters corresponding to each fall hidden danger element. According to the target warning parameters corresponding to each fall hidden danger element, the first emergency monitoring data is warned to obtain fall hidden danger prompt data.

[0074] The emergency monitoring data processing process provided in the embodiment of the present application includes an analysis thread and an artificial intelligence recognition thread. The analysis thread and the artificial intelligence recognition thread are set in an emergency monitoring data processing device. The emergency monitoring data processing device can be a camera device, a terminal device, etc. For example, the camera device can be a smart camera or a video camera. The terminal device can be a mobile phone, a tablet computer, a computer, etc. The emergency monitoring data processing device obtains the original emergency monitoring data and a preset data volume, and sets the original emergency monitoring data according to the preset data volume to obtain the first emergency monitoring data.

[0075] The emergency monitoring data processing process provided in the embodiment of the present application obtains the first emergency monitoring data that needs to be processed. The first emergency monitoring data is analyzed and processed to obtain the second emergency monitoring data and the analysis result characteristics. The artificial intelligence recognition thread is determined. The first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics are processed by no less than one derivative unit and no less than one decision unit to obtain no less than one transition emergency monitoring data. No less than one transition emergency monitoring data and the analysis result characteristics are processed by no less than one compression unit and no less than one decision unit to obtain the target warning parameters corresponding to each fall hazard factor.

[0076] On the basis of the above, an emergency fall prevention monitoring device for emergency monitoring is provided, and the device comprises: A data acquisition module, used for acquiring first emergency monitoring data to be processed, wherein the first emergency monitoring data includes at least one fall risk factor; a feature obtaining module, used for analyzing and processing the first emergency monitoring data to obtain second emergency monitoring data and analysis result features, wherein the second emergency monitoring data is used to represent the description details of each attribute in the first emergency monitoring data; a parameter obtaining module, configured to process the first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics to obtain a target warning parameter corresponding to each fall risk factor; The data prompt module is used to perform warning processing on the first emergency monitoring data according to the target warning parameters corresponding to each fall risk factor to obtain fall risk prompt data.

[0077] Based on the above, an emergency fall prevention monitoring system for emergency monitoring is shown, including a processor and a memory that communicate with each other, and the processor is used to read and execute a computer program from the memory to implement the above method.

[0078] Based on the above, a computer-readable storage medium is also provided, on which a computer program stored implements the above method when running.

[0079] In summary, based on the above scheme, the first emergency monitoring data that needs to be processed is obtained, and the first emergency monitoring data includes no less than one fall hidden danger element. The first emergency monitoring data is analyzed and processed to obtain the second emergency monitoring data and the analysis result characteristics, and the second emergency monitoring data is used to represent the description details of each attribute in the first emergency monitoring data. The first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics are processed to obtain the target warning parameters corresponding to each fall hidden danger element. According to the target warning parameters corresponding to each fall hidden danger element, the first emergency monitoring data is warned and processed to obtain the fall hidden danger prompt data. In the above process, the first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics can be processed to obtain the target warning parameters corresponding to each fall hidden danger element. And according to the target warning parameters corresponding to each fall hidden danger element, the first emergency monitoring data is warned and processed to obtain the fall hidden danger prompt data. It is possible to prompt the hidden danger point in advance, so as to reduce the risk of falling injuries for patients and their families.

[0080] It should be understood that the system and its modules shown above can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art will understand that the above methods and systems can be implemented using computer executable instructions and / or included in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software such as executed by various types of processors, and can also be implemented by a combination of the above hardware circuits and software (e.g., firmware).

[0081] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or any other possible beneficial effects.

Claims

1. An emergency fall prevention monitoring method for emergency monitoring, characterized in that: include: Acquire first emergency monitoring data that needs to be processed, wherein the first emergency monitoring data includes at least one fall risk factor; Analyze and process the first emergency monitoring data to obtain second emergency monitoring data and analysis result characteristics, wherein the second emergency monitoring data is used to represent description details of each attribute in the first emergency monitoring data; Processing the first emergency monitoring data, the second emergency monitoring data, and the analysis result characteristics to obtain a target warning parameter corresponding to each fall risk factor; According to the target warning parameters corresponding to each fall risk factor, the first emergency monitoring data is subjected to warning processing to obtain fall risk prompt data.

2. The method according to claim 1, characterized in that The first emergency monitoring data, the second emergency monitoring data and the analysis result characteristics are processed to obtain target warning parameters corresponding to each fall risk factor, including: Determining an artificial intelligence recognition thread, wherein the artificial intelligence recognition thread includes at least one derivation unit, at least one compression unit, and at least one decision unit; Processing the first emergency monitoring data, the second emergency monitoring data and the analysis result feature through the at least one derivation unit and the at least one decision unit to obtain at least one transition emergency monitoring data; The at least one compression unit and the at least one decision unit are used to process the at least one transition emergency monitoring data and the analysis result characteristics to obtain target warning parameters corresponding to each fall risk factor.

3. The method according to claim 2, characterized in that The first emergency monitoring data, the second emergency monitoring data and the analysis result feature are processed by the at least one derivation unit and the at least one decision unit to obtain at least one transition emergency monitoring data, including: Integrate the first emergency monitoring data and the second emergency monitoring data to obtain third emergency monitoring data; Processing the third emergency monitoring data through a first derivative unit to obtain a first emergency monitoring data mining result; Processing the first emergency monitoring data mining result and the analysis result feature by a first decision unit to obtain first transition emergency monitoring data; Processing the a-1th transition emergency monitoring data through the ath derivative unit to obtain the ath emergency monitoring data mining result; The ath decision unit processes the ath emergency monitoring data mining result and the analysis result feature to obtain the ath transition emergency monitoring data, where a is 2, 3, ..., X in sequence, and X is the number of the derivative units; Determining that the at least one transitional emergency monitoring data includes the X transitional emergency monitoring data.

4. The method according to claim 3, characterized in that The derivation unit includes a feature extraction unit and an encoding unit; the a-1th transition emergency monitoring data is processed by the a-th derivation unit to obtain the a-th emergency monitoring data mining result, including: Processing the a-1th transition emergency monitoring data by the feature extraction unit of the ath encoding unit to obtain the ath emergency monitoring hidden danger description content; The a-th emergency monitoring hidden danger description content is downsampled by the encoding unit of the a-th encoding unit to obtain the a-th emergency monitoring data mining result.

5. The method according to claim 3 or 4, characterized in that: The decision unit includes at least one dimensionless simplification unit, an important content identification unit and a data debugging unit; the ath emergency monitoring data mining result and the analysis result feature are processed by the ath decision unit to obtain the ath transition emergency monitoring data, including: Processing the ath emergency monitoring data mining result by a first dimensionless simplification unit to obtain a first dimensionless simplification result; Processing the first dimensionless simplification result and the analysis result feature by the important content identification unit to obtain first candidate hidden danger point monitoring data; The first candidate potential risk point monitoring data and the ath emergency monitoring data mining result are fused to obtain the second candidate potential risk point monitoring data; Processing the second candidate hidden danger point monitoring data by a second dimensionless simplification unit to obtain a second dimensionless simplification result; Processing the second dimensionless simplification result by the data debugging unit to obtain third candidate hidden danger point monitoring data; The second candidate hidden danger point monitoring data and the third candidate hidden danger point monitoring data are fused to obtain the ath transition emergency monitoring data.

6. The method according to claim 5, characterized in that The important content identification unit includes at least one feature extraction unit; the first dimensionless simplification result and the analysis result feature are processed by the important content identification unit to obtain first candidate hidden danger point monitoring data, including: Performing data simplification processing on the first dimensionless simplified result by a first feature extraction unit to obtain a first simplified result; Performing data simplification processing on the first dimensionless simplified result by a second feature extraction unit to obtain a second simplified result; Performing a first function processing on the first simplified result and the second simplified result to obtain important content identification emergency monitoring data, wherein the important content identification emergency monitoring data includes at least one fall risk factor and a matching relationship between each fall risk factor and other fall risk factors; Processing the first dimensionless simplified result by a third feature extraction unit to obtain important hidden danger description features; Performing second function processing on the important hidden danger description features and the analysis result features to obtain transition hidden danger point monitoring data; Determine the first candidate potential risk point monitoring data according to the transition potential risk point monitoring data and the important content identification emergency monitoring data; Wherein, determining the first candidate risk point monitoring data according to the transition risk point monitoring data and the important content identification emergency monitoring data includes: Simplifying the transition potential risk point monitoring data to obtain a third simplified result; Performing the first function processing on the third simplified result and the important content recognition result set to obtain a fourth simplified result; The fourth simplified result is repaired to obtain the first candidate hidden danger point monitoring data.

7. The method according to claim 2, characterized in that The at least one compression unit and the at least one decision unit process the at least one transition emergency monitoring data and the analysis result characteristics to obtain target warning parameters corresponding to each fall risk factor, including: Processing the Xth transition emergency monitoring data and the analysis result characteristics through the X+1th decision unit to obtain the first characteristic hidden danger point monitoring data; Fusing the first characteristic hidden danger point monitoring data and the Xth transition emergency monitoring data to obtain the first hidden danger data fusion result; The b-1th hidden danger data fusion result and the analysis result feature are processed by the X+bth decision unit to obtain the bth characteristic hidden danger point monitoring data, where b is 2, 3, ..., X in sequence, and the b-1th hidden danger data fusion result is obtained by fusing the b-1th characteristic hidden danger point monitoring data and the Xb-1th transition emergency monitoring data; Processing the b-th characteristic hidden danger point monitoring data through the b-th compression unit to obtain a b-th stripe hidden danger data fusion result; According to the fusion result of the X-th stripe hidden danger data, the target warning parameters corresponding to each fall hidden danger element are determined.

8. The method according to any one of claims 1 to 4, characterized in that: The first emergency monitoring data is analyzed and processed to obtain second emergency monitoring data and analysis result characteristics, including: Determine an analysis thread, wherein the analysis thread includes Z residual units and Z loss units, where Z is an integer greater than or equal to 1; Performing Z encoding unit processing on the first emergency monitoring data through the Z residual units to obtain key description features corresponding to each fall risk factor; Performing Z compression processing on the key description features corresponding to each fall risk factor through the Z loss units to obtain the second emergency monitoring data; Determine the characteristic emergency monitoring data output by the Zth loss unit as the analysis result feature.

9. The method according to any one of claims 1 to 4, characterized in that: According to the target warning parameters corresponding to each fall risk factor, the first emergency monitoring data is subjected to warning processing to obtain fall risk prompt data, including: Obtaining original emergency monitoring data corresponding to the first emergency monitoring data; determining a percentage between the first emergency monitoring data and the original emergency monitoring data; Determine a set warning parameter corresponding to each fall risk factor according to the percentage and the target warning parameter corresponding to each fall risk factor; In the first emergency monitoring data, determining the attribute value of each fall risk factor; For any fall hazard factor, determine the product of the attribute value of the fall hazard factor and the set warning parameter corresponding to the fall hazard factor as the target attribute value, and obtain the target attribute value corresponding to each fall hazard factor; Determine that the fall hazard warning data includes at least one fall hazard element and a target attribute value corresponding to each fall hazard element.

10. An emergency fall prevention monitoring system for emergency monitoring, characterized in that: The invention comprises a processor and a memory communicating with each other, wherein the processor is used to read a computer program from the memory and execute the computer program to implement the method according to any one of claims 1 to 9.

Images