Intelligent early warning management system and implementation method

Abstract

The present invention provides an intelligent early warning management system, including an intelligent recognition module, a forensics module, and an identification module; the intelligent recognition module is used to collect information factors of different objects one by one through an intelligent device, and the information factors are input into the system in real time after being collected; the forensics module is used to judge and preliminarily screen out the early warning objects according to certain rules based on the information factors collected by the intelligent recognition module, and store clue information formed by the early warning objects in the process of being identified and collected forensics; the identification module is used to perform monitoring value judgment according to the early warning objects in the forensics module and output the target objects according to certain principles. The present invention can improve the accuracy of the intelligent early warning management system for the scope of early warning objects, and effectively expand the application scope of the intelligent early warning management system.

Description

Technical Field

[0001] The present invention relates to the field of intelligent early warning, and in particular to an intelligent early warning management system and an implementation method thereof. Background Art

[0002] At present, in some transportation fields, although intelligent early warning management systems have been applied, their screening methods for identifying objects are too simple, which often leads to a large number of early warning objects, which is not conducive to accurate monitoring, or there are too few early warning objects, which makes it impossible to achieve the purpose of early warning. On the other hand, the design of the system is relatively simple, resulting in extremely limited application scenarios, which is not conducive to the effective promotion and application of intelligent early warning management systems.

[0003] In view of this, it is necessary to improve the intelligent early warning management system in the prior art to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to improve the accuracy of the intelligent early warning management system for the scope of early warning objects and effectively expand the application scope of the intelligent early warning management system.

[0005] To achieve the above-mentioned purpose, the present invention provides an intelligent early warning management system, including an intelligent recognition module, a forensics module, and an identification module;

[0006] The intelligent recognition module is used to collect information factors of different objects one by one through intelligent devices, and the information factors are input and managed into the system in real time after being collected;

[0007] The evidence collection module is used to judge and preliminarily screen out the warning objects according to certain rules based on the information factors collected by the intelligent recognition module, and store the clue information formed in the process of identifying and collecting evidence of the warning objects;

[0008] The identification module is used to perform monitoring value judgment according to the warning object in the evidence collection module and output the target object according to certain principles.

[0009] As a further improvement of the present invention, the intelligent identification module includes an extraction module for collecting information factors of the object, and an association module for implementing association setting of information factors from different objects.

[0010] As a further improvement of the present invention, the evidence collection module includes a determination module for screening and determining information factors, a storage module for storing and managing the information factors involved in the determination module, and a warning object output module for outputting objects according to the determination conclusions of the determination module.

[0011] As a further improvement of the present invention, the identification module includes an information flow for organizing and verifying input information factors, a monitoring value determination module for determining the monitoring value of the object output by the warning object output module in combination with the information flow, and a target identification module for determining the target value of the warning object with a certain monitoring value output by the monitoring value determination module.

[0012] As a further improvement of the present invention, the determination module includes a weight acquisition module for assigning weight values ​​to information factors to obtain key factors, and a threshold calculation module for performing threshold calculation based on the key factors obtained by the weight acquisition module and determining the overflow factor.

[0013] As a further improvement of the present invention, the threshold calculation method of the threshold calculation module is a single threshold setting method, a double threshold setting method, an association threshold setting method, a similarity threshold setting method, and a factor accumulation threshold setting method;

[0014] The single threshold setting method is to pre-set a threshold for a certain key factor, and when the value of the key factor reaches the preset threshold, the key factor is determined to be an overflow factor;

[0015] The dual threshold setting method is to pre-set an upper and lower threshold for a certain key factor to form a threshold interval. When the value of the key factor exceeds the threshold interval, the key factor is determined to be an overflow factor.

[0016] The associated threshold setting method is to pre-set a threshold for two or more key factors respectively, and set one of the key factors as the main factor. When the values ​​of the corresponding key factors reach the corresponding preset thresholds, the main factor is determined to be an overflow factor.

[0017] The similarity threshold setting method is to set the floating interval of a certain key factor in different ranges as a similarity value. When the frequency of occurrence of a certain similarity value reaches a preset threshold, the corresponding key factor is determined to be an overflow factor.

[0018] The factor accumulation threshold setting method is to set corresponding thresholds for more than one key factor, and set one of the key factors as the main factor. When all the key factors reach the preset threshold, the corresponding main factor is determined to be an overflow factor.

[0019] As a further improvement of the present invention, the single threshold setting method, the dual threshold setting method, the association threshold setting method, the similarity threshold setting method and the factor accumulation threshold setting method can be applied separately or two or more methods can be applied in combination.

[0020] The present application also provides a method for implementing an intelligent early warning management system, comprising the following steps:

[0021] S1 collects and integrates information factors of the identification object through the extraction module of the intelligent recognition module;

[0022] S2 uses the weight acquisition module to assign weights to the information factors of the screening object and obtain key factors;

[0023] The S3 key factor passes through the threshold calculation module to determine whether the identification object is added to the warning object output module;

[0024] S4 determines the monitoring value of the object through the monitoring value determination module according to the warning object output module and the information flow;

[0025] S5 stores the judgment clues of the warning objects with certain monitoring value through the storage module;

[0026] S6 outputs the target object through the target identification module.

[0027] As a further improvement of the present invention, the S5 stores the judgment clues of the warning objects with certain monitoring value through the storage module; at the same time, according to the association setting in the association module, determines the associated objects with the warning objects, and stores the information factors of the associated objects.

[0028] Compared with the prior art, the beneficial effects of the present invention are: by using a variety of threshold calculation methods alone or in combination, the needs of the intelligent early warning management system in different application scenarios are met, the scope of application is greatly expanded, and it is extremely conducive to promotion; through multi-level management of threshold calculation judgment, monitoring value judgment, and flexible positioning of target values, the accuracy of the warning object range is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is the structural diagram of the system;

[0030] Figure 2 This is a schematic diagram of the implementation process of this system;

[0031] Figure 3 A step diagram for implementing the method of this system. DETAILED DESCRIPTION

[0032] The present invention is described in detail below in conjunction with the various embodiments shown in the accompanying drawings, but it should be noted that these embodiments are not limitations of the present invention, and any equivalent transformations or substitutions in functions, methods, or structures made by ordinary technicians in the field based on these embodiments are all within the scope of protection of the present invention.

[0033] Please refer to Figure 1 As shown, Figure 1The structure diagram of the system includes an intelligent recognition module 10, a forensics module 20, and an identification module 30. The intelligent recognition module 10 is used to collect information factors of different objects one by one through an intelligent device, and the information factors are entered into the system in real time after collection; the forensics module 20 is used to judge and preliminarily screen out the warning objects according to certain rules based on the information factors collected by the intelligent recognition module, and store the clue information formed by the warning objects in the process of identification and evidence collection; the identification module 30 is used to perform monitoring value judgment according to the warning objects in the forensics module and output the final target objects according to certain principles.

[0034] The intelligent recognition module 10 includes an extraction module 101 for collecting information factors of an object, and an association module 102 for associating and managing information factors from different objects. In practical applications, the extraction module 101 can use the face recognition technology in the prior art to intelligently collect information factors related to people with people as the collection object. These information factors may include but are not limited to face features, natural attributes, collection time, collection area, frequency of occurrence, etc.; it can also use the Internet of Things recognition in the prior art to intelligently collect information factors related to objects with objects as the collection object. Objects may include but are not limited to vehicles, airplanes, high-speed rail and other means of transportation, mobile phones, smart wearables and other identifiable objects. Similarly, the information factors collected from these objects may include but are not limited to object features, attributes, collection time, collection area, frequency of occurrence, etc. The association module 102 associates two or more objects collected in the same collection behavior. That is, when an object is identified as a warning object, through the role of the association module 102, more effective and manageable warning objects can be further identified. The simultaneous appearance of these two or more objects can be generated by a single face recognition or a single IoT recognition or a single face and IoT recognition at the same time. In actual application scenarios, for example, when entering a parking lot entrance, the driver drives the vehicle through an intelligent recognition device that combines face recognition and IoT recognition to collect information about the driver and the vehicle at the same time, and performs association settings. Then, when the driver needs to go to the parking lot to find the vehicle, when entering the parking lot, the intelligent recognition system identifies the relevant information factors of the driver, and the vehicle location information can be directly output using the association settings without the driver actively sending a request to locate the vehicle in the parking lot. In particular, since the association setting of the object is not limited to the object category, it has a special role in most pursuit scenarios, such as luggage tracking at the airport, crowd locking in public places, and even IP locking in a virtual environment.

[0035] The evidence collection module 20 includes a determination module 201 for screening and determining information factors, a storage module 203 for storing and managing information factors involved in the determination module 201, and an early warning object output module 202 for outputting objects according to the determination conclusion of the determination module 201. The determination module 201 includes a weight acquisition module 211 for assigning weight values ​​to information factors to obtain key factors, and a threshold calculation module 221 for calculating thresholds and determining overflow factors based on the key factors obtained by the weight acquisition module 211. For the weight acquisition module 211, there are preset weight levels in this system, and factors with high weight levels are identified as key factors in the system. According to the needs of the actual scenario, different weight values ​​can be assigned to different information factors to achieve different weight levels. For example, when monitoring whether there are suspicious illegal persons entering and leaving a building, if the weight value on the information factor of entry and exit frequency is set to high, then the entry and exit frequency information factor is identified as a key factor.

[0036] The threshold calculation method of the threshold calculation module 221 is a single threshold setting method, a double threshold setting method, an associated threshold setting method, a similarity threshold setting method, and a factor accumulation threshold setting method:

[0037] 1) The single threshold setting method is to pre-set a threshold for a key factor. When the value of the key factor reaches the preset threshold, the key factor is determined to be an overflow factor. It is applicable to situations where the basic determining factor for object determination is single and non-floating. For example, in the scenario of a blood index test, the normal blood index state is negative and the abnormal state is positive. Then, when this blood index is a key factor in a certain scenario, its preset threshold should be "negative". Once it is not "negative", this key factor is determined to be an "overflow factor", and then it has this non-negative value in a specific scenario.

[0038] The subjects with “negative” blood as the information factor will be identified as early warning subjects.

[0039] 2) The double threshold setting method is to pre-set an upper and lower threshold for a key factor to form a threshold interval. When the value of the key factor exceeds the threshold interval, the key factor is judged as an overflow factor. It is suitable for situations where the basic determining factors for object judgment are relatively single and have floating properties within a certain range.

[0040] 3) The associated threshold setting method is to set a threshold for two or more key factors in advance, and set one of the key factors as the main factor. When the values ​​of the corresponding key factors all reach the corresponding preset thresholds, the main factor is determined to be an overflow factor; it is applicable to situations where the basic determining factors for object determination are associated and diverse, and need to be combined for determination; for example, for the black car market, to determine whether this car is a black car and whether this driver is operating illegally, it is necessary to combine two associated factors to determine, one is the frequency of vehicle identification and capture, and the other is the frequency of driver identification and capture. The two need to be combined to determine whether there is illegal operation. In particular, the threshold setting method for different key factors can adopt a single threshold setting method or a double threshold setting method.

[0041] 4) The similarity threshold setting method is to set the floating interval of a key factor in different ranges as the similarity value. When the frequency of occurrence of a similarity value reaches the preset threshold, the corresponding key factor is determined as an overflow factor. It is applicable to special situations where a certain determining factor of object determination has a certain degree of similarity ductility, such as a person appears multiple times in a similar area (parking lot). Similarly, the preset threshold here can be a single value or an interval range.

[0042] 5) The factor accumulation threshold setting method is to set corresponding thresholds for more than one key factor, and set one of the key factors as the main factor. When all the key factors reach the preset threshold, the corresponding main factor is determined as the overflow factor. It is suitable for situations where the basic determining factors of object determination are more complex and multiple factors need to be combined to form a specific chain of evidence. In particular, the setting of the preset threshold of this method can adopt the single threshold setting method, the double threshold setting method, or the similarity threshold setting method.

[0043] The above five methods can be applied separately according to actual needs, or two or more methods can be used in combination.

[0044] The identification module 30 includes an information flow 301 for sorting and checking input information factors, a monitoring value determination module 302 for determining the monitoring value of the object output by the warning object output module 202 in combination with the information flow, and a target identification module 303 for determining the target value of the warning object with a certain monitoring value output by the monitoring value determination module 302. The so-called input information factors here refer to information factors that are not identified and collected by the intelligent identification module 10, including but not limited to external database entry, additional information source system access, etc. In particular, the monitoring value in the monitoring value determination module 302 is a specific value calculated according to a certain influence relationship based on all factors that can affect a certain object in a certain scenario identification. Therefore, in different scenario identifications, the specific calculation method of the monitoring value will be different. In addition, the target value in the target identification module 303 is determined according to the monitoring value required for different scenario identifications. It can be a specific value or a numerical range.

[0045] Combine the following Figure 2 The implementation flow diagram of the system shown in FIG. Figure 3 The step diagram of the implementation method of the system shown in the figure further illustrates the implementation method 1 of the system:

[0046] S1 collects and integrates the information factors of the identification object through the extraction module 101 of the intelligent identification module 10. In particular, when there are two or more identification objects in a certain collection behavior, the association module 102 can be started at the same time to realize the association setting. In practical applications, unique information factors of different objects can be retrieved as association factors. For example, the ID card number of a person in the face recognition system is a unique information factor; the license plate number of a vehicle in the Internet of Things identification system is a unique information factor; when the driver driving the vehicle is identified and the corresponding information factor is collected, the driver's ID card number and the license plate number of the driving vehicle are set as the association factors of each other in this collection behavior.

[0047] S2 weights the information factors of the identified object and obtains the key factors through the weight acquisition module 211. The weight acquisition module 201 has been described in detail above and will not be repeated here.

[0048] S3 The key factor passes through the threshold calculation module 221 to determine whether the identification object is added to the warning object output module 202. If the key factor passes through the threshold calculation module 221 and is determined to be an overflow factor, the identification object corresponding to the key factor will be added to the warning object output module 202. The specific implementation method of the threshold calculation module 221 has been described one by one above, and will not be repeated here.

[0049] S4 determines the monitoring value of the object through the monitoring value determination module 302 according to the warning object output module 202 and the information flow 301. When the warning object is added to the warning object output module 202, the monitoring value determination module 302 will calculate the monitoring value of the warning object in combination with the information flow 301. The calculation method of the monitoring value can adopt the following formula:

[0050]

[0051] Among them, V is the monitoring value, V 关n is the value of the nth key factor, Y 关n is the threshold value of the nth key factor, n≥1, and V0 is the initial monitoring value corresponding to the warning object in the information flow 301. The initial monitoring value here is obtained by integrating and inferring the information flow 301 in the actual scenario. If there is no initial monitoring value of the warning object in the information flow 301, then V0 is set to 1. It can be concluded from the formula that when the value V of the key factor 关n Deviation Y 关n The larger the value V is, the greater the monitoring value V will be. In particular, the monitoring value can also be calculated by other formula calculation methods or other calculation methods that can achieve the same calculation effect.

[0052] S5 stores the judgment clues of the warning objects with certain monitoring value through the storage module 203. The so-called judgment clues are the judgment process of the influence of all the key factors involved in the above steps S1-S4 on the monitoring value.

[0053] S6 outputs the target object through the target identification module 303. As can be seen from the above, the target identification module 303 determines the target object through the target value. The target value can be calculated using the following formula:

[0054] T=V*P

[0055] Where T is the target value, V is the monitoring value in S4, and P is the precision value. The precision value P is set according to the needs of the actual scenario. The larger P is, the fewer target values ​​T can fall within the same range, so the precision is higher. The target value calculation method can also use other formula calculation methods or other implementation methods that can achieve the same calculation effect.

[0056] Implementation method 2:

[0057] The difference from the implementation method 1 is that S5 stores the judgment clues of the warning objects with certain monitoring value through the storage module 203; at the same time, according to the association setting in the association module 102, the associated objects with the warning objects are determined, and the information factors of the associated objects are stored. In this way, the expansion of the identification objects can be effectively realized. In some special scenarios, it is particularly important. For example, when a major health and safety accident occurs in a certain area, when someone is detected as an infection source, this method can quickly identify all other objects associated with this infection source.

[0058] The series of detailed descriptions listed above are only specific descriptions of feasible implementation methods of the present invention. They are not intended to limit the scope of protection of the present invention. Any equivalent implementation methods or changes that do not deviate from the technical spirit of the present invention should be included in the scope of protection of the present invention.

[0059] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0060] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. An intelligent early warning management system, characterized in that: It includes an intelligent recognition module, a forensics module, and an appraisal module; The intelligent recognition module is used to collect information factors of different objects one by one through intelligent devices, and the information factors are entered into the system for real-time management after being collected; The forensics module is used to judge according to the information factors collected by the intelligent recognition module according to certain rules and preliminarily screen out early-warning objects, and at the same time store the clue information formed during the process of identifying and collecting evidence of the early-warning objects; The appraisal module is used to determine the monitoring value of the early-warning objects in the forensics module and output target objects according to certain principles; The intelligent recognition module includes an extraction module for collecting information factors of objects and an association module for realizing the association setting of information factors from different objects; the association module performs association setting on two or more objects collected in the same collection behavior to identify another early-warning object that can be effectively identified and managed; The forensics module includes a judgment module for screening and judging information factors, a storage module for storing and managing the information factors participating in the judgment module, and an early-warning object output module for outputting objects according to the judgment conclusion of the judgment module; The appraisal module includes an information flow for sorting and verifying input information factors, a monitoring value judgment module for judging the monitoring value of the objects output by the early-warning object output module in combination with the information flow, and a target determination module for determining the target value of the early-warning objects with a certain monitoring value output by the monitoring value judgment module; the input information factors refer to the information factors that are not identified and collected by the intelligent recognition module; The calculation method of the target value adopts the following formula: T = V * P Where T is the target value, V is the monitoring value, P is the precision value, and the precision value P is set according to the needs of the actual scenario. The larger P is, the fewer target values T that can fall within the same range. Therefore, the higher the precision; The storage module stores the judgment clues of early-warning objects with a certain monitoring value; at the same time, according to the association setting in the association module, it determines the associated objects of the early-warning objects and stores the information factors of its associated objects.

2. The system according to claim 1, characterized in that: The calculation method of the monitoring value adopts the following formula: Where V is the monitoring value, V关n is the value of the nth key factor, Y关n is the threshold of the nth key factor, n ≥ 1, and V0 is the initial monitoring value corresponding to the early-warning object in the information flow.

3. The system according to claim 1, characterized in that: The judgment module includes a weight acquisition module for assigning weight values to information factors to obtain key factors, and a threshold calculation module for calculating the threshold of the key factors obtained by the weight acquisition module and judging the overflow factors; 4. The system according to claim 3, characterized in that: The threshold calculation method of the threshold calculation module is the single-threshold setting method, the double-threshold setting method, the association-threshold setting method, the similarity-threshold setting method, and the factor accumulation-threshold setting method; The single-threshold setting method is to preset a threshold for a certain key factor. When the value of the key factor reaches the preset threshold, the key factor is judged as an overflow factor; The dual threshold setting method is to pre-set an upper and lower threshold for a certain key factor to form a threshold interval. When the value of the key factor exceeds the threshold interval, the key factor is determined to be an overflow factor. The associated threshold setting method is to pre-set a threshold for two or more key factors respectively, and set one of the key factors as the main factor. When the values ​​of the corresponding key factors reach the corresponding preset thresholds, the main factor is determined to be an overflow factor. The similarity threshold setting method is to set the floating interval of a certain key factor in different ranges as a similarity value. When the frequency of occurrence of a certain similarity value reaches a preset threshold, the corresponding key factor is determined to be an overflow factor. The factor accumulation threshold setting method is to set corresponding thresholds for more than one key factor, and set one of the key factors as the main factor. When all the key factors reach the preset threshold, the corresponding main factor is determined to be an overflow factor.

5. The system according to claim 4, characterized in that: The single threshold setting method, the double threshold setting method, the association threshold setting method, the similarity threshold setting method and the factor accumulation threshold setting method can be applied separately or in combination of two or more methods.

6. A method for implementing an intelligent early warning management system according to any one of claims 1 to 5, characterized in that The following steps are involved: S1 collects and integrates information factors of the identification object through the extraction module of the intelligent recognition module; S2 uses the weight acquisition module to assign weights to the information factors of the screening object and obtain key factors; The S3 key factor passes through the threshold calculation module to determine whether the identification object is added to the warning object output module; S4 determines the monitoring value of the object through the monitoring value determination module according to the warning object output module and the information flow; S5 stores the judgment clues of the warning objects with certain monitoring value through the storage module; S6 outputs the target object through the target identification module.

Images