A security system decision-making method, device, computer equipment and storage medium based on multi-dimensional information fusion

Abstract

The present invention relates to a security system decision-making method, device, computer equipment and storage medium for multi-dimensional information fusion. The method comprises: obtaining image information, sound information and text information of a target area; constructing a D-type information storage medium of the image information according to the obtained image information, sound information and text information; p Value matrix, sound information D v Value matrix, text information D t Value matrix and final matrix D of fused data f , and the final matrix D of the fused data f Compare with the preset expert system matrix to determine the final matrix D of the fused data f Whether it can trigger the corresponding decision data; among them, D f =D p +D v +D t The present invention constructs the final matrix D of fusion data based on the acquisition of multi-dimensional data (including image information, sound information and text information). f , by transforming the fused data into the final matrix D f Comparing with the preset expert system matrix to obtain decision data can improve the accuracy of decision-making.

Description

Technical Field

[0001] The invention relates to a security system decision-making method, device, computer equipment and storage medium for multi-dimensional information fusion. Background Art

[0002] With the increasing improvement of people's living standards and the needs of urban development, intelligent security systems play an important role in urban security and social peace governance. With the development of information means today, the development of smart cities has entered a stage where how to reasonably apply the data obtained to give accurate management strategies, how to tap the potential content in the information, and be more efficient in management, and more modern and intelligent in management means. At the same time, how to use new technologies in modern applications in the field of smart city security, such as 5G technology, gait recognition and other technologies, enables staff to make corresponding security decisions based on the target's image, sound and other information through a simple information database.

[0003] However, in actual applications, the common practice is to use simple personal experience or even superficial information as the main representation of security decisions, which leads to the following technical difficulties:

[0004] First, image information, sound information and other information dimensions are different, and the amount of information they contain is also different. Even in the tracking of targets, the research methods and existing technologies have proposed different research directions and implementation methods. The traditional information analysis methods alone cannot reflect the relationship weight between multi-dimensional information and decision-making information, resulting in high complexity in information management, large amount of redundant information, inability to streamline, and inability to characterize. Second, there is a large amount of redundant information, low-quality information, unclear associations, and unclear standards for information fusion in multi-source information for the same target, which leads to isolation and insufficient matching of information processing, non-standard calculations, and the inability of computers to exert their maximum data analysis capabilities, resulting in a low level of intelligence in security decision-making. Summary of the invention

[0005] The purpose of the present invention is to propose a security system decision-making method for multi-dimensional information fusion, the method comprising: obtaining image information, sound information and text information of a target area; constructing a D-based image information according to the obtained image information, sound information and text information; p Value matrix, sound information D v Value matrix, text information D t Value matrix and final matrix D of fused data f , and the final matrix D of the fused data f Compare with the preset expert system matrix to determine the final matrix D of the fused data f Whether it can trigger the corresponding decision data; among them, D f =Dp +D v +D t .

[0006] The present invention constructs the final matrix D of fusion data based on the acquisition of multidimensional data (including image information, sound information and text information). f , by transforming the fused data into the final matrix D f Comparing with the preset expert system matrix to obtain decision data can improve the accuracy of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 It is a security system diagram with multi-dimensional information fusion;

[0008] Figure 2 is a block diagram of the internal structure of a computer device in one embodiment;

[0009] Figure 3 This is the overall system structure diagram of the security system with multi-dimensional information fusion. DETAILED DESCRIPTION

[0010] The present application scheme is further described below in conjunction with the accompanying drawings.

[0011] See also Figures 1 to 3 , a security system decision-making method based on multi-dimensional information fusion, comprising:

[0012] Obtain image information, sound information and text information of the target area;

[0013] Construct the D of image information based on the acquired image information, sound information and text information p Value matrix, sound information D v Value matrix, text information D t Value matrix and final matrix D of fused data f , and the final matrix D of the fused data f Compare with the preset expert system matrix to determine the final matrix D of the fused data f Whether it can trigger the corresponding decision data; among them, D f =D p +D v +D t .

[0014] This embodiment can obtain information of different dimensions from multi-dimensional security system terminals (including cameras, sensors capable of acquiring sound signals, text information entry terminals equipped for network security personnel, etc.), and the information of different dimensions includes information of three dimensions, namely image information, sound information and text information; for the information of the above three dimensions, the extraction of key characterization data is analyzed according to the algorithm. A security multi-dimensional database is established, and based on the security multi-dimensional database, multi-dimensional information units of entities, attributes and associations of security multi-dimensional information are generated. The multi-dimensional information units can be combined into a multi-dimensional information matrix, which can be used for subsequent research after being standardized.

[0015] Figure 1 The diagram of the multi-dimensional information fusion security system described in this embodiment is shown, including:

[0016] Parts ①, ②, and ③ within the dotted box I are sources of multidimensional information, including image information acquisition unit ①, sound information acquisition unit ②, and text information acquisition unit ③. The marks in the figure are just conceptual statements, mainly to indicate the multidimensionality of the information source.

[0017] The image information acquisition unit ① acquires image information which may be dynamic image information (such as video) or static image information (such as picture), mainly through a camera or other image acquisition device in the system that can acquire image information of each band;

[0018] The sound information acquisition unit ② represents a device for acquiring sound information (including environmental sound, special sound of the target, etc.), such as a sensor of sound signals;

[0019] Text information acquisition unit ③ includes text information input terminals (especially handheld terminals) equipped for network security personnel, which mainly refers to the information obtained by network security personnel, and standardized text information that can be obtained by network security personnel through standard selective software, that is, network security personnel's text reports, records, registrations and other information. According to the particularity of security monitoring, network security personnel are added as the main subject of collecting text information. In order to ensure the standardization and decision-making degree of data, the text information of handheld terminals has selected the established options as the main method.

[0020] The data collected by the above-mentioned information acquisition unit provides the most comprehensive original data for subsequent decision-making.

[0021] The dotted box II shows the image information acquisition unit ①, the sound information acquisition unit ② and the text information acquisition unit ③ directly connected to the dotted box I. In practice, the dotted box II is configured as a storage device for different information and a transmission link between information. The dotted box II may include various connection types, such as wired, wireless communication links or optical fiber cables, etc. The information acquired by the image information acquisition unit ①, the sound information acquisition unit ② and the text information acquisition unit ③ can be transmitted through the dotted box II and the server of the dotted box III. Due to the multi-source and structural differences of the data, the dotted box II can be configured to remove noise (i.e., denoising) for the initial screening of the data and the primary information acquisition. The dotted box II can be configured as a pre-data primary processor for denoising the image information, sound information and text information to purify the data. In the case of non-confidentiality, using the cloud system as a part of the component processing function of the dotted box II will achieve better results.

[0022] The main function of server ④ is to receive the image information, sound information and text information after data preprocessing in the dotted box II for fusion processing. The fusion method is to adopt the matrix method. By assigning different key characterization factors of information, different weights are considered when building the model. Finally, the multi-dimensional data is fused and the fusion evaluation index is given. If the evaluation index meets the preset threshold, the fusion is effective. The system can give corresponding decision data based on the fused data combined with the expert system.

[0023] In addition, in order to meet the real-time nature of decision-making, another reporting mode is provided. The corresponding data reporting level is given by weight, and according to the weight of the acquired image information, sound information and text information, it can be directly uploaded to the server④, and the server④ directly identifies the key characterization data and compares it with the preset expert database, and can quickly respond to the image information, sound information and text information.

[0024] Server ④ is a general representation of hardware and software. The environment for information storage is the hardware system, and the specific fusion algorithm construction depends on the fusion algorithm in server ④. The two can be carried out simultaneously or separately.

[0025] The implementation process of this embodiment is described in detail below, including the following steps:

[0026] Step 1, from Figure 1The image information acquisition unit ①, the sound information acquisition unit ② and the text information acquisition unit ③ shown in the figure respectively acquire the corresponding image information, sound information and text information. During the acquisition process, the image information guarantees the resolution before transmission. The hardware can also provide specific image information and sound information, such as using different cameras, sound sensors with different filters, and using standardized text information to acquire the corresponding information. The acquisition means can be a camera, a mobile phone, text records, a laptop computer record and other methods. Among them, the text information collected by the network security personnel can be packaged and reported in the image information and sound information, and transmitted together as a data packet group.

[0027] Step 2: After the data is acquired, data level preprocessing is performed based on software processing of image information, sound information, and text information. This mainly occurs in Figure 1 In the dotted box II, the image information, sound information and text information are processed preliminarily in the dotted box II to remove the corresponding noise information.

[0028] Step 3, construct entity information, attribute list, relationship attribute through the server ④ in the dotted box III, and construct preliminary information units. In this section, the information unit is presented in the form of vectors. And the D of the image information is constructed according to the following method p Value matrix, sound information D v Value matrix or text information D t Value matrix:

[0029]

[0030] Where ω1+ω2+ω3=1;

[0031] n represents obtaining n sets of image information, sound information or text information;

[0032] ω i It represents the weights of different components;

[0033] w ij It represents the weight of the difference between the abstract factors of image information, sound information or text information and the factors of the preset standard expert database;

[0034] α is Represents the value in the preset expert database;

[0035] α ij represents the jth attribute value of the i-th group of image information, sound information or text information; τ indicates that the target type is individually assigned a weight value, indicating that the type of target has a direct relationship with its possible consequences

[0036] δ indicates whether the target's abstract attribute also appears in other source information, indicating the cross-factor of the knowledge graph. The value of δ can be preset. For example, when the target's abstract attribute also appears in other source information, the value of δ is X; and when the target's abstract attribute does not appear in other source information, the value of δ is Y.

[0037] The D value represents the impact of the acquisition of n times of image information, sound information or text information on the total data set. The α of this algorithm ij The specific distribution can be reset within the system as needed and is not limited to this type of evaluation method. The basis for modification is the key determining factor in the actual application of the security monitoring system.

[0038] For example, obtaining vehicle movement image information in a key area, finding an abnormally moving vehicle, the vehicle driving to the area, repeatedly circling around, or frequent entry and exit of people, etc. After obtaining the image information, in step 3, a fixed sampling time within a period of time is used, such as the image information of the first sampling is abstracted as {τ, α 11 , α 12 , α 13 , α 14 ,δ} are the attribute lists of {target, behavior, representation, personnel situation, time, related source}, so this example is {vehicle, form, appearance, unknown, time, have}. The information describing the attributes is obtained by the algorithm in the system through the corresponding value. Because different values ​​have different effects on the final decision, that is, the impact on the result is different, so different weights are assigned. At the same time, combined with the corresponding thresholds in the expert database and different weight assignments, it is possible to calculate the image information of n groups of videos within a period of time. According to the above calculation company, D p Value matrix.

[0039] D with image information formed in step 3 p Value matrix, sound information D v Value matrix, text information D t The three types of matrices are sparsely constructed to construct the data matrix of the same space, and then fused to construct the final matrix D f , where D f =D p +D v +D t .

[0040] The final matrix is ​​used as the input matrix to compare with the preset expert system matrix in the same space to analyze the difference between the two. The influence of the expert matrix is ​​the corresponding decision matrix. The threshold is set to determine whether the final matrix can trigger the corresponding decision data.

[0041] In addition, step 3 may be provided with an emergency response mechanism, mainly the server ④ directly receives the data obtained by the image information acquisition unit ①, the sound information acquisition unit ② and the text information acquisition unit ③ respectively, and then makes a decision. This situation mainly deals with the processing of emergency time. The corresponding emergency characterization data can be set in the information collection process of the image information acquisition unit ①, the sound information acquisition unit ② and the text information acquisition unit ③. When the information center contains emergency characterization data, the data preprocessing of the dotted box II is skipped and directly enters the server ④ for processing. The server ④ can be configured with a local large-scale computer system, so the response speed is faster. Because it is an emergency, the information does not need to be fused, and the decision-making method based on the mapping of emergency characterization data can be directly triggered. For example, when the image information contains the image of "fire"; the sound information contains "large-scale human voice"; the text information contains key words such as "murder" and "riot", the corresponding decision information is directly retrieved without information fusion.

[0042] Step 4 is the display level of the whole system information, such as Figure 1 As shown, the dotted box IV shows the display device ⑤, which can be displayed in a centralized manner. In one display scene, there appear the replaced image information, sound waveform information, the comparison between the fused information index and the standard index, the decision selection and the distribution of decision information, etc. This facilitates management and avoids the influx of a large amount of information, which would cause managers to lose the decision-making and management of information.

[0043] Figure 2 The internal structure of the computer equipment in the system is shown. The computer equipment is distributed in the dotted box II and the dotted box III. If the dotted box II adopts the cloud system, then Figure 2 The main configuration shown is in the dotted frame III. It includes a processor, a memory, a network interface, an input device and a display screen connected through a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory.

[0044] The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement the relevant algorithm of the present invention. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can perform data analysis and decision mapping. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0045] The algorithm provided by the present invention is mainly implemented in the form of a computer program. The memory of the computer device can store the corresponding matrix analysis algorithm.

[0046] In one embodiment, a computer device is provided, the computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the following steps are implemented:

[0047] Obtaining input of image information, sound information and text information;

[0048] Calculate the above information through a processor;

[0049] Finally, the decision mapping result is obtained and the computer outputs the decision data.

[0050] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0051] Those skilled in the art will appreciate that all or part of the process implemented can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the process of the embodiments of the above methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0052] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0053] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0054] Figure 3 The system software structure diagram of the security system with multi-dimensional information fusion is shown in FIG. It can be seen that multi-dimensional information is not only the scope of the present invention, but also involves a lot of multi-dimensional information. The processing of data sources and preliminary decision making are described in step 2. Figure 1 Complete the corresponding work in the dotted box I and dotted box II. In terms of the summary of data processing capabilities, it is mainly Figure 1 The dashed box III part is completed. Figure 1 There are also corresponding functional applications in the dashed box III, including Figure 3 In "Application" and "Emergency Command" and "Event Notification / Warning".

[0055] The present invention has the following beneficial effects: by deeply mining multi-dimensional perception data to build a decision support information database, it can provide a scientific and unified decision-making method for smart city security, break down the barriers in the fields of security, public security, etc., and through the application of the system, it is found that in the process of applying the system, the decision-making speed is significantly accelerated and the decision-making accuracy is significantly improved.

Claims

1. A security system decision-making method based on multi-dimensional information fusion, characterized in that: The method includes: Obtain image information, sound information and text information of the target area; Construct the D of image information based on the acquired image information, sound information and text information p Value matrix, sound information D v Value matrix, text information D t Value matrix and final matrix D of fused data f , and the final matrix D of the fused data f Compare with the preset expert system matrix to determine the final matrix D of the fused data f Whether it can trigger the corresponding decision data; among them, D f =D p +D v +D t ; Image information D p Value matrix, sound information D v Value matrix or text information D t The value matrix can be calculated as follows: Where ω1+ω2+ω3=1; n represents obtaining n sets of image information, sound information or text information; ω i It represents the weights of different components; w ij It represents the weight of the difference between the abstract factors of image information, sound information or text information and the factors of the preset standard expert database; α is Represents the value in the preset expert database; α ij represents the j-th attribute value of the i-th group of image information, sound information or text information; τ indicates that the target categories are individually attached with weight values; δ indicates whether the abstract attribute of the target is also present in other source information.

2. A security system decision-making device with multi-dimensional information fusion, characterized in that: include: An information acquisition module is used to acquire image information, sound information and text information of a target area; The processor module is used to construct a D image information according to the acquired image information, sound information and text information. p Value matrix, sound information D v Value matrix, text information D t Value matrix and final matrix D of fused data f , and the final matrix D of the fused data f Compare with the preset expert system matrix to determine the final matrix D of the fused data f Whether it can trigger the corresponding decision data; among them, D f =D p +D v +D t ; Image information D p Value matrix, sound information D v Value matrix or text information D t The value matrix can be calculated as follows: Where ω1+ω2+ω3=1; n represents obtaining n sets of image information, sound information or text information; ω i It represents the weights of different components; w ij It represents the weight of the difference between the abstract factors of image information, sound information or text information and the factors of the preset standard expert database; α is Represents the value in the preset expert database; α ij represents the jth attribute value of the i-th group of image information, sound information or text information; τ indicates that the target categories are individually attached with weight values; δ indicates whether the abstract attribute of the target is also present in other source information.

3. The multi-dimensional information fusion security system decision-making device according to claim 2, characterized in that: The processor module includes a front-end data primary processor and a back-end server. The front-end data primary processor is configured to receive image information, sound information and text information obtained from the information acquisition module, and perform denoising on the image information, sound information and text information before sending them to the back-end server.

4. The multi-dimensional information fusion security system decision-making device according to claim 2, characterized in that: The information acquisition module includes an image information acquisition unit for acquiring image information, a sound information acquisition unit for acquiring sound information, and a text information acquisition unit for acquiring text information, wherein: The image information acquisition unit includes a camera; The sound information acquisition unit includes a sensor capable of acquiring a sound signal; The text information acquisition unit includes a text information input terminal equipped for network security personnel.

5. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the security system decision-making method of multi-dimensional information fusion as claimed in claim 1.

6. A computer-readable storage medium, characterized in that: Computer program instructions are stored thereon, and when the instructions are executed by the processor, the security system decision-making method of multi-dimensional information fusion described in claim 1 is implemented.

Images