A method, device, equipment and medium for explosion case scene reconstruction and suspect judgment

Abstract

The present application relates to the field of Bayesian network, and discloses a method, device, equipment and medium for explosion case scene reconstruction and suspect judgment, which can construct a Bayesian judgment network of the explosion case according to the condition elements, scene elements and consequence elements of the explosion case, generate a prior condition probability table of the Bayesian judgment network based on triangular fuzzy numbers and mean area method, take multiple elements, the correlation between different elements, part of criminal facts, preliminary assumptions, the occurred scene and the unoccurred scene obtained by surveying and analyzing the central scene as reasoning conditions, input the reasoning conditions into the Bayesian judgment network, make the Bayesian judgment network perform scene reconstruction and suspect judgment according to the reasoning conditions and the prior condition probability table, and obtain the coarse-grained criminal facts, fine-grained criminal scene, behavior and investigation suggestions generated and output by the Bayesian judgment network. The present application can effectively improve the intelligent investigation level and reduce the human resource consumption of investigators.

Description

Technical Field

[0001] This invention relates to the field of Bayesian networks, and in particular to a method, apparatus, equipment, and medium for scene reconstruction and suspect assessment in explosion cases. Background Technology

[0002] With the development of science and technology, case investigation techniques are constantly improving.

[0003] Bombing incidents refer to criminal acts or terrorist attacks that use explosives to cause destruction, injury, panic, or other illegal purposes. They are a major component of social security cases and are typical serious violent crimes that seriously affect social security and public safety.

[0004] Bombing cases are generally investigated by investigators based on experience and on-site information, which involves a low level of intelligence and a high consumption of human resources. Summary of the Invention

[0005] This invention provides a method, apparatus, equipment, and medium for scene reconstruction and suspect assessment in explosion cases, which addresses the shortcomings of low intelligence levels and high human resource consumption in explosion case investigation in related technologies, effectively improving the level of intelligent investigation and reducing the human resource consumption of investigators.

[0006] In a first aspect, the present invention provides a method for scene reconstruction and suspect assessment in explosion cases, including: Based on the conditional elements, situational elements, and consequence elements of the explosion case, a Bayesian judgment network for the explosion case is constructed. The prior conditional probability table of the Bayesian judgment network is generated based on the triangular fuzzy number and the mean area method. Multiple elements obtained through on-site investigation and analysis, the relationships between different elements, partial criminal facts, preliminary hypotheses, and scenarios that have occurred and have not occurred are used as inference conditions. These inference conditions are then input into the Bayesian judgment network, which performs scenario reconstruction and suspect assessment based on the inference conditions and the prior conditional probability table. This results in coarse-grained criminal facts, fine-grained criminal scenarios, behaviors, and investigation recommendations generated and output by the Bayesian judgment network. Upon receiving a new inference condition input by the user, the new inference condition is input into the Bayesian analysis network, enabling the Bayesian analysis network to perform scenario reconstruction and suspect assessment based on the new inference condition and the prior conditional probability table, until an end-of-assessment instruction is received, and the latest output information of the Bayesian analysis network is obtained; wherein, the new inference condition is generated by the user based on the coarse-grained crime facts, fine-grained crime scenarios, behaviors, and investigation suggestions output by the Bayesian analysis network.

[0007] Optionally, the conditional elements are the entities constituting the crime in the bombing case, including people, objects, and the environment; the scenario elements include the crime scenarios of the preparation stage, the execution stage, and the escape stage; and the consequence elements include items, traces, and information.

[0008] Optionally, constructing a Bayesian analysis network for the explosion case based on its conditional, contextual, and consequence elements includes: Based on the condition element, the scenario element, and the consequence element, respectively, create a corresponding condition assumption layer, scenario assumption layer, and consequence information layer, each of which includes at least one node; Based on the relationship between the conditional elements, the scenario elements, and the consequence elements, the parent node of the node in the consequence information layer is determined to be a node in the conditional assumption layer or the scenario assumption layer, and the parent node of the node in the scenario assumption layer is determined to be a node in the conditional assumption layer. Based on the dependencies between the specific components of the conditional elements, scenario elements, and consequence elements, and the parent-child relationships between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer, connection relationships are created between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer to construct the Bayesian judgment network.

[0009] Optionally, the prior conditional probability table includes the probability of the event occurring at the node given the probability of the event occurring at the parent node; The generation of the prior conditional probability table for the Bayesian judgment network based on the triangular fuzzy number and the mean area method includes: A probability questionnaire for the explosion case is generated based on leakage noise or a model. The probability questionnaire contains multiple questions that need to be answered using probability semantic values. The probability questionnaire was sent to multiple experts to complete, resulting in a completed questionnaire from each expert who answered using probability semantic values. Based on the established correspondence, the probability semantic value of each filling form is converted into a corresponding triangular fuzzy number; The arithmetic mean of the probability semantic values ​​in each of the filling forms is calculated to obtain the fuzzy probability average. The target probability is obtained by defuzzifying the average fuzzy probability using the mean area method. The target probability is normalized to obtain the probability of different states of each node; The prior condition probability table is generated based on the probability of different states of each node.

[0010] Secondly, the present invention provides a device for scene reconstruction and suspect assessment in explosion cases, comprising: The construction unit is used to construct a Bayesian judgment network for the explosion case based on the conditional elements, situational elements, and consequence elements of the explosion case. The generation unit is used to generate the prior conditional probability table of the Bayesian judgment network based on the triangular fuzzy number and the mean area method. As a unit, it is used to take multiple elements obtained through on-site investigation and analysis, the relationships between different elements, some criminal facts, preliminary assumptions, and scenarios that have occurred and scenarios that have not occurred as inference conditions; The analysis unit is used to input the reasoning conditions into the Bayesian analysis network, so that the Bayesian analysis network can perform scenario reproduction and suspect analysis based on the reasoning conditions and the prior condition probability table, and obtain the coarse-grained crime facts, fine-grained crime scenarios, behaviors and investigation suggestions generated and output by the Bayesian analysis network. The input unit is used to receive new inference conditions input by the user and input the new inference conditions into the Bayesian judgment network, so that the Bayesian judgment network can perform scenario reproduction and suspect judgment based on the new inference conditions and the prior conditional probability table until the judgment ends and the latest output information of the Bayesian judgment network is obtained; wherein, the new inference conditions are generated by the user based on the coarse-grained crime facts, fine-grained crime scenarios, behaviors and investigation suggestions output by the Bayesian judgment network.

[0011] Optionally, the conditional elements are the entities constituting the crime in the bombing case, including people, objects, and the environment; the scenario elements include the crime scenarios of the preparation stage, the execution stage, and the escape stage; and the consequence elements include items, traces, and information.

[0012] Optionally, the building unit is further configured to: Based on the condition element, the scenario element, and the consequence element, respectively, create a corresponding condition assumption layer, scenario assumption layer, and consequence information layer, each of which includes at least one node; Based on the relationship between the conditional elements, the scenario elements, and the consequence elements, the parent node of the node in the consequence information layer is determined to be a node in the conditional assumption layer or the scenario assumption layer, and the parent node of the node in the scenario assumption layer is determined to be a node in the conditional assumption layer. Based on the dependencies between the specific components of the conditional elements, scenario elements, and consequence elements, and the parent-child relationships between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer, connection relationships are created between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer to construct the Bayesian judgment network.

[0013] Optionally, the prior conditional probability table includes the probability of the event occurring at the node given the probability of the event occurring at the parent node; The generation unit is further configured to: A probability questionnaire for the explosion case is generated based on leakage noise or a model. The probability questionnaire contains multiple questions that need to be answered using probability semantic values. The probability questionnaire was sent to multiple experts to complete, resulting in a completed questionnaire from each expert who answered using probability semantic values. Based on the established correspondence, the probability semantic value of each filling form is converted into a corresponding triangular fuzzy number; The arithmetic mean of the probability semantic values ​​in each of the filling forms is calculated to obtain the fuzzy probability average. The target probability is obtained by defuzzifying the average fuzzy probability using the mean area method. The target probability is normalized to obtain the probability of different states of each node; The prior condition probability table is generated based on the probability of different states of each node.

[0014] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the explosion case scenario reconstruction and suspect assessment method described in the first aspect or any corresponding embodiment.

[0015] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the explosion case scenario reconstruction and suspect assessment method described in the first aspect or any corresponding embodiment thereof.

[0016] The present invention provides a method, apparatus, equipment, and medium for scene reconstruction and suspect assessment in explosion cases. This method can construct a Bayesian assessment network for explosion cases based on the conditional, situational, and consequence elements of the case. A prior conditional probability table for the Bayesian assessment network is generated based on triangular fuzzy numbers and the mean-area method. Multiple elements obtained through on-site investigation and analysis, the relationships between these elements, partial criminal facts, preliminary hypotheses, and both actual and non-occurring scenarios are used as inference conditions. These inference conditions are input into the Bayesian assessment network, enabling it to reconstruct the scene and assess suspects based on the inference conditions and the prior conditional probability table. The result is coarse-grained criminal facts, fine-grained criminal scenarios, behaviors, and investigation recommendations generated and output by the Bayesian network. Upon receiving new inference conditions input by the user, these conditions are fed into the Bayesian analytical network. The network then performs scenario reconstruction and suspect assessment based on the new inference conditions and a priori probability tables, continuing until a termination instruction is received. The network then outputs its latest information. The new inference conditions are generated by the user based on the coarse-grained crime facts, fine-grained crime scenarios, behaviors, and investigation suggestions output by the Bayesian network. This invention can construct a Bayesian analytical network based on the characteristics of bombing cases, utilizing it for scenario reconstruction and suspect assessment. This yields coarse-grained crime facts, fine-grained crime scenarios, behaviors, and investigation suggestions generated and output by the Bayesian network, effectively improving intelligent investigation capabilities and reducing the manpower consumption of investigators. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating a method for reconstructing a scenario and identifying a suspect in an explosion case, provided by an embodiment of the present invention; Figure 2 A schematic diagram illustrating the conditional elements of an explosion case provided in an embodiment of the present invention; Figure 3 A schematic diagram of scene elements for an explosion case provided in an embodiment of the present invention; Figure 4 A schematic diagram illustrating the consequences of an explosion incident, provided as an embodiment of the present invention; Figure 5 A Bayesian analysis network framework for explosion cases is provided in this embodiment of the invention. Figure 6A schematic diagram of the condition assumption layer and scenario assumption layer nodes for an explosion case provided in an embodiment of the present invention; Figure 7 A schematic diagram of the consequences information layer node of an explosion case provided in an embodiment of the present invention; Figure 8 A schematic diagram of a Bayesian judgment network provided in an embodiment of the present invention; Figure 9 A schematic diagram of a probabilistic semantic value and a corresponding triangular fuzzy number provided in an embodiment of the present invention; Figure 10 A flowchart illustrating another method for reconstructing explosion crime scenarios and identifying suspects, provided by an embodiment of the present invention; Figure 11 This invention provides a scenario reconstruction result output by a Bayesian analytical network in an embodiment of the invention. Figure 12 This is a schematic diagram of a device for reconstructing explosion crime scenarios and analyzing suspects, provided in an embodiment of the present invention. Figure 13 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0020] Among related technologies, the investigation and management of social security cases is one of the important areas of social security research. Social security cases are highly harmful, occur frequently, and often have serious consequences and wide-ranging impacts. A crucial task in the investigation phase of social security cases is to identify the perpetrators and their potential criminal acts.

[0021] Bombing incidents constitute a major component of public security cases, representing typical serious violent crimes that severely impact social and public safety. Detecting bombing incidents is highly challenging, and traditional experience is no longer sufficient to meet the demands for rapid and accurate investigation. Designing methods for bombing incident scenario reconstruction and suspect assessment has significant scientific value and practical guidance.

[0022] The following is combined with Figures 1-11 This invention describes a method for reconstructing explosion incident scenarios and identifying suspects.

[0023] like Figure 1As shown in the figure, this embodiment proposes a first method for reconstructing explosion case scenarios and identifying suspects. This method may include the following steps: S101. Based on the conditional elements, situational elements, and consequence elements of an explosion case, construct a Bayesian analysis network for the explosion case.

[0024] Specifically, this embodiment can construct the conditional elements, scenario elements, and consequence elements of an explosion case based on the characteristics of the case.

[0025] Optionally, the conditional elements are the entities that constitute the crime in the bombing case, including people, objects, and the environment; the situational elements include the crime scene during the preparation, execution, and escape phases; and the consequence elements include items, traces, and information.

[0026] Specifically, such as Figure 2 The bombing case illustrated uses the following elements as conditions. These elements are the entities that constitute the act of committing the crime, and the act itself can be seen as the interaction between these elements. The elements are categorized into three types based on their attributes: people, objects, and environment. People are further divided into perpetrators (offenders), victims, and other individuals, based on their roles in the case. The perpetrator is the subject who commits the crime, the victim is the person who suffers the consequences of the crime, and other individuals include witnesses and other individuals related to the case. Similarly, objects are further divided into tools used in the crime, tools used for transfer, and other objects. Tools used in the crime refer to the tools used by the perpetrator when committing the violent crime; tools used for transfer refer to the tools used by the perpetrator to enter or leave the scene; other objects refer to all other objects related to the crime but not included in the above categories. The environment includes the central scene and related scenes. The central scene (also called the main scene) refers to the place where the perpetrator commits the violent crime, and related scenes refer to other places associated with the crime.

[0027] like Figure 3The illustrated example illustrates the elements of a bombing crime scenario. This embodiment can analyze these elements. In the preparation phase, eight scenarios are analyzed: reconnaissance, preparing transfer tools (self-supplied), preparing transfer tools (others), preparing explosive device materials (self-supplied), preparing explosive device materials (others), manufacturing the explosive device, forming a gang, and skills training. In the execution phase, four scenarios are identified: entering the crime scene, placing the explosive device, detonating the explosive (on-site detonation), and detonating the explosive (off-site detonation). Entering the crime scene refers to the perpetrator entering the location of the crime through various means; placing the explosive device refers to the perpetrator installing tools for detonation at the crime scene; detonating the explosive (on-site detonation) refers to the perpetrator detonating the explosive at the scene; detonating the explosive (off-site detonation) refers to the perpetrator not being at the scene but detonating the explosive remotely. In the escape phase, the perpetrator may engage in two actions: providing medical assistance and destroying evidence. Providing medical assistance refers to the perpetrator rescuing themselves or others; destroying evidence refers to the perpetrator cleaning up the scene and discarding the tools used in the crime.

[0028] like Figure 4 The consequences of the bombing case are described below. These consequences are information generated by the act of committing the crime. They are categorized into three types: objects, traces, and information. Objects are elements that primarily connect a person or object through their material composition; these can be further divided into body parts, bloodstains, biological materials, trace substances, and other items. Trace elements are elements that primarily connect a person or object through their external structural features; these can be further divided into fingerprints, footprints, burn marks, and other traces. Unlike objects and traces, information elements primarily aid in the case's deduction through the information they contain. These include witness testimonies, audiovisual materials, and electronic data, mainly corresponding to evidence other than physical evidence.

[0029] Optionally, step S101 includes: Based on the condition elements, scenario elements, and consequence elements, create corresponding condition assumption layers, scenario assumption layers, and consequence information layers, each containing at least one node. Based on the relationships between conditional elements, scenario elements, and consequence elements, determine that the parent node of a node in the consequence information layer is a node in the conditional assumption layer or the scenario assumption layer, and determine that the parent node of a node in the scenario assumption layer is a node in the conditional assumption layer. Based on the dependencies between specific components in the conditional elements, scenario elements, and consequence elements, and the parent-child relationships between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer, connection relationships are created between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer to construct a Bayesian judgment network.

[0030] Specifically, this embodiment can construct a Bayesian network node for a reasoning model of an explosion case.

[0031] like Figure 5 As shown, in this embodiment, when conducting reasoning in bombing cases, a Bayesian network node for the bombing case reasoning model can be established based on the actual work processes of case investigation and in accordance with the principles of scientific rigor, systematicity, and operability. Based on expert consultation and suggestions, a Bayesian network reasoning model is constructed to further conduct investigations and interrogations. Based on the principle of operability, within the framework of all elements of a bombing case, a Bayesian network model is applied to construct the framework of the bombing case investigation reasoning model.

[0032] Specifically, the nodes of the conditional assumption layer and scenario assumption layer of the explosion case are as follows: Figure 6 As shown, the consequence information layer nodes are as follows: Figure 7 As shown.

[0033] Specifically, such as Figure 8 As shown, this implementation can determine the dependencies, i.e., the connections, between nodes in a Bayesian network. The connections between nodes in a Bayesian network are determined by the results of causal relationship and element association analysis. The parent nodes of the consequence information layer nodes are the condition hypothesis layer nodes and the scenario hypothesis layer nodes, because the corresponding conditional and scenario elements cause the occurrence of consequence elements. The parent nodes of the scenario hypothesis layer nodes are the hypothesis nodes involving the perpetrator, because the perpetrator has implemented various scenarios. Based on the nodes and their connections, the network structure of the Bayesian network can be determined.

[0034] S102. Generate a prior conditional probability table for Bayesian judgment networks based on triangular fuzzy numbers and mean area method.

[0035] Optionally, the prior conditional probability table includes the probability of a node's event occurring given the probability of its parent node's event occurring. Step S102 includes: A probability questionnaire for explosion cases is generated based on leakage noise or a model. The probability questionnaire contains multiple questions that need to be answered using probability semantic values. The probability questionnaire was sent to multiple experts to complete, and each expert's completed questionnaire was obtained by using probability semantic values. Based on the established correspondence, the probability semantic value of each form is converted into a corresponding triangular fuzzy number; The arithmetic mean of the probability semantic values ​​for each completed form is calculated to obtain the fuzzy probability average. The target probability is obtained by defuzzifying the average fuzzy probability using the mean area method; The target probability is normalized to obtain the probability of different states of each node; A prior conditional probability table is generated based on the probability of different states of each node.

[0036] Specifically, this embodiment needs to determine the network parameters in the logical network structure, namely the prior probability and conditional probability of the Bayesian network. Since real-world investigative case data is difficult to obtain, expert experience is relied upon to acquire the probability parameters. To facilitate the acquisition of expert experience, semantic variables can be introduced to characterize the magnitude of the probability, and the probability can be linked to the triangular fuzzy number, such as... Figure 9 As shown, this is to take into account the experience of different experts.

[0037] Specifically, the probability semantic value given by the kth expert can be converted into a triangular fuzzy number according to the table above, which can be denoted as ; if there are a total q If there are multiple experts, then the arithmetic mean of their evaluations needs to be used to synthesize the results. Therefore, the average of the fuzzy probabilities can be expressed as: .

[0038] in, It is the left endpoint (minimum possible value) of the triangular fuzzy number. It is the peak value (most likely value) of the triangular fuzzy number. It is the right endpoint (maximum possible value) of the triangular fuzzy number.

[0039] Next, in this embodiment, the mean area method can be used to defuzzify the fuzzy probabilities, transforming them into precise probabilities. Finally, to satisfy the requirement that the sum of the probabilities of different states of a node is 1, the probabilities need to be normalized, thus obtaining the probability of each node's state.

[0040] It should be noted that this embodiment can use an expert questionnaire to obtain probability parameters. Generally, for a node with n parent nodes, when each node has two states (yes and no), then for this node, 2... n This presents a problem. Therefore, designing tens of thousands of investigation questions is required for the logical network structure of explosion-related cases, which is both inefficient and significantly increases the burden on experts. To address this, leakage noise or a model can be used to simplify the number of questions. Thus, for a node with n parent nodes, only n+1 questions need to be designed. Using leakage noise or a model, a questionnaire was designed for station reporting cases, and experts were surveyed. A total of 312 survey results were collected, and the probability parameters of the logical network structure were obtained.

[0041] Specifically, this embodiment can further calculate the conditional probabilities of child nodes in a Bayesian network. For any variable in a Bayesian Network (BN), and belonging to the probability space { The set of} The elements correspond one-to-one. The probability space of a Bayesian network consists of three parts. The first part is the space... , It is the structure or diagram of a Bayesian network, in which These are nodes (or variables) in a Bayesian network. It is the set of directed edges between variables in a network. The variables and directed edges together form a directed acyclic graph (DAG). The second part is the space. ,yes The space of all possible states The third part is space. It is all related to and Related The probability distribution. Definition It is the set of all possible Bayesian networks (BN). .in, It is an adjoint parameter A directed acyclic graph, .parameter , is a node A conditional probability table (CPT) is a table that lists the nodes in tabular form. Each state is related to its parent node The conditional probability of a state, i.e., P( When a node has no parent node, its conditional probability table is directly derived from the prior probability distribution P(…). Given. Any node The conditional probability table can be initialized based on the user's prior knowledge or learned through a training set. For the node space... Sampling is One observation at each node in the database. For node space of d Compilation of sub-sampling. When all values ​​of all variables are known, it is considered that... There are no missing values. This assumes that each sample is taken individually. They are independent of each other and have the same unknown distribution.

[0042] It should be noted that Bayesian networks are mathematical models based on the application of Bayesian rules to analyze the obtained data. When variables are given... The value of the child node Bayes' theorem can be used to calculate variables. The posterior probability distribution is calculated using formula (1): ; in, Prior probability yes state ( The known probability distribution of (). Likelihood function , including with The instantiation conditional probability of contiguous sub-variables. When all instantiated variables are known... Therefore, there is The marginalization of observed variables, This explains the relationship between instantiated variables and... The relationship between all possible states is given by formula (2): ; in, It is a variable All p The instantiation value of the j-th variable of each child node. The posterior probability is given by Representation, also known as The marginal probability is expressed as its confidence level, representing the probability of its occurrence given the evidence. Finally, can be It is deduced from the above formula (1).

[0043] S103. The multiple elements obtained through the investigation and analysis of the central scene, the relationships between different elements, some criminal facts, preliminary assumptions, and scenarios that have occurred and have not occurred are used as inference conditions.

[0044] Specifically, in this embodiment, relevant information obtained by investigators from conducting on-site investigations and analyses can be used as inference conditions to be input into the Bayesian analytical network for processing.

[0045] S104. Input the reasoning conditions into the Bayesian judgment network so that the Bayesian judgment network can reproduce the scenario and judge the suspect based on the reasoning conditions and the prior condition probability table, and obtain the coarse-grained crime facts, fine-grained crime scenarios, behaviors and investigation suggestions generated and output by the Bayesian judgment network.

[0046] Specifically, in this embodiment, the determined inference conditions can be input into the Bayesian judgment network. At this time, the Bayesian judgment network can perform scenario reproduction and suspect assessment based on the received inference conditions and prior condition probability table, generate coarse-grained crime facts, fine-grained crime scenarios, behaviors and investigation suggestions, and output them to the user.

[0047] S105. Upon receiving a new inference condition input by the user, the new inference condition is input into the Bayesian judgment network, so that the Bayesian judgment network can perform scenario reproduction and suspect judgment based on the new inference condition and the prior condition probability table, until the judgment ends and the latest output information of the Bayesian judgment network is obtained; wherein, the new inference condition is generated by the user based on the coarse-grained crime facts, fine-grained crime scenarios, behaviors and investigation suggestions output by the Bayesian judgment network.

[0048] Specifically, users can determine whether new inference conditions need to be constructed based on the information output by the Bayesian judgment network. When it is determined that new inference conditions need to be constructed, users can input the constructed new inference conditions into the Bayesian judgment network. The Bayesian judgment network will then perform another scenario reproduction and suspect assessment based on the new inference conditions and the prior condition probability table, obtaining new information output by the Bayesian judgment network. The user will then determine whether new inference conditions need to be constructed again based on the information output by the Bayesian judgment network, until the user selects the input end command for the assessment.

[0049] The explosion case scenario reconstruction and suspect assessment method proposed in this embodiment can construct a Bayesian assessment network for explosion cases based on the conditional elements, scenario elements, and consequence elements of the explosion case. A prior conditional probability table for the Bayesian assessment network is generated based on triangular fuzzy numbers and the mean-area method. Multiple elements obtained through on-site investigation and analysis, the relationships between different elements, partial criminal facts, preliminary hypotheses, and both actual and non-occurring scenarios are used as inference conditions. These inference conditions are input into the Bayesian assessment network, enabling it to reconstruct the scenario and assess the suspect based on the inference conditions and the prior conditional probability table. This results in coarse-grained criminal facts, fine-grained criminal scenarios, behaviors, and investigation suggestions generated and output by the Bayesian assessment network. Upon receiving new inference conditions input by the user, these conditions are fed into the Bayesian analytical network. The network then performs scenario reconstruction and suspect assessment based on the new inference conditions and a priori probability tables, continuing until a termination instruction is received. The network then outputs its latest information. The new inference conditions are generated by the user based on the coarse-grained crime facts, fine-grained crime scenarios, behaviors, and investigation suggestions output by the Bayesian network. This embodiment can construct a Bayesian analytical network based on the characteristics of bombing cases. By utilizing this network for scenario reconstruction and suspect assessment, the network generates and outputs coarse-grained crime facts, fine-grained crime scenarios, behaviors, and investigation suggestions, effectively improving intelligent investigation capabilities and reducing the manpower consumption of investigators.

[0050] like Figure 10As shown in this embodiment, in the second method for reconstructing explosion case scenarios and identifying suspects, this method can complete the construction of a Bayesian analytical network, design and develop the Bayesian analytical network, integrate the Bayesian analytical network into the analytical system, and execute the following process: (1) Input elements and the relationships between elements: Input the elements obtained from the field survey and the relationships between the elements obtained from the inspection or investigation into the system.

[0051] (2) Provide definite facts of the crime and possible hypotheses: Based on the case details and relevant information (such as video surveillance, witness testimonies, expert experience, etc.), definite facts of the crime (such as item A being the tool used in the crime) and possible hypotheses (such as person B being the perpetrator) can be preliminarily provided. In this step, the above information needs to be entered into the system to clarify the reasoning scope of the Bayesian analytical network.

[0052] (3) Provide scenarios that are certain to have occurred and scenarios that are certain not to have occurred: Based on the case details and relevant information (such as video surveillance, witness testimonies, expert experience, etc.), we can preliminarily determine which scenarios (criminal acts) occurred and which scenarios did not occur. In this step, the above information can be entered into the system so that the analysis model can provide results more quickly and accurately.

[0053] (4) Judgment and reasoning: Based on the information input in the first three steps, the Bayesian judgment network can perform judgment and reasoning on the case, which will produce three results: 1) Coarse-grained criminal facts: It gives the definite criminal facts (conditional elements) and possible criminal hypotheses, as well as the behavioral scenarios or consequences supporting them; 2) Fine-grained criminal scenarios and behaviors: The model can output the definite criminal scenarios that have occurred and the possible criminal scenarios, the temporal relationship of the scenarios and the location of the occurrence, as well as the scenarios or evidence supporting them; 3) Investigation suggestions: Based on the qualitative and quantitative results of the conditional hypothesis layer and the behavioral hypothesis layer, it gives corresponding investigation suggestions to strengthen the chain of evidence. (5) Output of reasoning results: This step outputs and displays the reasoning results and investigation suggestions generated by the analysis.

[0054] (6) Judgment of change of assumption conditions: Based on the analysis and reasoning results, the user needs to judge whether the reasoning conditions need to be changed. If they need to be changed, the process (7) needs to be switched to determine the new reasoning conditions. If they do not need to be changed, the process (8) continues.

[0055] (7) Determine new inference conditions: This step can modify the input provided to the Bayesian judgment network in steps (1)-(3) for re-judgment.

[0056] (8) Storing elements and results into the database: This step requires storing all the element information of the case and the final case result (i.e. the real result of the conditional hypothesis layer and the behavioral hypothesis layer) into the database for the purpose of analyzing the update and optimization of the Bayesian analysis network.

[0057] After the Bayesian analysis network is updated and optimized, this embodiment can input explosion case investigation information into the optimized Bayesian analysis network for analysis and judgment, and obtain valid conditional element hypotheses and scenario element hypotheses, providing support for explosion case scenario reconstruction and suspect identification.

[0058] Using a real case as input, the basic facts are as follows: A series of explosions occurred at the south gate of a unit in a certain city, with a total of three explosions occurring on the east and west sides of the south gate's roadside green belt and the opposite side of the road's south green belt. One person died and several were injured. By collecting on-site investigation information, forensic examination information, investigative information, and other information, these were input into a Bayesian analytical network. The final Bayesian analytical network's analysis of the crime's process is shown below. Figure 11 As shown.

[0059] By comparing the conclusions (criminal facts, criminal scenario, and investigative suggestions) given by the Bayesian judgment network with the actual case, it can be found that: (1) the reverse reasoning results given by the judgment model are relatively reasonable and based on evidence, which can help investigators to reconstruct the crime; (2) the judgment model can provide effective investigative suggestions in actual cases and play an important role in discovering new evidence.

[0060] The explosion case scenario reconstruction and suspect assessment method proposed in this embodiment can construct a Bayesian assessment network based on the characteristics of the explosion case, optimize and update the Bayesian assessment network using relevant data, and use the optimized Bayesian assessment network to reconstruct the scenario and assess the suspect, ensuring the accuracy of the assessment, further improving the level of intelligent investigation and reducing the consumption of human resources.

[0061] like Figure 12 As shown in the figure, this embodiment proposes a device for scene reconstruction and suspect assessment in explosion cases. The device may include: Construction unit 101 is used to construct a Bayesian judgment network for explosion cases based on the conditional elements, situational elements, and consequence elements of explosion cases. The generation unit 102 is used to generate a prior conditional probability table of the Bayesian judgment network based on the triangular fuzzy number and the mean area method. Unit 103 is used to take multiple elements obtained through the investigation and analysis of the central scene, the relationship between different elements, some criminal facts, preliminary assumptions, and scenarios that have occurred and scenarios that have not occurred as inference conditions. The analysis unit 104 is used to input the reasoning conditions into the Bayesian analysis network, so that the Bayesian analysis network can reproduce the scenario and analyze the suspect based on the reasoning conditions and the prior condition probability table, and obtain the coarse-grained crime facts, fine-grained crime scenarios, behaviors and investigation suggestions generated and output by the Bayesian analysis network. The input unit 105 is used to receive new inference conditions input by the user and input them into the Bayesian judgment network. The Bayesian judgment network then performs scenario reconstruction and suspect assessment based on the new inference conditions and the prior condition probability table until it receives an end-of-assessment instruction and obtains the latest output information of the Bayesian judgment network. The new inference conditions are generated by the user based on the coarse-grained crime facts, fine-grained crime scenarios, behaviors, and investigation suggestions output by the Bayesian judgment network.

[0062] It should be noted that the processing procedures of the construction unit 101, generation unit 102, processing unit 103, judgment unit 104, and input unit 105, and their beneficial effects, can be referred to respectively. Figure 1 Steps S101 to S105 are not described in detail here.

[0063] Optionally, the conditional elements are the entities that constitute the crime in the bombing case, including people, objects, and the environment; the situational elements include the crime scene during the preparation, execution, and escape phases; and the consequence elements include items, traces, and information.

[0064] Optionally, building unit 101 is also used for: Based on the condition elements, scenario elements, and consequence elements, create corresponding condition assumption layers, scenario assumption layers, and consequence information layers, each containing at least one node. Based on the relationships between conditional elements, scenario elements, and consequence elements, determine that the parent node of a node in the consequence information layer is a node in the conditional assumption layer or the scenario assumption layer, and determine that the parent node of a node in the scenario assumption layer is a node in the conditional assumption layer. Based on the dependencies between specific components in the conditional elements, scenario elements, and consequence elements, and the parent-child relationships between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer, connection relationships are created between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer to construct a Bayesian judgment network.

[0065] Optionally, the prior conditional probability table includes the probability of an event occurring at a node given the probability of the event occurring at its parent node; The generating unit 102 is also used for: A probability questionnaire for explosion cases is generated based on leakage noise or a model. The probability questionnaire contains multiple questions that need to be answered using probability semantic values. The probability questionnaire was sent to multiple experts to complete, and each expert's completed questionnaire was obtained by using probability semantic values. Based on the established correspondence, the probability semantic value of each form is converted into a corresponding triangular fuzzy number; The arithmetic mean of the probability semantic values ​​for each completed form is calculated to obtain the fuzzy probability average. The target probability is obtained by defuzzifying the average fuzzy probability using the mean area method; The target probability is normalized to obtain the probability of different states of each node; A prior conditional probability table is generated based on the probability of different states of each node.

[0066] The explosion case scenario reconstruction and suspect assessment device proposed in this embodiment can construct a Bayesian assessment network for explosion cases based on the conditional elements, scenario elements, and consequence elements of the explosion case. A prior conditional probability table for the Bayesian assessment network is generated based on triangular fuzzy numbers and the mean-area method. Multiple elements obtained through on-site investigation and analysis, the relationships between different elements, partial criminal facts, preliminary hypotheses, and both actual and non-actual scenarios are used as inference conditions. These inference conditions are input into the Bayesian assessment network, enabling it to reconstruct the scenario and assess the suspect based on the inference conditions and the prior conditional probability table. The result is coarse-grained criminal facts, fine-grained criminal scenarios, behaviors, and investigation suggestions generated and output by the Bayesian assessment network. Upon receiving new inference conditions input by the user, these conditions are fed into the Bayesian analytical network. The network then performs scenario reconstruction and suspect assessment based on the new inference conditions and a priori probability tables, continuing until a termination instruction is received. The network then outputs its latest information. The new inference conditions are generated by the user based on the coarse-grained crime facts, fine-grained crime scenarios, behaviors, and investigation suggestions output by the Bayesian network. This embodiment can construct a Bayesian analytical network based on the characteristics of bombing cases. By utilizing this network for scenario reconstruction and suspect assessment, the network generates and outputs coarse-grained crime facts, fine-grained crime scenarios, behaviors, and investigation suggestions, effectively improving intelligent investigation capabilities and reducing the manpower consumption of investigators.

[0067] In this embodiment, the explosion case scenario reconstruction and suspect analysis device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0068] This invention also provides a computer device having the above-described features. Figure 12 The device shown is for reconstructing the scene of an explosion and analyzing suspects.

[0069] Please see Figure 13 The present invention provides a schematic diagram of the structure of a computer device according to an optional embodiment. The computer device includes one or more processors 10, a memory 20, and interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces. The various components are interconnected via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, multiple processors and / or multiple buses can be used with multiple memories, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 13 Take a processor 10 as an example.

[0070] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0071] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0072] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function. The data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0073] Memory 20 may include volatile memory, such as random access memory. Memory may also include non-volatile memory, such as flash memory, hard disk, or solid-state drive. Memory 20 may also include combinations of the above types of memory.

[0074] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0075] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for scene reconstruction and suspect assessment in explosion cases, characterized in that, include: Based on the conditional elements, situational elements, and consequence elements of the explosion case, a Bayesian judgment network for the explosion case is constructed. The prior conditional probability table of the Bayesian judgment network is generated based on the triangular fuzzy number and the mean area method. Multiple elements obtained through on-site investigation and analysis, the relationships between different elements, partial criminal facts, preliminary hypotheses, and scenarios that have occurred and have not occurred are used as inference conditions. These inference conditions are then input into the Bayesian judgment network, which performs scenario reconstruction and suspect assessment based on the inference conditions and the prior conditional probability table. This results in coarse-grained criminal facts, fine-grained criminal scenarios, behaviors, and investigation recommendations generated and output by the Bayesian judgment network. Upon receiving a new inference condition input by the user, the new inference condition is input into the Bayesian analysis network, enabling the Bayesian analysis network to perform scenario reconstruction and suspect assessment based on the new inference condition and the prior conditional probability table, until an end-of-assessment instruction is received, and the latest output information of the Bayesian analysis network is obtained; wherein, the new inference condition is generated by the user based on the coarse-grained crime facts, fine-grained crime scenarios, behaviors, and investigation suggestions output by the Bayesian analysis network.

2. The method according to claim 1, characterized in that, The conditional elements are the entities constituting the crime in the bombing case, including people, objects, and the environment; the scenario elements include the crime scenarios of the preparation stage, the execution stage, and the escape stage; the consequence elements include items, traces, and information.

3. The method according to claim 1, characterized in that, The process of constructing a Bayesian analytical network for an explosion case based on its conditional, situational, and consequence elements includes: Based on the condition element, the scenario element, and the consequence element, respectively, create a corresponding condition assumption layer, scenario assumption layer, and consequence information layer, each of which includes at least one node; Based on the relationship between the conditional elements, the scenario elements, and the consequence elements, the parent node of the node in the consequence information layer is determined to be a node in the conditional assumption layer or the scenario assumption layer, and the parent node of the node in the scenario assumption layer is determined to be a node in the conditional assumption layer. Based on the dependencies between the specific components of the conditional elements, scenario elements, and consequence elements, and the parent-child relationships between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer, connection relationships are created between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer to construct the Bayesian judgment network.

4. The method according to claim 1, characterized in that, The prior conditional probability table includes the probability of an event occurring for the node given the probability of an event occurring for the parent node; The generation of the prior conditional probability table for the Bayesian judgment network based on the triangular fuzzy number and the mean area method includes: A probability questionnaire for the explosion case is generated based on leakage noise or a model. The probability questionnaire contains multiple questions that need to be answered using probability semantic values. The probability questionnaire was sent to multiple experts to complete, resulting in a completed questionnaire from each expert who answered using probability semantic values. Based on the established correspondence, the probability semantic value of each filling form is converted into a corresponding triangular fuzzy number; The arithmetic mean of the probability semantic values ​​in each of the filling forms is calculated to obtain the fuzzy probability average. The target probability is obtained by defuzzifying the average fuzzy probability using the mean area method. The target probability is normalized to obtain the probability of different states of each node; The prior condition probability table is generated based on the probability of different states of each node.

5. A device for reconstructing explosion crime scenarios and identifying suspects, characterized in that, include: The construction unit is used to construct a Bayesian judgment network for the explosion case based on the conditional elements, situational elements, and consequence elements of the explosion case. The generation unit is used to generate the prior conditional probability table of the Bayesian judgment network based on the triangular fuzzy number and the mean area method. As a unit, it is used to take multiple elements obtained through on-site investigation and analysis, the relationships between different elements, some criminal facts, preliminary assumptions, and scenarios that have occurred and scenarios that have not occurred as inference conditions; The analysis unit is used to input the reasoning conditions into the Bayesian analysis network, so that the Bayesian analysis network can perform scenario reproduction and suspect analysis based on the reasoning conditions and the prior condition probability table, and obtain the coarse-grained crime facts, fine-grained crime scenarios, behaviors and investigation suggestions generated and output by the Bayesian analysis network. The input unit is used to receive new inference conditions input by the user and input the new inference conditions into the Bayesian judgment network, so that the Bayesian judgment network can perform scenario reproduction and suspect judgment based on the new inference conditions and the prior conditional probability table until the judgment ends and the latest output information of the Bayesian judgment network is obtained; wherein, the new inference conditions are generated by the user based on the coarse-grained crime facts, fine-grained crime scenarios, behaviors and investigation suggestions output by the Bayesian judgment network.

6. The apparatus according to claim 5, characterized in that, The conditional elements are the entities constituting the crime in the bombing case, including people, objects, and the environment; the scenario elements include the crime scenarios of the preparation stage, the execution stage, and the escape stage; the consequence elements include items, traces, and information.

7. The apparatus according to claim 5, characterized in that, The building unit is also used for: Based on the condition element, the scenario element, and the consequence element, respectively, create a corresponding condition assumption layer, scenario assumption layer, and consequence information layer, each of which includes at least one node; Based on the relationship between the conditional elements, the scenario elements, and the consequence elements, the parent node of the node in the consequence information layer is determined to be a node in the conditional assumption layer or the scenario assumption layer, and the parent node of the node in the scenario assumption layer is determined to be a node in the conditional assumption layer. Based on the dependencies between the specific components of the conditional elements, scenario elements, and consequence elements, and the parent-child relationships between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer, connection relationships are created between different nodes in the conditional hypothesis layer, scenario hypothesis layer, and consequence information layer to construct the Bayesian judgment network.

8. The apparatus according to claim 5, characterized in that, The prior conditional probability table includes the probability of an event occurring for the node given the probability of an event occurring for the parent node; The generation unit is further configured to: A probability questionnaire for the explosion case is generated based on leakage noise or a model. The probability questionnaire contains multiple questions that need to be answered using probability semantic values. The probability questionnaire was sent to multiple experts to complete, resulting in a completed questionnaire from each expert who answered using probability semantic values. Based on the established correspondence, the probability semantic value of each filling form is converted into a corresponding triangular fuzzy number; The arithmetic mean of the probability semantic values ​​in each of the filling forms is calculated to obtain the fuzzy probability average. The target probability is obtained by defuzzifying the average fuzzy probability using the mean area method. The target probability is normalized to obtain the probability of different states of each node; The prior condition probability table is generated based on the probability of different states of each node.

9. A computer device, characterized in that, include: The device includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform the explosion case scenario reconstruction and suspect assessment method as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the explosion case scenario reconstruction and suspect assessment method as described in any one of claims 1 to 4.

Images