Battlefield situation awareness management method and system based on panoramic vision

Abstract

The invention relates to the field of situation awareness, in particular to a battlefield situation awareness management method and system based on panoramic vision. Comprising the following steps: S1, acquiring panoramic vision data, preprocessing the panoramic vision data, and then classifying view field ranges; s2, according to a view field range classification result, constructing a range expansion and contraction model, and obtaining battlefield situation prediction factors by using the range expansion and contraction model; s3, classifying battlefield situation security levels according to the battlefield situation prediction factors, and determining visual display content according to a security level classification result; and S4, forming a closed-loop system by the above steps. Through panoramic vision data preprocessing and view field range classification, the data quality is ensured; a range expansion model is constructed, and the situation awareness precision is improved; dynamically adjusting a buffer area, and optimizing an advancing strategy; the closed-loop system is continuously optimized, long-term stability and reliability are ensured, and battlefield situation awareness efficiency and effect are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of situation awareness, and in particular to a battlefield situation awareness management method and system based on panoramic vision. Background Art

[0002] As the complexity and uncertainty of modern warfare increase, battlefield situation awareness has become the key to victory or defeat. Traditional battlefield situation awareness methods mainly rely on single sensors or cameras with limited viewing angles, which makes it difficult to obtain information comprehensively and in real time, resulting in limited accuracy and timeliness of decision support. Specifically, traditional methods have the following main problems: 1. Incomplete information acquisition: Traditional sensors and cameras usually only cover a limited field of view and cannot achieve 360-degree panoramic field of view coverage, which makes it easy to miss key information.

[0003] 2. Lack of real-time performance: Traditional methods have slow data processing and transmission speeds and cannot achieve real-time updates, which affects the timeliness of decision-making.

[0004] 3. Limited data processing capabilities: Traditional devices have limited computing power and find it difficult to process large amounts of complex data, resulting in inefficient information processing.

[0005] 4. Poor system flexibility: Traditional systems are complex to build and maintain, require professional technicians, and are difficult to troubleshoot, which affects rapid deployment and maintenance.

[0006] To solve these problems, panoramic vision technology came into being. Panoramic vision technology uses multi-cameras and fisheye lenses to achieve 360-degree panoramic field coverage, providing all-round and blind-angle battlefield information. However, panoramic vision technology also has some disadvantages: 1. Huge amount of data: The amount of image data generated is huge, which has high requirements for storage and transmission, affecting decision-making efficiency.

[0007] 2. High computing power requirements: Image stitching, correction, and real-time processing require powerful computing power. Edge devices must have high-performance processors and sufficient memory, otherwise image processing delays will increase.

[0008] 3. System complexity: The system is complex to build and maintain, requiring professional technicians, making troubleshooting difficult and affecting rapid deployment and maintenance.

[0009] 4. Poor environmental adaptability: Extreme conditions (such as high temperature, low temperature or strong electromagnetic interference) affect performance, causing electronic equipment to be unstable and affecting the normal operation of the system.

[0010] Therefore, in order to overcome the limitations of traditional methods in information acquisition, processing and transmission, this study proposes a battlefield situation awareness management method and system based on panoramic vision. The system aims to provide more comprehensive, real-time and accurate battlefield situation information by optimizing data processing flow, enhancing computing power, and improving system flexibility and adaptability, thereby improving the efficiency and accuracy of decision support. Summary of the invention

[0011] In order to overcome the shortcomings of traditional methods in information acquisition, processing and transmission, the present invention provides a battlefield situation awareness management method and system based on panoramic vision.

[0012] The technical implementation scheme of the present invention is: a battlefield situation awareness management method based on panoramic vision, comprising the following steps: S1: Obtain panoramic visual data and preprocess it, then classify the field of view; S2: Based on the field of view classification results, a range expansion model is constructed, and the battlefield situation prediction factor is obtained using the range expansion model; S3: Classify the battlefield situation security level according to the battlefield situation prediction factors, and determine the visualization display content according to the security level classification results; S4: The above steps form a closed-loop system and optimize the system by continuously collecting new data and feedback.

[0013] Preferably, the obtaining of panoramic visual field range data and preprocessing, and then classifying the field range, includes: the preprocessing includes data cleaning and standardization; the classifying of the field range includes dividing the field range into a front view area, a rear view area, a side view area, an upper view area, and a bottom view area; Among them, the angles of the front viewing area, the rear viewing area, the side viewing area, and the top viewing area are the first observation angles, and the angle of the bottom viewing area is the second observation angle.

[0014] Preferably, constructing a range telescopic model according to the field of view range classification result, and obtaining a battlefield situation prediction factor using the range telescopic model, includes: constructing a telescopic range model using the data of the front viewing area, rear viewing area, and side viewing area of ​​the first observation angle; obtaining a training set of the data of the front viewing area, rear viewing area, and side viewing area, inputting the training set data into the telescopic range model to obtain a trained telescopic range model, and obtaining the battlefield situation prediction factor using the trained telescopic range model.

[0015] Preferably, the step of obtaining the trained telescopic range model includes: constructing a fan-shaped area based on the field of view range and the farthest observation distance as the radius; dividing the fan-shaped area into a first area and a second area, wherein the side close to the visual sensor is the first area, and the side far from the visual sensor is the second area; The specific division basis is as follows: first, a comprehensive value is obtained based on the computational complexity, the amount of computational data, and the computational resource requirements; then, the boundary between the first area and the second area is determined in combination with the SVM classification model; finally, the area of ​​each inefficient area in the second area that does not meet the real-time battlefield requirements is obtained through formula calculation.

[0016] Preferably, the comprehensive value is first obtained according to the computational complexity, the amount of computational data and the computational resource requirements, and then the boundary between the first area and the second area is determined in combination with the SVM classification model, and finally the area of ​​each inefficient area in the second area that does not meet the real-time battlefield requirements is obtained by formula calculation, including: the comprehensive value calculation formula is as follows,

[0017] in, is the comprehensive value, is the weight of computational complexity, is the weight of computing resources, To calculate the weight of the data volume, is the normalized value of the computational complexity, is the normalized value of computing resources, To calculate the normalized value of the data volume; The total area of ​​the second region is calculated as follows,

[0018] in, is the total area of ​​the second region, is the field of view in radians, For the farthest observation distance, is the radius of the first region; The area calculation formula for each inefficient area in the second region is as follows:

[0019] in, For the first , ) inefficient area, is the angular increment of the area of ​​each inefficient region, in radians, is the radius increment of each inefficient region area, is the radius of the first region, is the radius index of the inefficient area, is the angle index of the inefficient region area; The normalized calculation formula is as follows:

[0020] in, is the original value of a feature, is the minimum value of this feature, is the maximum value of this feature, is the normalized value, ranging from [0, 1]; The core calculation formula of the SVM classification model is as follows:

[0021] in, is the weight vector, is the input feature vector, is the bias term.

[0022] Preferably, the battlefield situation security level is classified according to the battlefield situation prediction factor, and the visual display content is determined according to the security level classification result, including: obtaining the area of ​​each inefficient area in the second area, constructing a buffer area at the junction of the first area and the second area, and the two endpoints of the buffer area are respectively located on the boundary line of the first area and the second area; constructing an angle between the two endpoints of the buffer area to the center line between the two inefficient areas to form an angle area; determining the curvature of the vertex position of the angle area, using the change of the curvature as the data source of the prediction factor, and using the prediction factor to classify the battlefield situation security level; the buffer area is defined as a dynamic area set at the junction of the first area and the second area, which is used to verify and confirm the reliability of the panoramic visual data; The specific basis for dividing the area is: obtain the starting point of the area of ​​each inefficient area in the second area, connect the two ends of the buffer area with the starting points of the two inefficient areas respectively, and form two lines; if the intersection of the two lines is located at the farthest end of the second area, the area where the intersection is located is determined as the angle construction area.

[0023] Preferably, determining the curvature of the position includes: the curvature is calculated using the following formula:

[0024] in, is the curvature, f(x) is the function of the curve; is the first-order derivative of f(x), is the second-order derivative of f(x).

[0025] Preferably, the construction of prediction factors to classify battlefield situation safety levels includes: recording the calculated curvature values ​​of each position, and recording the changes of the curvature of each position over time, with the recording time interval being t seconds; extracting the changes of the curvature of the front view area, the side view area, and the rear view area over time from the recorded data; using the changes of the curvature in each orientation over time as a prediction factor sequence; using the LSTM model and the prediction factor sequence as input to predict the best route for the next step; using the prediction factor sequence to generate a recommended route; using the difference between the best route and the recommended route as a criterion for judging the safety level; using the RNN model and the difference value as input to judge the safety level; Among them, the use of the difference between the best route and the recommended route as the criterion for judging the safety level includes: obtaining second observation angle data to supplement and verify the difference between the best route and the recommended route; if driving along the best route, using the second observation angle data to verify the rationality of the best route; if the second observation angle data meets the standard of the best route, increasing the recommended position value of the recommended route; if the second observation angle data does not meet the standard of the best route, reducing the recommended position value of the recommended route.

[0026] Preferably, the above steps constitute a closed-loop system, and the system is optimized by continuously collecting new data and feedback, including: obtaining the difference between the best route and the recommended route, and using the difference result as the telescopic range in the range telescopic model, that is, the buffer area; the buffer area is defined as a dynamic area set around the travel path, which is used to adjust and optimize the travel strategy; If the difference between the best route and the recommended route is negative, the buffer area approaches the first area, and the specific operation is to move the boundary of the buffer area k meters toward the first area; if the difference between the best route and the recommended route is positive, the buffer area approaches the second area, and the specific operation is to move the boundary of the buffer area k meters toward the second area; if the difference between the best route and the recommended route is 0, the second area expands outward according to the Transformer model, and the specific operation is to expand the boundary of the second area; then the above steps constitute a closed-loop system, and the system is optimized by continuously collecting new data and feedback to ensure continuous improvement and performance enhancement of the system.

[0027] A battlefield situation awareness management system based on panoramic vision, including: Data acquisition and preprocessing module: used to acquire panoramic visual data and perform preprocessing; Field of view classification module: used to classify the field of view; Range expansion model module: used to build a range expansion model and obtain battlefield situation prediction factors; Security level classification and visualization module: used to classify battlefield situation security levels according to battlefield situation prediction factors and determine the visualization display content; Feedback optimization module: used to optimize the system by continuously collecting new data and feedback; Communication and storage module: responsible for data transmission and storage.

[0028] Beneficial effects: First, by preprocessing the panoramic visual data and classifying the field of view, the quality and consistency of the data are ensured. Secondly, a range expansion model is constructed, and the battlefield situation prediction factors are used to improve the accuracy of situational awareness. Then, intuitive decision support is provided through security level classification and visualization. In addition, the buffer area is dynamically adjusted, and the travel strategy is flexibly optimized according to the difference between the best route and the recommended route, which enhances the adaptability and safety of the system. Finally, through the construction of a closed-loop system, new data and feedback are continuously collected, which realizes the continuous optimization and performance improvement of the system and ensures the long-term stability and reliability of the system. These measures jointly improve the efficiency and effectiveness of battlefield situation awareness and provide strong support for decision makers. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a flowchart of a battlefield situation awareness management method and system based on panoramic vision of the present invention; Figure 2 The present invention is a structural schematic diagram of a battlefield situation awareness management system based on panoramic vision. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] Embodiment 1: A battlefield situation awareness management method based on panoramic vision, such as Figure 1 As shown, the following steps are included: S1: Obtain panoramic visual data and preprocess it, then classify the field of view; S2: Based on the field of view classification results, a range expansion model is constructed, and the battlefield situation prediction factor is obtained using the range expansion model; S3: Classify the battlefield situation security level according to the battlefield situation prediction factors, and determine the visualization display content according to the security level classification results; S4: The above steps form a closed-loop system and optimize the system by continuously collecting new data and feedback.

[0032] Obtaining panoramic vision field of view data and preprocessing it, and then classifying the field of view, including: the preprocessing includes data cleaning and standardization; the classifying the field of view includes dividing the field of view into a front view area, a rear view area, a side view area, an upper view area, and a bottom view area; Among them, the angles of the front viewing area, the rear viewing area, the side viewing area, and the top viewing area are the first observation angles, and the angle of the bottom viewing area is the second observation angle.

[0033] Further explanation is that the field of view manages and analyzes data more effectively through clear classification and processing steps, improving the overall performance and accuracy of the system. Field of view front area: the field of view in front of the tank, mainly used to detect the enemy and obstacles in front. Field of view rear area: the field of view behind the tank, mainly used to monitor the enemy and friendly dynamics behind. Field of view side area: the field of view on both sides of the tank, mainly used to monitor the safety of the flanks. Field of view upper area: the field of view above the tank, mainly used to detect aerial threats such as drones and helicopters. Field of view bottom area: the field of view below the tank, mainly used to detect obstacles and potential traps on the ground. Through clear classification and processing steps, data can be managed and analyzed more effectively, improving the overall performance and accuracy of the system. Through the above field of view classification and data processing process, the system can identify and classify targets faster, provide real-time battlefield situation information, and help tank commanders make more informed decisions; the reason why the bottom area is classified separately is that it is mainly used to detect obstacles and potential traps on the ground.

[0034] According to the field of view range classification result, a range telescopic model is constructed, and battlefield situation prediction factors are obtained by using the range telescopic model, including: using the front view area, rear view area, and side view area data of the first observation angle to construct a telescopic range model; obtaining a training set of the front view area, rear view area, and side view area data, inputting the training set data into the telescopic range model to obtain a trained telescopic range model, and obtaining the battlefield situation prediction factors by using the trained telescopic range model.

[0035] A further explanation is that the model is constructed using data from the first observation angle (forward viewing area, rear viewing area, and side viewing area) to ensure the pertinence and effectiveness of the model. By constructing and training the model using data from a specific area, the characteristics of the battlefield situation can be better captured and reflected, and the accuracy and reliability of the model can be improved. Training data is obtained from the forward viewing area, rear viewing area, and side viewing area to ensure that the model can process situation information from different directions and improve the generalization ability of the model. Using the trained telescopic range model, battlefield situation prediction factors are extracted to provide support for subsequent situation analysis and decision-making. By constructing and training the model using data from a specific area, the characteristics of the battlefield situation can be better captured and reflected, and the accuracy and reliability of the model can be improved.

[0036] Obtaining a trained telescopic range model, including: constructing a fan-shaped area based on the field of view range and the farthest observation distance as a radius; dividing the fan-shaped area into a first area and a second area, wherein a side close to the visual sensor is the first area, and a side far from the visual sensor is the second area; The specific division basis is as follows: first, a comprehensive value is obtained based on the computational complexity, the amount of computational data, and the computational resource requirements; then, the boundary between the first area and the second area is determined in combination with the SVM classification model; finally, the area of ​​each inefficient area in the second area that does not meet the real-time battlefield requirements is obtained through formula calculation.

[0037] A further explanation is that, based on the field of view range and the farthest observation distance as the radius, a sector area is constructed to ensure that the model covers the entire observation range. The sector area is divided into the first area and the second area, and the areas close to the visual sensor and far away from the visual sensor are clearly distinguished, which helps to concentrate resources and attention. A comprehensive value is obtained based on the computational complexity, the amount of computational data, and the computational resource requirements to ensure the efficiency and feasibility of the model. The boundary between the first area and the second area is determined in combination with the SVM classification model to improve the accuracy and reliability of the division. The area of ​​each inefficient area in the second area that does not meet the real-time battlefield requirements is obtained through formula calculation to ensure that the model can accurately identify and process key areas. Through clear regional division and scientific division basis, data can be managed and analyzed more effectively to improve the overall performance and accuracy of the system.

[0038] First, the comprehensive value is obtained according to the computational complexity, the amount of computational data, and the computational resource requirements. Then, the boundary between the first area and the second area is determined by combining the SVM classification model. Finally, the area of ​​each inefficient area in the second area that does not meet the real-time battlefield requirements is obtained by formula calculation, including: The comprehensive value calculation formula is as follows,

[0039] in, is the comprehensive value, is the weight of computational complexity, is the weight of computing resources, To calculate the weight of the data volume, is the normalized value of the computational complexity, is the normalized value of computing resources, To calculate the normalized value of the data volume; Further explanation is, : Comprehensive value, indicating the comprehensive computational burden of a certain area, : The weight of computational complexity, reflecting the importance of computational complexity in the comprehensive value. : The weight of computing resources, reflecting the importance of computing resources in the comprehensive value. : The weight of the calculated data volume, reflecting the importance of the calculated data volume in the comprehensive value. : The normalized value of the computational complexity, indicating the computational complexity of the area. : The normalized value of computing resources, indicating the computing resource requirements of the region. : The normalized value of the calculated data volume, which represents the calculated data volume of the area; the specific value of the weight can be determined through experiments or expert experience to ensure the accuracy and operability of the calculation.

[0040] The total area of ​​the second region is calculated as follows,

[0041] in, is the total area of ​​the second region, is the field of view in radians, For the farthest observation distance, is the radius of the first region; The calculation formula for the area of ​​each inefficient region in the second region is as follows:

[0042] in, For the first , ) inefficient area, is the angular increment of the area of ​​each inefficient region, in radians, is the radius increment of each inefficient region area, is the radius of the first region, is the radius index of the inefficient area, is the angle index of the inefficient region area; The normalized calculation formula is as follows:

[0043] in, is the original value of a feature, is the minimum value of this feature, is the maximum value of this feature, is the normalized value, ranging from [0, 1]; The core calculation formula of the SVM classification model is as follows:

[0044] in, is the weight vector, is the input feature vector, is the bias term.

[0045] Further explanation is, is the weight vector, indicating the direction of the hyperplane, is the input feature vector, which represents the characteristics of the sample. is the bias term, which represents the intercept of the hyperplane.

[0046] The battlefield situation security level is classified according to the battlefield situation prediction factor, and the visual display content is determined according to the security level classification result, including: obtaining the area of ​​each inefficient area in the second area, constructing a buffer area at the junction of the first area and the second area, and the two end points of the buffer area are respectively located on the boundary line of the first area and the second area; constructing an angle from the two end points of the buffer area to the center line between the two inefficient areas to form an angle area; determining the curvature of the vertex position of the angle area, using the change of curvature as the data source of the prediction factor, and using the prediction factor to classify the battlefield situation security level; the buffer area is defined as a dynamic area set at the junction of the first area and the second area, which is used to verify and confirm the reliability of the panoramic visual data; The specific basis for dividing the area is: obtain the starting point of the area of ​​each inefficient area in the second area, connect the two ends of the buffer area with the starting points of the two inefficient areas respectively, and form two lines; if the intersection of the two lines is located at the farthest end of the second area, the area where the intersection is located is determined as the angle construction area.

[0047] Further explanation is that it is clearly pointed out that the area of ​​each inefficient area in the second area needs to be obtained to ensure the integrity and accuracy of the data. Constructing a buffer area at the junction of the first area and the second area helps to smooth the transition and reduce misjudgment. Constructing an angle between the two inefficient areas with the two endpoints of the buffer area to obtain the area where the angle is constructed helps to identify the key area more accurately. Using the change in curvature as the data source of the prediction factor helps to capture subtle changes in terrain and environment. Using the prediction factor to classify the battlefield situation security level improves the accuracy and reliability of the classification. It is clearly pointed out that the starting point of the area of ​​each inefficient area in the second area needs to be obtained to ensure the accuracy and integrity of the data. Connecting the two ends of the buffer area with the starting points of the two inefficient areas to form a line. If the intersection of the lines intersects at the farthest end of the second area, it is determined to be the angle construction area, and the curvature of the area vertex is calculated to ensure the scientificity and accuracy of the division.

[0048] Further explanation is that the first buffer area: Definition: The panoramic vision reliability buffer area is a dynamic area set at the intersection of the first area and the second area, which is used to verify and confirm the reliability of the panoramic vision data and ensure the accuracy and security of the data. The second buffer area: Definition: The route optimization buffer area is a dynamic area set around the travel path based on the known reliability of the panoramic vision data, which is used to adjust and optimize the travel strategy to ensure the best travel route and safety. Relationship description: Dependency: The first buffer area: Through the preprocessing and verification of the panoramic vision data, the reliability and accuracy of the data are ensured, as the intersection area of ​​the first area and the second area. The second buffer area: After the reliability of the panoramic vision data is verified in the first buffer area, the route optimization buffer area is constructed based on these reliable data to optimize the travel route and ensure the best travel strategy. Logical order: Step 1: Construct the panoramic vision reliability buffer area to verify the reliability of the panoramic vision data. Step 2: Based on the verification results of the first buffer area, construct the route optimization buffer area to optimize the travel route.

[0049] Determine the curvature of the position, including: The curvature is calculated as follows,

[0050] in, is the curvature, f(x) is the function of the curve; is the first-order derivative of f(x), is the second-order derivative of f(x).

[0051] A further explanation is that the derivation method is: the function is derived according to the existing analytical method, numerical method or other method.

[0052] The construction of the prediction factor to classify the battlefield situation security level includes: recording the curvature calculation value of each position, and recording the change of the curvature of each position over time, the recording time interval is t seconds; extracting the change of the curvature of the front view area, the side view area, and the rear view area over time from the recorded data; using the change of the curvature in each direction over time as a prediction factor sequence; using the LSTM model, taking the prediction factor sequence as input, to predict the best route for the next step; using the prediction factor sequence to generate a recommended route; using the difference between the best route and the recommended route as a criterion for judging the security level; using the RNN model, taking the difference value as input, to judge the security level; Among them, the use of the difference between the best route and the recommended route as the criterion for judging the safety level includes: obtaining second observation angle data to supplement and verify the difference between the best route and the recommended route; if driving along the best route, using the second observation angle data to verify the rationality of the best route; if the second observation angle data meets the standard of the best route, increasing the recommended position value of the recommended route; if the second observation angle data does not meet the standard of the best route, reducing the recommended position value of the recommended route.

[0053] Further explanation is that the calculated curvature value of each position is recorded to ensure the integrity and accuracy of the data. At the same time, the change of curvature of each position over time is recorded to capture the dynamic changes of terrain and environment. Front view area, side view area, rear view area: obtain the change of curvature of the front view area, side view area, and rear view area over time to ensure the diversity and representativeness of the data. LSTM model: Use the LSTM model to predict the best route for the next step to improve the accuracy and reliability of the prediction. RNN model: Use the RNN model to judge the safety level to ensure that the assessment of the safety level is more scientific and accurate. The difference between the best route and the recommended route is used as the criterion for judging the safety level to ensure the comprehensiveness and objectivity of the assessment.

[0054] The above steps constitute a closed-loop system, and the system is optimized by continuously collecting new data and feedback, including: obtaining the difference between the best route and the recommended route, and using the difference result as the telescopic range in the range telescopic model, that is, the buffer area; the buffer area is defined as a dynamic area set around the travel path, which is used to adjust and optimize the travel strategy; If the difference between the best route and the recommended route is negative, the buffer area approaches the first area, and the specific operation is to move the boundary of the buffer area k meters toward the first area; if the difference between the best route and the recommended route is positive, the buffer area approaches the second area, and the specific operation is to move the boundary of the buffer area k meters toward the second area; if the difference between the best route and the recommended route is 0, the second area expands outward according to the Transformer model, and the specific operation is to expand the boundary of the second area; then the above steps constitute a closed-loop system, and the system is optimized by continuously collecting new data and feedback.

[0055] Further explanation is that the difference between the best route and the recommended route can be calculated by using the Euclidean distance; Obtain the difference between the best route and the recommended route: Purpose: Evaluate the gap between the current recommended route and the ideal route to determine the range that needs to be adjusted. Method: Calculate the difference between the best route and the recommended route. By quantifying the difference, the buffer area is adjusted more accurately to improve the adaptability and accuracy of the system. The difference result is used as the telescopic range in the range telescoping model, that is, the buffer area: Definition: The buffer area is a dynamic area set around the travel path to adjust and optimize the travel strategy. Function: The buffer area is dynamically adjusted according to the difference in the travel route to ensure the flexibility and safety of the travel strategy. Dynamic adjustment of the buffer area effectively responds to travel needs in different environments and improves the robustness and adaptability of the system. Adjustment of the buffer area: Negative value situation: Condition: The difference between the best route and the recommended route is negative. Operation: Move the boundary of the buffer area k meters toward the first area. When the recommended route is better than the best route, reduce the buffer area to reduce unnecessary resource consumption and improve travel efficiency. Positive value situation: Condition: The difference between the best route and the recommended route is positive. Operation: Move the boundary of the buffer area k meters toward the second area. When the recommended route is not as good as the best route, expand the buffer area to increase the safety margin and improve the safety of travel. Zero value situation: Condition: The difference between the best route and the recommended route is 0. Operation: Expand the boundary of the second area outward according to the Transformer model. When the recommended route is consistent with the best route, expand the second area to cover more potential risk areas and improve the system's foresight and response capabilities. The operating mechanism of the closed-loop system: Data collection: Continuously collect new data, including real-time data and environmental data of the route. Feedback mechanism: Feedback to the system based on the collected data and actual travel results. Optimization process: Use feedback data to adjust model parameters and optimize travel strategies and buffer area settings. Through the closed-loop system, the system continuously learns and improves to improve the accuracy and safety of the route and ensure the long-term stability and reliability of the system.

[0056] Example 2: Based on Example 1, a battlefield situation awareness management system based on panoramic vision includes:

[0057] Data acquisition and preprocessing module: used to acquire panoramic visual data and perform preprocessing; Field of view classification module: used to classify the field of view; Range expansion model module: used to build a range expansion model and obtain battlefield situation prediction factors; Security level classification and visualization module: used to classify battlefield situation security levels according to battlefield situation prediction factors and determine the visualization display content; Feedback optimization module: used to optimize the system by continuously collecting new data and feedback; Communication and storage module: responsible for data transmission and storage.

[0058] The above is a detailed introduction to the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A battlefield situation awareness management method based on panoramic vision, characterized in that: The following steps are involved: S1: Obtain panoramic visual data and preprocess it, then classify the field of view; S2: Based on the field of view classification results, a range expansion model is constructed, and the battlefield situation prediction factor is obtained using the range expansion model; S3: Classify the battlefield situation security level according to the battlefield situation prediction factors, and determine the visualization display content according to the security level classification results; S4: The above steps form a closed-loop system and optimize the system by continuously collecting new data and feedback.

2. A battlefield situation awareness management method based on panoramic vision as claimed in claim 1, characterized in that: The obtaining of panoramic visual field range data and preprocessing, and then classifying the field range, includes: the preprocessing includes data cleaning and standardization; the classifying of the field range includes dividing the field range into a front view area, a rear view area, a side view area, an upper view area, and a bottom view area; Among them, the angles of the front viewing area, the rear viewing area, the side viewing area, and the top viewing area are the first observation angles, and the angle of the bottom viewing area is the second observation angle.

3. A battlefield situation awareness management method based on panoramic vision as claimed in claim 1, characterized in that: The method of constructing a range telescopic model according to the field of view classification result and obtaining a battlefield situation prediction factor using the range telescopic model includes: constructing a telescopic range model using the data of the front view area, the rear view area, and the side view area at a first observation angle; obtaining a training set of the data of the front view area, the rear view area, and the side view area, inputting the training set data into the telescopic range model to obtain a trained telescopic range model, and obtaining the battlefield situation prediction factor using the trained telescopic range model.

4. A battlefield situation awareness management method based on panoramic vision as claimed in claim 3, characterized in that: The method of obtaining a trained telescopic range model includes: constructing a fan-shaped area based on the field of view range and the farthest observation distance as a radius; dividing the fan-shaped area into a first area and a second area, wherein a side close to the visual sensor is the first area, and a side far from the visual sensor is the second area; The specific division basis is as follows: first, a comprehensive value is obtained based on the computational complexity, the amount of computational data, and the computational resource requirements; then, the boundary between the first area and the second area is determined in combination with the SVM classification model; finally, the area of ​​each inefficient area in the second area that does not meet the real-time battlefield requirements is obtained through formula calculation.

5. A battlefield situation awareness management method based on panoramic vision as claimed in claim 4, characterized in that: The method first obtains a comprehensive value according to the computational complexity, the amount of computational data, and the computational resource requirements, then determines the boundary between the first area and the second area in combination with the SVM classification model, and finally obtains the area of ​​each inefficient area in the second area that does not meet the real-time battlefield requirements through a formula calculation, including: the comprehensive value calculation formula is as follows, in, is the comprehensive value, is the weight of computational complexity, is the weight of computing resources, To calculate the weight of the data volume, is the normalized value of the computational complexity, is the normalized value of computing resources, To calculate the normalized value of the data volume; The total area of ​​the second region is calculated as follows, in, is the total area of ​​the second region, is the field of view in radians, For the farthest observation distance, is the radius of the first region; The area calculation formula for each inefficient area in the second region is as follows: in, For the first , ) inefficient area, is the angular increment of the area of ​​each inefficient region, in radians, is the radius increment of each inefficient area, is the radius of the first region, is the radius index of the inefficient area, is the angle index of the inefficient region area; The normalized calculation formula is as follows: in, is the original value of a feature, is the minimum value of this feature, is the maximum value of this feature, is the normalized value, ranging from [0, 1]; The core calculation formula of the SVM classification model is as follows: in, is the weight vector, is the input feature vector, is the bias term.

6. A battlefield situation awareness management method based on panoramic vision as claimed in claim 1, characterized in that: The battlefield situation security level is classified according to the battlefield situation prediction factor, and the visual display content is determined according to the security level classification result, including: obtaining the area of ​​each inefficient area in the second area, constructing a buffer area at the junction of the first area and the second area, and the two end points of the buffer area are respectively located on the boundary line of the first area and the second area; constructing an angle between the two end points of the buffer area and the center line between the two inefficient areas to form an angle area; determining the curvature of the vertex position of the angle area, using the change of the curvature as the data source of the prediction factor, and using the prediction factor to classify the battlefield situation security level; the buffer area is defined as a dynamic area set at the junction of the first area and the second area, which is used to verify and confirm the reliability of the panoramic visual data; The specific basis for dividing the area is: obtain the starting point of the area of ​​each inefficient area in the second area, connect the two ends of the buffer area with the starting points of the two inefficient areas respectively, and form two lines; if the intersection of the two lines is located at the farthest end of the second area, the area where the intersection is located is determined as the angle construction area.

7. A battlefield situation awareness management method based on panoramic vision as claimed in claim 6, characterized in that: Determining the curvature of the position includes: the curvature calculation formula is as follows, in, is the curvature, f(x) is the function of the curve; is the first-order derivative of f(x), is the second-order derivative of f(x).

8. A battlefield situation awareness management method based on panoramic vision as claimed in claim 6, characterized in that: The construction of the prediction factor to classify the battlefield situation security level includes: recording the curvature calculation value of each position, and recording the change of the curvature of each position over time, the recording time interval is t seconds; extracting the change of the curvature of the front view area, the side view area, and the rear view area over time from the recorded data; using the change of the curvature in each direction over time as a prediction factor sequence; using the LSTM model, taking the prediction factor sequence as input, to predict the best route for the next step; using the prediction factor sequence to generate a recommended route; using the difference between the best route and the recommended route as a criterion for judging the security level; using the RNN model, taking the difference value as input, to judge the security level; Among them, the use of the difference between the best route and the recommended route as the criterion for judging the safety level includes: obtaining second observation angle data to supplement and verify the difference between the best route and the recommended route; if driving along the best route, using the second observation angle data to verify the rationality of the best route; if the second observation angle data meets the standard of the best route, increasing the recommended position value of the recommended route; if the second observation angle data does not meet the standard of the best route, reducing the recommended position value of the recommended route.

9. A battlefield situation awareness management method based on panoramic vision as claimed in claim 1, characterized in that: The above steps constitute a closed-loop system, and the system is optimized by continuously collecting new data and feedback, including: obtaining the difference between the best route and the recommended route, and using the difference result as the telescopic range in the range telescopic model, that is, the buffer area; the buffer area is defined as a dynamic area set around the travel path, which is used to adjust and optimize the travel strategy; If the difference between the best route and the recommended route is negative, the buffer area approaches the first area, and the specific operation is to move the boundary of the buffer area k meters toward the first area; if the difference between the best route and the recommended route is positive, the buffer area approaches the second area, and the specific operation is to move the boundary of the buffer area k meters toward the second area; if the difference between the best route and the recommended route is 0, the second area expands outward according to the Transformer model, and the specific operation is to expand the boundary of the second area; then the above steps constitute a closed-loop system, and the system is optimized by continuously collecting new data and feedback to ensure continuous improvement and performance enhancement of the system.

10. A battlefield situation awareness management system based on panoramic vision for use in any one of claims 1-9, characterized in that: Included are: Data acquisition and preprocessing module: used to acquire panoramic visual data and perform preprocessing; Field of view classification module: used to classify the field of view; Range expansion model module: used to build a range expansion model and obtain battlefield situation prediction factors; Security level classification and visualization module: used to classify battlefield situation security levels according to battlefield situation prediction factors and determine the visualization display content; Feedback optimization module: used to optimize the system by continuously collecting new data and feedback; Communication and storage module: responsible for data transmission and storage.

Images