A communication control system and method based on artificial intelligence

Abstract

The present invention relates to the field of security management technology. The present invention discloses a communication control system and method based on artificial intelligence, including obtaining the number of vehicles entering an edge control area, determining the number of monitoring equipment based on the number of vehicles, sending a standard moving speed, then obtaining the license plate number of the vehicle, judging whether to adopt a first mode or a second mode based on the license plate number, processing a real-time vehicle image set based on the license plate number to obtain a target vehicle image set, inputting the target vehicle image set into a pre-built leakage monitoring model, obtaining a monitoring category, obtaining a regional image corresponding to each post, determining a post status based on the regional image, and selecting a target post according to the post status. In this way, the present invention can control vehicles entering an area near a military camp, and when a soldier inside a vehicle takes a photo of a scene in a nearby area through a mobile phone, the corresponding vehicle is immediately warned, thereby achieving strict confidentiality of the location of the military camp.

Description

Technical Field

[0001] The present invention relates to the field of security management technology, and more specifically, to an artificial intelligence-based communication management and control system and method. Background Art

[0002] Smart military camps refer to making full use of new generation information technologies such as "Internet of Things", "cloud computing" and "big data processing", through the construction, transformation and integration of various information infrastructure and the innovation of information support environment, closely focusing on the needs of the troops, deeply integrating and developing and utilizing military information resources, and implanting digital and intelligent elements into the training, daily management, comprehensive support, office work and other businesses of the troops, changing the daily office and business processing methods of the military camps, making the military camp environment more intelligent, the military work more efficient, and the military management more intensive, and promoting the information construction of the troops to a higher level. In the existing technology, the communication management in the military camp is usually the management of the internal part of the military camp, such as personnel control or gun control. Due to the particularity of the military camp, the area near the military camp also needs to be controlled;

[0003] For example, the Chinese patent with publication number CN111369222A provides a military camp control method and related equipment, which receives raw data collected by a data collection device and sent by a data collection device, processes the raw data through a data layer and an intermediate layer, obtains business type data, and sends a military camp management instruction to a client device. The military camp management instruction carries the business type data, and the military camp management instruction is used to instruct the client device to control the military camp based on the business type data through the business layer;

[0004] Although the above patent controls the military camp through data, the above equipment also controls the internal system of the military camp through data. However, in reality, due to the particularity of the military camp, the area near the military camp may be radio shielded. Therefore, it is necessary to control the vehicles entering the area near the military camp, and the soldiers inside the vehicle use their mobile phones to take pictures of the scenery of the nearby area, which may easily lead to the leakage of the location of the military camp.

[0005] In view of this, the present invention proposes a communication control system and method based on artificial intelligence to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a communication management and control system and method based on artificial intelligence.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A communication control method based on artificial intelligence, comprising:

[0009] S10: Obtain the number of vehicles entering the edge control area, determine the number of monitoring devices based on the number of vehicles, and send a standard moving speed;

[0010] S20: Acquire the license plate number of the vehicle, and determine whether to use the first mode or the second mode based on the license plate number, wherein the first mode is to only take pictures without recognition, and the second mode is to take pictures and perform recognition;

[0011] S30: Based on the real-time vehicle image set entering the intermediate control area obtained in the second mode, the real-time vehicle image set is processed based on the license plate number to obtain a target vehicle image set, and the target vehicle image set is input into a pre-built leakage monitoring model to obtain a monitoring category, which includes leakage warning and non-warning. The intermediate control area is smaller than the edge control area;

[0012] S40: When entering the core control area, obtain the area image corresponding to each position, determine the position status based on the area image, and select the target position according to the position status, and the core control area is smaller than the intermediate control area.

[0013] Furthermore, the method for determining the number of monitoring devices based on the number of vehicles includes:

[0014] The number of monitoring devices is determined based on the number of vehicles and the mapping relationship.

[0015] Furthermore, the method for generating the standard moving speed includes:

[0016] The mileage and driving period corresponding to the vehicle are obtained, the initial driving speed is calculated based on the mileage and driving period, the real-time wind data and real-time precipitation are obtained, the initial driving speed, the real-time wind data and the real-time precipitation are input into the speed correction model, and the standard moving speed is obtained.

[0017] Furthermore, the method for constructing the velocity correction model includes:

[0018] A sample data set is obtained, wherein the sample data set includes a historical initial driving speed, historical wind data, historical precipitation, and a historical standard moving speed. The sample data set is divided into a sample training set and a sample test set, and a regression network is constructed. The historical initial driving speed, historical wind data, and historical precipitation in the sample training set are used as input data of the regression network, and the historical standard moving speed in the sample training set is used as output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the real-time standard moving speed. The initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as a speed correction model.

[0019] Further, the method of determining whether to adopt the first mode or the second mode based on the license plate number includes:

[0020] The license plate number is input into a pre-built license plate database to determine whether the license plate number belongs to the license plate database. When the license plate number does not belong to the license plate database, the second mode is adopted. When the license plate number belongs to the license plate database, it is further determined whether the license plate number is in the white list in the license plate database. If the license plate number is not in the white list, the second mode is adopted. If the license plate number is in the white list, the first mode is adopted.

[0021] Further, the method for determining whether the license plate number belongs to the license plate database includes:

[0022] Traverse the license plate database and calculate the similarity between the license plate number and each license plate data in the license plate database. If there is a license plate number with a similarity greater than a preset similarity threshold, it is determined that the license plate number is in the license plate database. If not, it is determined that the license plate number is not in the license plate database.

[0023] Furthermore, the method of processing the real-time vehicle image set based on the license plate number to obtain the target vehicle image set includes:

[0024] The characters in the license plate number that are not numbers are eliminated to obtain the target digital sequence, and the first N non-zero digits of the target digital sequence are taken. Based on the first N non-zero digits, the corresponding real-time vehicle images in the real-time vehicle image set are marked respectively, so as to obtain the target vehicle image set, where N is an integer greater than 0.

[0025] Furthermore, it also includes:

[0026] If the signal connection with the target vehicle cannot be established, the target vehicle image set is subjected to feature extraction processing according to the preset W extraction depths and feature pyramid network to obtain the W-layer feature map corresponding to the target vehicle image set. Figure 1 One-to-one, compare the W-layer feature map with the preset feature database to obtain the comparison result, and determine whether to stop the target vehicle based on the comparison result. The comparison result includes feature consistency and feature inconsistency;

[0027] Methods for determining whether to stop the target vehicle based on the comparison results include:

[0028] When the comparison result is that the features are inconsistent, the target vehicle is stopped; when the comparison result is that the features are consistent, the target vehicle is not stopped.

[0029] When the comparison result is that the features are consistent, the real-time moving trajectory of the target vehicle is further obtained according to the W-layer feature map, and the probability of the license plate being copied is determined according to the real-time moving trajectory. When the probability of the license plate being copied is greater than the preset probability threshold, the target vehicle is stopped.

[0030] Furthermore, the method for obtaining the real-time moving trajectory of the target vehicle according to the W-layer feature graph includes:

[0031] Based on the target detection algorithm, the target vehicle in the W-layer feature map is located, and R vehicle movement features in the W-layer feature map are extracted and spliced ​​to obtain the real-time movement trajectory of the target vehicle.

[0032] Furthermore, the method for determining the probability of a duplicate card according to the real-time moving trajectory includes:

[0033] The real-time position coordinates of the target vehicle are determined according to the real-time moving trajectory, and the coordinates of the nearest checkpoint in Q preset standard trajectories are determined based on the real-time position coordinates. The corresponding trajectory weight is assigned according to the distance between the real-time position coordinates and the checkpoint coordinates, and the similarity between the real-time moving trajectory and the Q preset standard trajectories is calculated. The trajectory matching degree is calculated according to the trajectory weight and similarity, and the inverse of the trajectory matching degree is used as the probability of duplication.

[0034] A communication control system based on artificial intelligence, which is used to implement the above-mentioned communication control method based on artificial intelligence, includes:

[0035] Speed ​​control module: used to obtain the number of vehicles entering the edge control area, determine the number of monitoring devices based on the number of vehicles, and send the standard moving speed;

[0036] Mode selection module: used to obtain the license plate number of the vehicle, and determine whether to use the first mode or the second mode based on the license plate number. The first mode is to only take pictures without recognition, and the second mode is to take pictures and recognize;

[0037] Control module: used to obtain a real-time vehicle image set entering the intermediate control area based on the second mode, process the real-time vehicle image set based on the license plate number to obtain a target vehicle image set, input the target vehicle image set into a pre-built leakage monitoring model, and obtain a monitoring category, which includes leakage warning and non-warning. The intermediate control area is smaller than the edge control area;

[0038] Position selection module: used to obtain the regional image corresponding to each position when entering the core control area, determine the position status based on the regional image, and select the target position according to the position status. The core control area is smaller than the intermediate control area.

[0039] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements the above-mentioned communication control method based on artificial intelligence.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present invention first obtains the number of vehicles entering the edge control area, determines the number of monitoring equipment based on the number of vehicles, and sends a standard moving speed, then obtains the license plate number of the vehicle, determines whether to use the first mode or the second mode based on the license plate number, obtains a real-time vehicle image set entering the intermediate control area based on the second mode, processes the real-time vehicle image set based on the license plate number to obtain a target vehicle image set, inputs the target vehicle image set into a pre-built leakage monitoring model, obtains the monitoring category, and when entering the core control area, obtains the regional image corresponding to each post, determines the post status based on the regional image, and selects the target post according to the post status. In this way, the present invention can control vehicles entering the area near the military camp, and when the soldiers inside the vehicle use mobile phones to shoot the scenery of the nearby area, the target vehicle is determined according to the license plate number and the corresponding real-time vehicle image set, and the corresponding target vehicle is immediately warned, so as to achieve strict confidentiality of the location of the military camp. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a flow chart of a communication control method based on artificial intelligence in the present invention;

[0043] Figure 2 It is a communication control system based on artificial intelligence in the present invention;

[0044] Figure 3 A schematic diagram of a computer-readable storage medium in the present invention. DETAILED DESCRIPTION

[0045] 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.

[0046] Example 1

[0047] See also Figure 1 As shown, this embodiment discloses a communication control method based on artificial intelligence, including:

[0048] S10: Obtain the number of vehicles entering the edge control area, determine the number of monitoring devices based on the number of vehicles, and send a standard moving speed;

[0049] It should be noted that the edge control area refers to the outer area near the military camp. First, the vehicle needs to enter the edge control area, and then enter the middle control area and the core control area in turn. The number of vehicles can be directly obtained by the infrared sensor;

[0050] Methods for determining the number of monitoring devices based on the number of vehicles include:

[0051] Determine the number of monitoring devices based on the number of vehicles and the mapping relationship;

[0052] It should be understood that the monitoring device may be a drone. Usually, the driving conditions of vehicles in the monitoring area can be realized by low-altitude satellites. However, the peripheral area near the military camp is usually equipped with radio shielding equipment, which makes it impossible for low-altitude satellites to perform effective monitoring. Therefore, when a vehicle enters the peripheral area near the military camp, it is necessary to start the corresponding drone, and the drone follows the vehicle for real-time monitoring. The number of vehicles and the number of monitoring devices may be in a one-to-one correspondence, or one monitoring device may correspond to two vehicles. This embodiment does not limit this.

[0053] Standard movement speed generation methods include:

[0054] Obtaining the corresponding mileage and driving period of the vehicle, calculating the initial driving speed based on the mileage and driving period, obtaining real-time wind data and real-time precipitation, inputting the initial driving speed, real-time wind data and real-time precipitation into a speed correction model, and obtaining a standard moving speed;

[0055] In this embodiment, the mileage is usually a fixed value pre-stored in the system. For example, when the edge control area is a circular area, the mileage is usually the radius of the circular area, and the center of the circular area is the location of the military camp. The driving period refers to the difference between the specified arrival time and the time when the vehicle enters the edge control area. It can be understood that due to the discipline of the military camp, when the vehicle enters the edge control area, it is necessary to arrive at the military camp at the specified time. In order to make subsequent corrections, the arrival time specified in this embodiment can be advanced accordingly to facilitate subsequent adjustments, and avoid the situation where the military camp cannot arrive at the specified time. For example, when it is specified that the vehicle must arrive at the military camp before 3 pm, the embodiment can set 2 pm as the specified arrival time and calculate the initial driving speed;

[0056] It can be understood that the real-time wind data includes the real-time wind value and the wind direction. The positive or negative real-time wind value indicates the wind direction. When the wind direction is the same as the flying direction of the monitoring device, the real-time wind value is positive. When the wind direction is different from the flying direction of the monitoring device, the real-time wind value is negative. In this embodiment, in order for the monitoring device to better monitor the vehicle, the monitoring device and the vehicle are kept at the same moving speed. Taking the monitoring device as a drone as an example, the flying speed of the drone should be the same as the driving speed of the vehicle, so that the drone can monitor the corresponding vehicle in real time. The real-time wind data mentioned above mainly affects the drone, while the real-time precipitation affects the driving speed of the vehicle.

[0057] It should be added that when the wind direction in the real-time wind data is the same as the flight direction of the drone, the real-time wind value is proportional to the flight speed of the drone. When the wind direction is different from the flight direction of the drone, the real-time wind value is inversely proportional to the flight speed of the drone. For the safety of vehicle driving, the real-time precipitation is inversely proportional to the driving speed of the vehicle. It can be understood that the final standard moving speed requires both the vehicle to drive at the standard moving speed and the drone to fly at the standard moving speed. This embodiment takes into account the impact of real-time wind data on the monitoring equipment and the impact of real-time precipitation on vehicle driving. Through correction, the vehicle and the monitoring equipment can move at the same speed, so that the monitoring equipment can monitor the corresponding vehicle throughout the entire process.

[0058] The method for constructing the velocity correction model includes:

[0059] Acquire a sample data set, wherein the sample data set includes a historical initial driving speed, historical wind data, historical precipitation, and a historical standard moving speed; divide the sample data set into a sample training set and a sample test set; construct a regression network; use the historical initial driving speed, historical wind data, and historical precipitation in the sample training set as input data of the regression network; use the historical standard moving speed in the sample training set as output data of the regression network; train the regression network to obtain an initial regression network for predicting the real-time standard moving speed; use the sample test set to test the initial regression network; output an initial regression network that satisfies a preset error value as a speed correction model; and the initial regression network is preferably a neural network model;

[0060] It should be noted that the historical initial driving speed mentioned above is obtained by calculation before the experiment, while the historical wind data and historical precipitation are directly measured by sensors in the experimental environment, or can be generated by simulation in the experimental environment. For example, ANSYS is a multi-physics simulation software widely used in the engineering field, which can simulate the dynamic behavior of vehicles and aircraft, and consider the influence of environmental factors such as wind and rainfall on their movement. Similarly, the corresponding historical standard moving speed is obtained through simulation and combined with expert experience;

[0061] S20: Acquire the license plate number of the vehicle, and determine whether to use the first mode or the second mode based on the license plate number, wherein the first mode is to only take pictures without recognition, and the second mode is to take pictures and perform recognition;

[0062] In some embodiments, the license plate number can be obtained in a variety of ways, such as directly obtaining the license plate number image through a camera on a monitoring device, and segmenting the license plate number image to obtain the corresponding license plate number. For example, the color clustering method is to cluster the pixels in the image according to their colors to achieve image segmentation. In license plate number recognition, algorithms such as K-means clustering can be used to divide the colors in the image into several categories, and then segment the license plate number according to its color features;

[0063] The method for determining whether to adopt the first mode or the second mode based on the license plate number includes:

[0064] Input the license plate number into a pre-built license plate database, and determine whether the license plate number belongs to the license plate database. When the license plate number does not belong to the license plate database, the second mode is adopted. When the license plate number belongs to the license plate database, further determine whether the license plate number is in a white list in the license plate database. If the license plate number is not in the white list, the second mode is adopted. If the license plate number is in the white list, the first mode is adopted.

[0065] Methods for determining whether a license plate number belongs to a license plate database include:

[0066] Traverse the license plate database, calculate the similarity between the license plate number and each license plate data in the license plate database, if there is a license plate number with a similarity greater than a preset similarity threshold, then it is determined that the license plate number is in the license plate database, if not, then it is determined that the license plate number is not in the license plate database;

[0067] It should be understood that the above-mentioned license plate database is a collection of historical license plate numbers, and the license plate data mentioned above is also the historical license plate number. In this embodiment, the purpose of taking pictures and identifying them through monitoring equipment is to quickly warn soldiers when they take pictures of the scenery near the military camp through their mobile phones. However, it also reduces the consumption of computing resources and the pressure on the system. The vehicles transporting soldiers all have corresponding license plate numbers. By identifying the license plate number, it is determined whether the current vehicle is a vehicle transporting soldiers, so as to further decide whether to identify it.

[0068] In some embodiments, there are several methods for calculating the similarity between the license plate number and each license plate data in the license plate database. For example, the license plate number is converted into a first feature vector, the license plate data is converted into a second feature vector, and the cosine similarity between the first feature vector and the second feature vector is calculated. The value range of the cosine similarity is between -1 and 1. The closer the value is to 1, the more similar the two vectors are, and the closer the value is to -1, the less similar the two vectors are.

[0069] S30: Based on the real-time vehicle image set entering the intermediate control area obtained in the second mode, the real-time vehicle image set is processed based on the license plate number to obtain a target vehicle image set, and the target vehicle image set is input into a pre-built leakage monitoring model to obtain a monitoring category, which includes leakage warning and non-warning. The intermediate control area is smaller than the edge control area;

[0070] It should be noted that the real-time vehicle image set includes M real-time vehicle images, where M is an integer greater than 1, and each real-time vehicle image corresponds to at least one vehicle;

[0071] The method for processing the real-time vehicle image set based on the license plate number to obtain the target vehicle image set includes:

[0072] The characters of the license plate number that are not digits are removed to obtain the target digit sequence, the first N digits of the target digit sequence that are not 0 are taken, and the corresponding real-time vehicle images in the real-time vehicle image set are marked based on the digits of the first N digits that are not 0, so as to obtain the target vehicle image set, where N is an integer greater than 0;

[0073] It should be added that the license plate number is usually composed of Chinese characters, letters and numbers. In this embodiment, Chinese characters and letters are removed, only numbers are retained, and the numbers are combined in the original order to generate a target digital sequence. For example, the target digital sequence is 40269. If N is 3, the three numbers 4, 2 and 6 are taken respectively, and the corresponding real-time vehicle images are marked. The corresponding real-time vehicle image refers to that the order of the image in the real-time vehicle image set meets the preset rules. It can be understood that the real-time vehicle image set is a collection of M real-time vehicle images in a certain order, then the 4th real-time vehicle image, the 2nd real-time vehicle image and the 6th real-time vehicle image in the real-time vehicle image set can be marked respectively, or they can be marked according to the multiple method, for example, the 8th real-time vehicle image, the 4th real-time vehicle image and the 12th real-time vehicle image in the real-time vehicle image set are marked respectively, and this embodiment does not limit this;

[0074] Methods for labeling real-time vehicle images include:

[0075] Obtaining a target view display function and a watermark sequence, and watermarking the real-time vehicle image based on the target view display function and the watermark sequence;

[0076] It can be understood that the target view display function refers to the watermark code in OpenCV, and the watermark order in this embodiment is the same as the order of obtaining numbers. Following the above, if the three numbers 4, 2 and 6 are taken respectively, the watermark order is also to first watermark the 4th real-time vehicle image, then watermark the 2nd real-time vehicle image, and finally watermark the 6th real-time vehicle image. The purpose of this is to achieve a strong association between the target vehicle image set and the license plate number. For example, the target digital sequence corresponding to the first vehicle is 40269, and the target digital sequence corresponding to the second vehicle is 60249. The three numbers taken out are 2, 4 and 6. If the watermark is performed in the order of 2, 4, and 6, the first vehicle and the second vehicle cannot be distinguished. Therefore, the watermark order in this embodiment is the same as the order of obtaining numbers, which achieves a strong association between the target vehicle image set and the license plate number, which is convenient for quickly locating the corresponding license plate number according to the target vehicle image set and issuing a warning to the corresponding vehicle.

[0077] The construction method of the leakage monitoring model includes:

[0078] Obtain i groups of training data, where i is a positive integer greater than 1, and the training data includes a historical target vehicle image set and a historical monitoring category. The historical target vehicle image set and the historical monitoring category are used as sample sets, and the sample set is divided into a training set and a test set. A classifier is constructed, and the historical target vehicle image set in the training set is used as input data, and the historical monitoring category in the training set is used as output data. The classifier is trained to obtain an initial classifier, and the initial classifier is tested using the test set, and the initial classifier having an accuracy greater than a preset threshold is output as a leakage monitoring model, and the classifier is preferably a naive Bayes model or a support vector machine model;

[0079] In this embodiment, the real-time vehicle image set is detected by the leakage monitoring model. When the monitoring category is leakage warning, it indicates that a soldier has taken pictures of the scenery near the military camp with a mobile phone. It is necessary to immediately trace the vehicle corresponding to the real-time vehicle image set and warn the corresponding vehicle to avoid leakage of the military camp location.

[0080] S40: when entering the core control area, obtaining the regional image corresponding to each post, determining the post status based on the regional image, and selecting the target post according to the post status, wherein the core control area is smaller than the intermediate control area;

[0081] In this embodiment, the post refers to the post where the soldier is on guard, and the post status includes a handover state and a non-handover state. The handover state refers to the soldier is handing over the post;

[0082] The method of determining the job status based on the regional image and selecting the target job according to the job status includes:

[0083] The number of personnel in the regional image is obtained. When the number of personnel is greater than 1, the position status is a handover status. When the number of personnel is not greater than 1, the position status is a non-handover status. Any position in a non-handover status is selected as the target position.

[0084] It should be added that when a vehicle enters the core control area, it indicates that the vehicle is about to arrive at the military camp, so soldiers are required to check the vehicle. Then, it is best to avoid the situation where soldiers are taking over their posts when the vehicle arrives at the military camp, which can improve the efficiency of the inspection. There are many methods for obtaining the number of personnel in the regional image, such as using the YOLO target detection algorithm to identify personnel in the regional image. The YOLO target detection algorithm is a prior art and will not be described in detail in this embodiment.

[0085] In this embodiment, the number of vehicles entering the edge control area is first obtained, the number of monitoring devices is determined based on the number of vehicles, and the standard moving speed is sent, then the license plate number of the vehicle is obtained, and the first mode or the second mode is adopted based on the license plate number. A real-time vehicle image set entering the intermediate control area is obtained based on the second mode, and the real-time vehicle image set is processed based on the license plate number to obtain a target vehicle image set, which is input into a pre-built leakage monitoring model to obtain a monitoring category. When entering the core control area, the regional image corresponding to each post is obtained, the post status is determined based on the regional image, and the target post is selected according to the post status. In this way, this embodiment can control vehicles entering the area near the military camp, and when the soldiers inside the vehicle use their mobile phones to take pictures of the nearby area, the corresponding vehicles are immediately alerted, thereby achieving strict confidentiality of the location of the military camp.

[0086] Example 2

[0087] Based on Example 1, a communication control method based on artificial intelligence also includes:

[0088] If the signal connection with the target vehicle cannot be established, the target vehicle image set is subjected to feature extraction processing according to the preset W extraction depths and feature pyramid network to obtain the W-layer feature map corresponding to the target vehicle image set. Figure 1 One-to-one, compare the W-layer feature map with the preset feature database to obtain the comparison result, and determine whether to stop the target vehicle based on the comparison result. The comparison result includes feature consistency and feature inconsistency;

[0089] Methods for determining whether to stop the target vehicle based on the comparison results include:

[0090] When the comparison result is that the features are inconsistent, the target vehicle is stopped; when the comparison result is that the features are consistent, the target vehicle is not stopped.

[0091] When the comparison result is that the features are consistent, the real-time moving trajectory of the target vehicle is further obtained according to the W-layer feature map, and the probability of the license plate being copied is determined according to the real-time moving trajectory. When the probability of the license plate being copied is greater than the preset probability threshold, the target vehicle is stopped.

[0092] It should be noted that the feature pyramid network is a deep learning model structure for computer vision tasks. Its main feature is to enhance the model's detection ability for objects of different scales by constructing a multi-scale feature map pyramid from high resolution to low resolution. The feature pyramid network uses a multi-layer feature map of a basic convolutional neural network (such as ResNet) to extract features from shallow to deep layers. These feature maps have different resolutions and can capture information of different scales.

[0093] Taking W as 3 as an example, if the target vehicle is a military truck, three layers of feature maps will be generated when feature extraction is performed on the target vehicle image set. The first layer of feature maps: contains some basic shape, color, texture and other features, which are used to capture the overall appearance features of the military truck. The second layer of feature maps: extracts more abstract and advanced features, such as the features of the vehicle's tires, body structure, and compartment, in order to more accurately identify the detailed features of the military truck. The third layer of feature maps: contains deeper features, such as vehicle logos, armored structures unique to military vehicles, weapon systems and other features.

[0094] Comparing the W-layer feature map with the preset feature database can be to vectorize the feature map, calculate the real-time similarity between the vectors, and pre-set the corresponding similarity threshold. When the real-time similarity is greater than the similarity threshold, the comparison result is that the features are consistent, and when it does not appear, the comparison result is that the features are inconsistent.

[0095] The method for obtaining the real-time moving trajectory of the target vehicle according to the W-layer feature map includes:

[0096] Based on the target detection algorithm, the target vehicle in the W-layer feature map is located, and R vehicle movement features in the W-layer feature map are extracted and spliced ​​to obtain the real-time movement trajectory of the target vehicle.

[0097] In this embodiment, the target detection algorithm can be a Faster R-CNN algorithm or a YOLO algorithm. The vehicle movement feature here can be the tilt angle of the vehicle tire, the steering angle of the vehicle, or the displacement change of the vehicle tire. The method of splicing R vehicle movement features can be to convert the R vehicle movement features into vectors in sequence, and splice them in order to form a comprehensive vector, and input the comprehensive vector into an existing neural network model for trajectory prediction to obtain a real-time movement trajectory. The advantage of this embodiment of obtaining the real-time movement trajectory through a W-layer feature map instead of signal positioning is that the correlation and timing information between different features are fully explored, which improves the comprehensive understanding and prediction accuracy of the target vehicle's movement behavior.

[0098] Methods for determining deck probability based on real-time movement trajectories include:

[0099] The real-time position coordinates of the target vehicle are determined according to the real-time moving trajectory, and the coordinates of the nearest checkpoint in Q preset standard trajectories are determined based on the real-time position coordinates. The corresponding trajectory weight is assigned according to the distance between the real-time position coordinates and the checkpoint coordinates, and the similarity between the real-time moving trajectory and the Q preset standard trajectories is calculated. The trajectory matching degree is calculated according to the trajectory weight and similarity, and the inverse of the trajectory matching degree is used as the probability of duplication.

[0100] Methods for calculating the trajectory matching degree based on trajectory weight and similarity include:

[0101] TMS= ;

[0102] Where TMS is the trajectory matching degree, For the The trajectory weight corresponding to the standard trajectory, For the The similarity between the standard trajectory and the real-time moving trajectory.

[0103] It should be added that the preset standard trajectory mentioned above refers to the trajectory route with the destination being the military camp center, which is located in the core control area. It can be understood that in this embodiment, the ultimate goal is to accurately control vehicles entering the area near the military camp. Several detection points are set on the trajectory route leading to the military camp center. When the target vehicle is a fake license plate vehicle, it will not move towards the nearest detection point. Therefore, in this embodiment, the corresponding trajectory weight is allocated according to the distance between the real-time position coordinates and the checkpoint coordinates. The closer the distance between the real-time position coordinates and the checkpoint coordinates, the greater the weight allocated to the corresponding trajectory. The advantage of this is that by allocating a larger weight to the trajectory close to the checkpoint, the system can respond more sensitively to the trajectory deviation of the vehicle when approaching the checkpoint. This method can detect abnormal trajectories in a timely manner and improve the system's ability to detect abnormal behavior.

[0104] In the above, calculating the similarity between the real-time moving trajectory and the Q preset standard trajectories can be achieved by calculating the cosine value between the real-time moving trajectory and the preset standard trajectory. The value of the cosine similarity is between -1 and 1. The closer the cosine value is to 1, the closer the directions of the two trajectories are.

[0105] In this embodiment, the target vehicle is first monitored for leaks. When a leak is found, the target vehicle can be contacted and warned based on the corresponding license plate number. However, when the contact cannot be made in time, or the target vehicle is a fake license plate vehicle, the characteristics of the target vehicle should be obtained. When inconsistencies are found after comparison, it is necessary to stop the vehicle immediately. When consistency is found, it may be due to signal reasons that the contact cannot be made in time. The real-time moving trajectory of the target vehicle can be further judged. When the probability of fake license plates is not greater than the preset probability threshold, it indicates that the target vehicle is not a fake license plate vehicle, and it only needs to be intercepted at the route checkpoint. When the probability of fake license plates is greater than the preset probability threshold, it indicates that the target vehicle is likely to be a fake license plate vehicle and needs to be stopped immediately. This can effectively deal with potential risks such as fake license plate vehicles and improve management and control efficiency and accuracy.

[0106] Example 3

[0107] like Figure 2 As shown, this embodiment provides a communication control system based on artificial intelligence on the basis of embodiment 1, including:

[0108] Speed ​​control module: used to obtain the number of vehicles entering the edge control area, determine the number of monitoring devices based on the number of vehicles, and send the standard moving speed;

[0109] It should be noted that the edge control area refers to the outer area near the military camp. First, the vehicle needs to enter the edge control area, and then enter the middle control area and the core control area in turn. The number of vehicles can be directly obtained by the infrared sensor;

[0110] Methods for determining the number of monitoring devices based on the number of vehicles include:

[0111] Determine the number of monitoring devices based on the number of vehicles and the mapping relationship;

[0112] It should be understood that the monitoring device may be a drone. Usually, the driving conditions of vehicles in the monitoring area can be realized by low-altitude satellites. However, the peripheral area near the military camp is usually equipped with radio shielding equipment, which makes it impossible for low-altitude satellites to perform effective monitoring. Therefore, when a vehicle enters the peripheral area near the military camp, it is necessary to start the corresponding drone, and the drone follows the vehicle for real-time monitoring. The number of vehicles and the number of monitoring devices may be in a one-to-one correspondence, or one monitoring device may correspond to two vehicles. This embodiment does not limit this.

[0113] Standard movement speed generation methods include:

[0114] Obtaining the corresponding mileage and driving period of the vehicle, calculating the initial driving speed based on the mileage and driving period, obtaining real-time wind data and real-time precipitation, inputting the initial driving speed, real-time wind data and real-time precipitation into a speed correction model, and obtaining a standard moving speed;

[0115] In this embodiment, the mileage is usually a fixed value pre-stored in the system. For example, when the edge control area is a circular area, the mileage is usually the radius of the circular area, and the center of the circular area is the location of the military camp. The driving period refers to the difference between the specified arrival time and the time when the vehicle enters the edge control area. It can be understood that due to the discipline of the military camp, when the vehicle enters the edge control area, it is necessary to arrive at the military camp at the specified time. In order to make subsequent corrections, the arrival time specified in this embodiment can be advanced accordingly to facilitate subsequent adjustments, and avoid the situation where the military camp cannot arrive at the specified time. For example, when it is specified that the vehicle must arrive at the military camp before 3 pm, the embodiment can set 2 pm as the specified arrival time and calculate the initial driving speed;

[0116] It can be understood that the real-time wind data includes the real-time wind value and the wind direction. The positive or negative real-time wind value indicates the wind direction. When the wind direction is the same as the flying direction of the monitoring device, the real-time wind value is positive. When the wind direction is different from the flying direction of the monitoring device, the real-time wind value is negative. In this embodiment, in order for the monitoring device to better monitor the vehicle, the monitoring device and the vehicle are kept at the same moving speed. Taking the monitoring device as a drone as an example, the flying speed of the drone should be the same as the driving speed of the vehicle, so that the drone can monitor the corresponding vehicle in real time. The real-time wind data mentioned above mainly affects the drone, while the real-time precipitation affects the driving speed of the vehicle.

[0117] It should be added that when the wind direction in the real-time wind data is the same as the flight direction of the drone, the real-time wind value is proportional to the flight speed of the drone. When the wind direction is different from the flight direction of the drone, the real-time wind value is inversely proportional to the flight speed of the drone. For the safety of vehicle driving, the real-time precipitation is inversely proportional to the driving speed of the vehicle. It can be understood that the final standard moving speed requires both the vehicle to drive at the standard moving speed and the drone to fly at the standard moving speed. This embodiment takes into account the impact of real-time wind data on the monitoring equipment and the impact of real-time precipitation on vehicle driving. Through correction, the vehicle and the monitoring equipment can move at the same speed, so that the monitoring equipment can monitor the corresponding vehicle throughout the entire process.

[0118] It should be noted that the historical initial driving speed mentioned above is obtained by calculation before the experiment, while the historical wind data and historical precipitation are directly measured by sensors in the experimental environment, or can be generated by simulation in the experimental environment. For example, ANSYS is a multi-physics simulation software widely used in the engineering field, which can simulate the dynamic behavior of vehicles and aircraft, and consider the influence of environmental factors such as wind and rainfall on their movement. Similarly, the corresponding historical standard moving speed is obtained through simulation and combined with expert experience;

[0119] Mode selection module: used to obtain the license plate number of the vehicle, and determine whether to use the first mode or the second mode based on the license plate number. The first mode is to only take pictures without recognition, and the second mode is to take pictures and recognize;

[0120] In some embodiments, the license plate number can be obtained in a variety of ways, such as directly obtaining the license plate number image through a camera on a monitoring device, and segmenting the license plate number image to obtain the corresponding license plate number. For example, the color clustering method is to cluster the pixels in the image according to their colors to achieve image segmentation. In license plate number recognition, algorithms such as K-means clustering can be used to divide the colors in the image into several categories, and then segment the license plate number according to its color features;

[0121] The method for determining whether to adopt the first mode or the second mode based on the license plate number includes:

[0122] Input the license plate number into a pre-built license plate database, and determine whether the license plate number belongs to the license plate database. When the license plate number does not belong to the license plate database, the second mode is adopted. When the license plate number belongs to the license plate database, further determine whether the license plate number is in a white list in the license plate database. If the license plate number is not in the white list, the second mode is adopted. If the license plate number is in the white list, the first mode is adopted.

[0123] Methods for determining whether a license plate number belongs to a license plate database include:

[0124] Traverse the license plate database, calculate the similarity between the license plate number and each license plate data in the license plate database, if there is a license plate number with a similarity greater than a preset similarity threshold, then it is determined that the license plate number is in the license plate database, if not, then it is determined that the license plate number is not in the license plate database;

[0125] It should be understood that the above-mentioned license plate database is a collection of historical license plate numbers, and the license plate data mentioned above is also the historical license plate number. In this embodiment, the purpose of taking pictures and identifying them through monitoring equipment is to quickly warn soldiers when they take pictures of the scenery near the military camp through their mobile phones. However, it also reduces the consumption of computing resources and the pressure on the system. The vehicles transporting soldiers all have corresponding license plate numbers. By identifying the license plate number, it is determined whether the current vehicle is a vehicle transporting soldiers, so as to further decide whether to identify it.

[0126] In some embodiments, there are several methods for calculating the similarity between the license plate number and each license plate data in the license plate database. For example, the license plate number is converted into a first feature vector, the license plate data is converted into a second feature vector, and the cosine similarity between the first feature vector and the second feature vector is calculated. The value range of the cosine similarity is between -1 and 1. The closer the value is to 1, the more similar the two vectors are, and the closer the value is to -1, the less similar the two vectors are.

[0127] Control module: used to obtain a real-time vehicle image set entering the intermediate control area based on the second mode, process the real-time vehicle image set based on the license plate number to obtain a target vehicle image set, input the target vehicle image set into a pre-built leakage monitoring model, and obtain a monitoring category, which includes leakage warning and non-warning. The intermediate control area is smaller than the edge control area;

[0128] It should be noted that the real-time vehicle image set includes M real-time vehicle images, where M is an integer greater than 1, and each real-time vehicle image corresponds to at least one vehicle;

[0129] The method for processing the real-time vehicle image set based on the license plate number to obtain the target vehicle image set includes:

[0130] The characters of the license plate number that are not digits are removed to obtain the target digit sequence, the first N digits of the target digit sequence that are not 0 are taken, and the corresponding real-time vehicle images in the real-time vehicle image set are marked based on the digits of the first N digits that are not 0, so as to obtain the target vehicle image set, where N is an integer greater than 0;

[0131] It should be added that the license plate number is usually composed of Chinese characters, letters and numbers. In this embodiment, Chinese characters and letters are removed, only numbers are retained, and the numbers are combined in the original order to generate a target digital sequence. For example, the target digital sequence is 40269. If N is 3, the three numbers 4, 2 and 6 are taken respectively, and the corresponding real-time vehicle images are marked. The corresponding real-time vehicle image refers to that the order of the image in the real-time vehicle image set meets the preset rules. It can be understood that the real-time vehicle image set is a collection of M real-time vehicle images in a certain order, then the 4th real-time vehicle image, the 2nd real-time vehicle image and the 6th real-time vehicle image in the real-time vehicle image set can be marked respectively, or they can be marked according to the multiple method, for example, the 8th real-time vehicle image, the 4th real-time vehicle image and the 12th real-time vehicle image in the real-time vehicle image set are marked respectively, and this embodiment does not limit this;

[0132] Methods for labeling real-time vehicle images include:

[0133] Obtaining a target view display function and a watermark sequence, and watermarking the real-time vehicle image based on the target view display function and the watermark sequence;

[0134] It can be understood that the target view display function refers to the watermark code in OpenCV, and the watermark order in this embodiment is the same as the order of obtaining numbers. Following the above, if the three numbers 4, 2 and 6 are taken respectively, the watermark order is also to first watermark the 4th real-time vehicle image, then watermark the 2nd real-time vehicle image, and finally watermark the 6th real-time vehicle image. The purpose of this is to achieve a strong association between the target vehicle image set and the license plate number. For example, the target digital sequence corresponding to the first vehicle is 40269, and the target digital sequence corresponding to the second vehicle is 60249. The three numbers taken out are 2, 4 and 6. If the watermark is performed in the order of 2, 4, and 6, the first vehicle and the second vehicle cannot be distinguished. Therefore, the watermark order in this embodiment is the same as the order of obtaining numbers, which achieves a strong association between the target vehicle image set and the license plate number, which is convenient for quickly locating the corresponding license plate number according to the target vehicle image set and issuing a warning to the corresponding vehicle.

[0135] In this embodiment, the real-time vehicle image set is detected by the leakage monitoring model. When the monitoring category is leakage warning, it indicates that a soldier has taken pictures of the scenery near the military camp with a mobile phone. It is necessary to immediately trace the vehicle corresponding to the real-time vehicle image set and warn the corresponding vehicle to avoid leakage of the military camp location.

[0136] Position selection module: used for obtaining the regional image corresponding to each position when entering the core control area, determining the position status based on the regional image, and selecting the target position according to the position status, wherein the core control area is smaller than the intermediate control area;

[0137] In this embodiment, the post refers to the post where the soldier is on guard, and the post status includes a handover state and a non-handover state. The handover state refers to the soldier is handing over the post;

[0138] The method of determining the job status based on the regional image and selecting the target job according to the job status includes:

[0139] The number of personnel in the regional image is obtained. When the number of personnel is greater than 1, the position status is a handover status. When the number of personnel is not greater than 1, the position status is a non-handover status. Any position in a non-handover status is selected as the target position.

[0140] It should be added that when a vehicle enters the core control area, it indicates that the vehicle is about to arrive at the military camp, so soldiers are required to check the vehicle. In this case, it is best to avoid the situation where soldiers are taking over their posts when the vehicle arrives at the military camp. This can improve the efficiency of the inspection. There are many ways to obtain the number of people in the regional image, such as using the YOLO target detection algorithm to identify people in the regional image. The YOLO target detection algorithm is a prior art and will not be described in detail in this embodiment.

[0141] Example 4

[0142] The present embodiment discloses an electronic device, including a power supply, an interface, a keyboard, a memory, a central processing unit, and a computer program stored in the memory and executable on the central processing unit. When the central processing unit executes the computer program, an artificial intelligence-based communication control method provided by the above methods is implemented. The interface includes a network interface and a data interface. The network interface includes a wired or wireless interface, and the data interface includes an input or output interface.

[0143] Since the electronic device introduced in this embodiment is an electronic device used to implement a communication control method based on artificial intelligence in the embodiment of the present application, based on the communication control method based on artificial intelligence introduced in the embodiment of the present application, the technical personnel of the field can understand the specific implementation of the electronic device of the present embodiment and its various variations, so how the electronic device implements the method in the embodiment of the present application is not described in detail here. As long as the technical personnel of the field implement the electronic device used in the communication control method based on artificial intelligence in the embodiment of the present application, it belongs to the scope of protection of this application.

[0144] Example 5

[0145] like Figure 3 As shown, this embodiment discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the above-mentioned communication control method based on artificial intelligence is implemented.

[0146] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0147] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A communication control method based on artificial intelligence, characterized in that: include: S10: Obtain the number of vehicles entering the edge control area, determine the number of monitoring devices based on the number of vehicles, and send a standard moving speed; S20: Acquire the license plate number of the vehicle, and determine whether to use the first mode or the second mode based on the license plate number, wherein the first mode is to only take pictures without recognition, and the second mode is to take pictures and perform recognition; S30: Based on the real-time vehicle image set entering the intermediate control area obtained in the second mode, the real-time vehicle image set is processed based on the license plate number to obtain a target vehicle image set, and the target vehicle image set is input into a pre-built leakage monitoring model to obtain a monitoring category, which includes leakage warning and non-warning. The intermediate control area is smaller than the edge control area. When the monitoring category is leakage warning, the vehicle corresponding to the real-time vehicle image set is traced, and the corresponding vehicle is warned; S40: when entering the core control area, obtaining the regional image corresponding to each post, determining the post status based on the regional image, and selecting the target post according to the post status, wherein the core control area is smaller than the intermediate control area; The construction method of the leakage monitoring model includes: Obtain i groups of training data, where i is a positive integer greater than 1, and the training data includes a historical target vehicle image set and a historical monitoring category. The historical target vehicle image set and the historical monitoring category are used as sample sets, and the sample sets are divided into a training set and a test set. A classifier is constructed, and the historical target vehicle image set in the training set is used as input data, and the historical monitoring category in the training set is used as output data. The classifier is trained to obtain an initial classifier, and the initial classifier is tested using the test set, and the initial classifier having an accuracy greater than a preset threshold is output as a leakage monitoring model.

2. According to claim 1, a communication control method based on artificial intelligence is characterized in that: The method for determining the number of monitoring devices based on the number of vehicles includes: The number of monitoring devices is determined based on the number of vehicles and the mapping relationship.

3. The communication control method based on artificial intelligence according to claim 1 is characterized in that: The method for generating the standard moving speed includes: The mileage and driving period corresponding to the vehicle are obtained, the initial driving speed is calculated based on the mileage and driving period, the real-time wind data and real-time precipitation are obtained, the initial driving speed, the real-time wind data and the real-time precipitation are input into the speed correction model, and the standard moving speed is obtained.

4. According to claim 3, a communication control method based on artificial intelligence is characterized in that: The method for constructing the speed correction model comprises: A sample data set is obtained, wherein the sample data set includes a historical initial driving speed, historical wind data, historical precipitation, and a historical standard moving speed. The sample data set is divided into a sample training set and a sample test set, and a regression network is constructed. The historical initial driving speed, historical wind data, and historical precipitation in the sample training set are used as input data of the regression network, and the historical standard moving speed in the sample training set is used as output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the real-time standard moving speed. The initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as a speed correction model.

5. The communication control method based on artificial intelligence according to claim 1, characterized in that: The method for determining whether to adopt the first mode or the second mode based on the license plate number includes: The license plate number is input into a pre-built license plate database to determine whether the license plate number belongs to the license plate database. When the license plate number does not belong to the license plate database, the second mode is adopted. When the license plate number belongs to the license plate database, it is further determined whether the license plate number is in the white list in the license plate database. If the license plate number is not in the white list, the second mode is adopted. If the license plate number is in the white list, the first mode is adopted.

6. According to claim 5, a communication control method based on artificial intelligence is characterized in that: The method for determining whether a license plate number belongs to a license plate database comprises: Traverse the license plate database and calculate the similarity between the license plate number and each license plate data in the license plate database. If there is a license plate number with a similarity greater than a preset similarity threshold, it is determined that the license plate number is in the license plate database. If not, it is determined that the license plate number is not in the license plate database.

7. The communication control method based on artificial intelligence according to claim 1, characterized in that: The method for processing the real-time vehicle image set based on the license plate number to obtain the target vehicle image set includes: The characters in the license plate number that are not numbers are eliminated to obtain the target digital sequence, and the first N non-zero digits of the target digital sequence are taken. Based on the first N non-zero digits, the corresponding real-time vehicle images in the real-time vehicle image set are marked respectively, so as to obtain the target vehicle image set, where N is an integer greater than 0.

8. The communication control method based on artificial intelligence according to claim 7 is characterized in that: Also includes: If it is impossible to establish a signal connection with the target vehicle, the target vehicle image set is subjected to feature extraction processing according to the preset W extraction depths and feature pyramid network to obtain W-layer feature maps corresponding to the target vehicle image set. The W extraction depths correspond to the W-layer feature maps one by one. The W-layer feature maps are compared with the preset feature database to obtain the comparison results. According to the comparison results, it is determined whether to stop the target vehicle. The comparison results include feature consistency and feature inconsistency. Methods for determining whether to stop the target vehicle based on the comparison results include: When the comparison result shows that the features are inconsistent, the target vehicle is stopped; when the comparison result shows that the features are consistent, the target vehicle is not stopped; When the comparison result is that the features are consistent, the real-time moving trajectory of the target vehicle is further obtained according to the W-layer feature map, and the probability of the license plate being copied is determined according to the real-time moving trajectory. When the probability of the license plate being copied is greater than the preset probability threshold, the target vehicle is stopped.

9. The communication control method based on artificial intelligence according to claim 8, characterized in that: The method for obtaining the real-time moving trajectory of the target vehicle according to the W-layer feature graph includes: Based on the target detection algorithm, the target vehicle in the W-layer feature map is located, and R vehicle movement features in the W-layer feature map are extracted and spliced ​​to obtain the real-time movement trajectory of the target vehicle.

10. The communication control method based on artificial intelligence according to claim 8, characterized in that: The method for determining the probability of a duplicate card according to the real-time moving trajectory comprises: The real-time position coordinates of the target vehicle are determined according to the real-time moving trajectory, and the coordinates of the nearest checkpoint in Q preset standard trajectories are determined based on the real-time position coordinates. The corresponding trajectory weight is assigned according to the distance between the real-time position coordinates and the checkpoint coordinates, and the similarity between the real-time moving trajectory and the Q preset standard trajectories is calculated. The trajectory matching degree is calculated according to the trajectory weight and similarity, and the inverse of the trajectory matching degree is used as the probability of duplication.

11. A communication control system based on artificial intelligence, which is used to implement a communication control method based on artificial intelligence as described in any one of claims 1 to 10, characterized in that: include: speed Control module: used to obtain the number of vehicles entering the edge control area, determine the number of monitoring devices based on the number of vehicles, and send the standard moving speed; Mode selection module: used to obtain the license plate number of the vehicle, and determine whether to use the first mode or the second mode based on the license plate number. The first mode is to only take pictures without recognition, and the second mode is to take pictures and recognize; Control module: used to obtain a real-time vehicle image set entering the intermediate control area based on the second mode, process the real-time vehicle image set based on the license plate number to obtain a target vehicle image set, input the target vehicle image set into a pre-built leakage monitoring model, and obtain a monitoring category, which includes leakage warning and non-warning. The intermediate control area is smaller than the edge control area; Position selection module: used to obtain the regional image corresponding to each position when entering the core control area, determine the position status based on the regional image, and select the target position according to the position status. The core control area is smaller than the intermediate control area.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, it implements the communication control method based on artificial intelligence as described in any one of claims 1 to 10.

Images