Campus monitoring system applied to smart campus

Abstract

The invention discloses a campus monitoring system applied to a smart campus, and relates to the technical field of campus monitoring, and the campus monitoring system comprises a position acquisition module, a position analysis module, a campus monitoring module, a grade division module, an early warning terminal and a database. Whether the students are in the school is judged according to the real-time positions of the students, if not, whether the students are safe is analyzed according to the total number of the students in the activity area and traffic information, if yes, whether the activity area of the students is safe is analyzed, if not, early warning is conducted, and if yes, whether the behaviors of the students are abnormal is analyzed; and if the student is abnormal, whether the student is safe is analyzed according to the real-time position of the student, if the student is not safe, early warning is performed, and the danger levels of other students are re-divided, so that whether the off-school activity of the student is safe can be known, the danger of the student can be found in time, and the safety of the student is effectively guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of campus monitoring, and in particular to a campus monitoring system applied to a smart campus. Background Art

[0002] With the rapid development of society, the campus environment is becoming increasingly complex. The pressure students face in study and life is gradually increasing, and abnormal behaviors occur from time to time. Building a precise and efficient campus monitoring system has become an urgent need for the construction of smart campuses.

[0003] Existing technology, such as the application with publication number CN112132711A, discloses a campus monitoring system for smart campuses, including a campus monitoring module, a facial recognition module, a student evaluation module and a main controller. The campus monitoring module is connected to the cloud database and the Beidou positioning module. The Beidou positioning module is used to locate the participants in malicious incidents or disputes in real time. The campus monitoring module monitors the learning scene to stop malicious incidents or disputes in real time. The facial recognition module is used to identify and record the collected student facial information. The notification module reminds the teachers in the teaching department, and the preset module is used to set the number of times the student's face is collected. When the number of times the provocative student participates in a malicious incident or dispute reaches the preset number threshold, the wireless signal transceiver module sends a signal to the mobile terminal of the student's parents.

[0004] In response to the above solution, the applicant of the present invention found that the above technology has at least the following technical problems: 1. When analyzing student safety, the existing technology lacks analysis of whether students are safe outside of school. When students are outside of school, it is impossible to know whether the students' activities are safe.

[0005] 2. When students do not have any malicious incidents or disputes, danger may still occur. The existing technology analyzes whether students are safe based on real-time monitoring of whether students have any malicious incidents or disputes. It cannot effectively guarantee the safety of students and cannot detect students' dangers in time. Summary of the Invention

[0006] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide a campus monitoring system applied to smart campuses.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solution: The present invention provides a campus monitoring system applied to a smart campus, including the following modules: a location acquisition module, a location analysis module, a campus monitoring module, a level classification module, an early warning terminal and a database.

[0008] The location acquisition module is used to equip students with smart terminals integrated with UWB positioning and obtain the students' locations in real time.

[0009] The location analysis module is used to compare the real-time location of students with the coverage of the school to determine whether the students are on campus.

[0010] The campus guardianship module includes an off-campus guardianship unit and an on-campus guardianship unit.

[0011] The off-campus guardianship unit is used to analyze whether the potential safety hazards in the activity area of students are high when the students are having activities off campus. When the potential safety hazards in the activity area of students are high, it means the students are not safe, and at this time, a warning is issued.

[0012] The on-campus guardianship unit is used to analyze whether the activity area of students is safe when the students are having activities on campus. When the activity area of students is not safe, a warning is issued. If it is safe, it analyzes whether the behavior of students is abnormal. If it is abnormal, it analyzes whether the students are safe. If they are not safe, a warning is issued.

[0013] The grade division module is used to identify the behaviors of other students in the activity area and re-divide the risk levels of other students when the activity area of students on campus is not safe.

[0014] The warning terminal is used to issue a warning when the students are not safe or the activity area is not safe.

[0015] The database is used to store various danger signs, various dangerous postures, various building areas, various dangerous destinations, various classroom areas, historical activity tracks in the activity area, off-campus activity locations of each student, various dangerous behaviors, and information of each student.

[0016] The beneficial effects of the present invention are as follows: 1. The present invention provides a campus guardianship system applied to a smart campus. The real-time location of students is obtained through an intelligent terminal. Whether the students are on campus is judged according to the real-time location of the students. If not on campus, it analyzes whether the students are safe according to the total number of students in the activity area and various traffic information. If on campus, it analyzes whether the activity area of the students is safe. If not safe, a warning is issued. If safe, it analyzes whether the behavior of the students is abnormal. If abnormal, it analyzes whether the students are safe according to the real-time location of the students. If not safe, a warning is issued. At the same time, the risk levels of other students are re-divided, so as to know whether the off-campus activities of students are safe and timely discover the dangers of students, effectively ensuring the safety of students.

[0017] 2. The present invention judges whether the students are on campus according to the real-time location of the students. If not on campus, it obtains the activity area of the students and analyzes whether the potential safety hazards in the activity area of the students are high according to various traffic information in the activity area of the students and the total number of students in the activity area. When the potential safety hazards in the activity area are high, it means the students are not safe, and at this time, a warning is issued. When the students are off campus, it can know whether the activities of the students are safe.

[0018] 3. The present invention analyzes whether a student is engaged in activities on a plane based on the student's real-time position. If the student is engaged in activities on a plane, it identifies other students within the activity area, various identifiers within the activity area, and the student density, classifies the risk levels of other students, and analyzes whether the student's activity area is safe. If the student is engaged in activities at a height, it analyzes whether the student's activity area is safe based on the risk levels of other students, the student's real-time position, and the real-time position height. When the student's activity area is unsafe, it issues a warning. When the student's activity area is safe, it analyzes whether the student's behavior is abnormal. If it is abnormal, it determines whether the student is safe based on the change of the student's real-time position. If the student is unsafe, it issues a warning, thus promptly detecting the student's danger and effectively ensuring the safety of the student. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Figure 1 It is a schematic diagram of the system structure connection of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0022] Please refer to Figure 1 As shown, the present invention provides a campus guardianship system applied to a smart campus, including the following modules: a position acquisition module, a position analysis module, a campus guardianship module, a level classification module, a warning terminal, and a database.

[0023] The position acquisition module is connected to the position analysis module, the position analysis module is connected to the campus guardianship module, the campus guardianship module is connected to the warning terminal, and the database is connected to the position analysis module, the campus guardianship module, and the level classification module.

[0024] The position acquisition module is used to equip students with intelligent terminals integrated with UWB positioning and collect the positions of students in real time.

[0025] It should be noted that only when students are at school, the intelligent terminals are used to collect the positions of students in real time.

[0026] The position analysis module is used to compare the real-time position of the student with the coverage of the school to determine whether the student is on campus.

[0027] It should be noted that the coverage of the school is obtained from the database, and the real-time position of the student is compared with the coverage of the school. If the real-time position of the student is within the coverage of the school, it means that the student is on campus. If the real-time position of the student is not within the school, it means that the student is off campus.

[0028] The campus guardianship module includes an off-campus guardianship unit and a behavior guardianship unit.

[0029] The off-campus guardianship unit is used to analyze whether the potential safety hazards in the activity area of the student are high when the student is off campus. When the potential safety hazards in the activity area of the student are high, it means that the student is not safe, and an alarm is issued at this time.

[0030] In a specific embodiment, the off-campus guardianship unit has the following specific process: when the student is off campus, the off-campus activity locations of each student are obtained from the database, and the off-campus activity locations of each student are clustered to obtain each off-campus activity area. The real-time position of the student is compared with each off-campus activity area. If the real-time position of the student is not within each off-campus activity area, it means that the potential safety hazards in the activity area of the student are high. At this time, an alarm is issued to the student through the intelligent terminal and the student's counselor is notified. If the real-time position of the student is within a certain off-campus activity area, this off-campus activity area is called a marked area, the number of students in the marked area is counted, and according to the number of students in the marked area and the traffic information of the marked area, it is analyzed whether the potential safety hazards in the marked area are high. If the potential safety hazards in the marked area are high, an alarm is issued to the student through the intelligent terminal.

[0031] It should be noted that this system is connected to the traffic management system to obtain each traffic information from the traffic management system. Each traffic information includes traffic flow, pedestrian flow, vehicle speed, etc.

[0032] In the above, the process of analyzing whether the potential safety hazards in the marked area are high is as follows: the number of students in the marked area and the traffic information in the marked area are normalized and input into the safety analysis model, and the danger index of the marked area is output. If the danger index of the marked area is 1, it means that the potential safety hazards in the marked area are high. If the danger index of the marked area is 0, it means that the potential safety hazards in the marked area are low.

[0033] It should be noted that the safety analysis model: In the formula, A a represents the a-th traffic information, A a ' represents the standard information of the a-th traffic information, and α represents the traffic accident index. Represents a preset danger coefficient threshold, Q represents the total number of students in the marked area, Q≥1, a represents the number of each traffic information, e represents the natural constant, a = 1, 2, 3,..., b, b represents the total number of traffic information, and both a and b are positive integers.

[0034] It should also be noted that the standard information of each traffic information is the maximum value used to analyze whether each traffic information is abnormal, obtained from the traffic management system. When a certain traffic information is higher than the standard information of this traffic information, it means that this traffic information is abnormal. If a certain traffic information is lower than the standard information of this traffic information, it means that this traffic information is normal.

[0035] Among them, the preset danger coefficient threshold is the critical value used to judge whether the potential safety hazard in the marked area is high, set by relevant staff. When the danger coefficient is higher than the preset danger coefficient threshold, it means that the potential safety hazard in the marked area is high. When the danger coefficient is lower than the preset danger coefficient threshold, it means that the potential safety hazard in the marked area is low.

[0036] The on-campus guardianship unit is used to analyze whether the activity area of students is safe when students are having activities on campus. When the activity area of students is not safe, it gives an early warning. If it is safe, it analyzes whether the behavior of students is abnormal. If it is abnormal, it analyzes whether the students are safe. If they are not safe, it gives an early warning.

[0037] In a specific embodiment, the on-campus guardianship unit has the following specific process: When students are having activities on campus, obtain the real-time position height of the students and compare it with the preset height threshold. If the real-time position height of the students is less than or equal to the preset height threshold, it means that the students are having activities on the plane. If the real-time position height of the students is higher than the preset height threshold, it means that the students are having activities at a high place.

[0038] It should be noted that the preset height threshold is the threshold used to evaluate whether students are having activities on the plane, set by relevant staff.

[0039] Cluster the positions of the students at N activity times to obtain the activity area of the students, obtain the surveillance video of the activity area, and analyze whether the activity area of the students is safe according to the surveillance video of the activity area and the positions of the students at N activity times. If it is not safe, give an early warning prompt to the students through the intelligent terminal. If it is safe, analyze whether the behavior of the students is abnormal. If it is abnormal, analyze whether the students are safe. If they are not safe, the intelligent terminal gives an early warning prompt to the students and notifies the security personnel.

[0040] In the above, when students are engaged in activities on a plane, analyze whether the activity area of the students is safe. The specific process is as follows: Obtain the surveillance video of the activity area and input it into the deep learning model to identify other students in the activity area, the total number of students in the activity area, each identifier in the activity area, and the student density. Call other students each marked student, compare each identifier in the activity area with each dangerous identifier in the data, and call each identifier in the activity area that is the same as the dangerous identifier in the database each marked identifier. Count the total number of marked identifiers in the activity area, and at the same time divide the danger levels of each marked student, and count the total number of marked students in each danger level. Determine the safety coefficient of the activity area according to the total number of marked students in each danger level, the total number of students in the activity area, the total number of marked identifiers in the activity area, and the student density. Compare the safety coefficient of the activity area with the preset first safety coefficient threshold. If the safety coefficient of the activity area is lower than the preset first safety coefficient threshold, it means that the activity area of the students is not safe. If the safety coefficient of the activity area is higher than the preset first safety coefficient threshold, it means that the activity area of the students is safe.

[0041] It should be noted that the identifiers include rivers, fences, lawns, street lamps, and various warning and indication identifiers, etc. Among them, the various warning and indication identifiers include "Pay attention to safety" identifiers, "No passage" identifiers, "Be careful of slipping" identifiers, speed limit identifiers, etc.

[0042] Among them, the various dangerous identifiers include rivers, "Pay attention to safety" identifiers, "Be careful of slipping" identifiers, etc.

[0043] It should also be noted that normalize the total number of marked students in each danger level, the total number of students in the activity area, the total number of marked identifiers in the activity area, and the student density, then In the formula B c represents the total number of marked students in danger level c, B represents the total number of students in the activity area, D represents the student density, E represents the total number of marked identifiers in the activity area, w c represents the weight coefficient of danger level c, and c represents each danger level. When c = 1, it represents the first-level danger. When c = 2, it represents the second-level danger. When c = 3, it represents the third-level danger. represents the safety coefficient of the activity area.

[0044] In the above, the preset first safety coefficient threshold is the minimum value used to judge whether the activity area is safe and is set by relevant staff.

[0045] Among the above, the risk levels of each marked student are classified as follows: Obtain all the information of each marked student from the database, determine the risk coefficient of each marked student according to all the information of each marked student, compare the risk coefficient of each marked student with the first risk coefficient and the second risk coefficient. If the risk coefficient of a certain marked student is less than the first risk coefficient, it means that the risk level of this marked student is a first-level risk. If the risk coefficient of a certain marked student is not less than the first risk coefficient and not higher than the second risk coefficient, it means that the risk level of this marked student is a second-level risk. If the risk coefficient of a certain marked student is higher than the second risk coefficient, it means that the risk level of this marked student is a third-level risk. Use this method to classify the risk levels of each marked student.

[0046] It should be noted that all the information of each marked student includes the number of disciplinary actions, the number of notifications, and the number of interviews, etc.

[0047] It should also be noted that if all the information of each marked student is normalized, then In the formula, F d represents the f-th piece of information of the d-th marked student, γ d represents the risk coefficient of the d-th marked student, f represents the number of each piece of information, f = 1, 2, 3,..., g, and g represents the total number of all pieces of information.

[0048] Among them, the first risk coefficient and the second risk coefficient are critical values used to classify the risk levels of each marked student, and are set by relevant staff.

[0049] Among the above, when a student is carrying out activities at a high place, analyze whether the activity area of the student is safe. The specific process is as follows: Obtain the total number of marked students in each risk level, the total number of students in the activity area, and the student density in the activity area. Obtain each building area on campus from the database, and compare the activity area with each building area on campus. If the activity area is within a certain building area on campus, it means that the student is carrying out activities in this building, and this building is called the marked building. Calculate the distance between the positions of the student at N activity times and the edge of the marked building area. According to the position height of the student at N activity times, the distance between the positions of the student at N activity times and the edge of the marked building area, the total number of students in each risk level in the activity area, the total number of students in the activity area, and the student density, determine the safety coefficient of the activity area. Compare the safety coefficient of the activity area with the preset second safety coefficient threshold. If the safety coefficient of the activity area is higher than the preset second safety coefficient threshold, it means that the activity area of the student is safe. If the safety coefficient of the activity area is lower than the preset second safety coefficient threshold, it means that the activity area of the student is unsafe.

[0050] It should be noted that the position height at N activity moments of the student, the distance between the position at N activity moments and the edge of the marked building area, the total number of students in each danger level within the activity area, the total number of students in the activity area, and the student density are normalized, then In the formula, G h represents the position height at the h-th activity moment, and H h represents the distance between the position at the h-th activity moment and the edge of the marked building area, β represents the safety coefficient of the activity area, h represents the number of each activity moment, h = 1, 2, 3,..., N, N represents the total number of activity moments, and both h and N are positive integers.

[0051] It should also be noted that the preset second safety coefficient threshold is the minimum value used to judge whether the activity area is safe and is set by relevant staff.

[0052] In the above, the specific process of analyzing whether the student's behavior is abnormal is as follows: Obtain the positions of the student at N activity moments, calculate the average distance between the positions at each activity moment, compare the average distance between the positions at each activity moment with the preset distance threshold. If the average distance between the positions at each activity moment is less than the preset distance threshold, it means the student's behavior is abnormal. If the average distance between each activity moment is greater than the preset distance threshold, input the positions at each activity moment into the deep learning model, output the student's current activity trajectory and action destination, and analyze whether the student's current activity trajectory is the same as the historical activity trajectory. If the student's current activity trajectory is the same as the historical activity trajectory, it means the student's behavior is normal. If the student's current activity trajectory is different from the historical activity trajectory, it means the student's behavior is abnormal.

[0053] It should be noted that the preset distance threshold is the critical value used to judge whether the student's position has changed significantly and is set by relevant staff. When the average distance between the positions at each activity moment is less than the preset distance threshold, it means the student's position has not changed significantly. When the average distance between the positions at each activity moment is greater than the preset distance threshold, it means the student's position has changed significantly.

[0054] Among the above, the process of analyzing whether the student's current activity trajectory is the same as the historical activity trajectory is as follows: Obtain the positions of each point on each historical activity trajectory of the student in the activity area from the database. According to the positions of each point on each historical activity trajectory of the student in the activity area and the positions of each point on the student's current activity trajectory, determine the degree of overlap between each historical activity trajectory and the current activity trajectory. Compare the degree of overlap between each historical activity trajectory and the current activity trajectory with a preset overlap threshold. If the degree of overlap between a certain historical activity trajectory and the current activity trajectory is higher than the preset overlap threshold, it means that the student's current activity trajectory is the same as the historical activity trajectory. If the degree of overlap between each historical activity trajectory and the current activity trajectory is less than the preset overlap threshold, it means that the student's current activity trajectory is different from the historical activity trajectory.

[0055] It should be noted that In the formula represents the position of the j-th point on the i-th historical activity trajectory, and (x j , y j , z j ) represents the position of the j-th point on the current activity trajectory. ε i represents the degree of overlap between the i-th historical activity trajectory and the current activity trajectory. j represents the serial number of each point on the activity trajectory, j = 1, 2, 3,..., k, where k represents the total number of points on the activity trajectory. i represents the serial number of each historical activity trajectory. j, k, and i are all positive integers.

[0056] It should also be noted that the preset overlap threshold is the minimum value used to evaluate whether the student's current activity trajectory is a certain historical activity trajectory. If the degree of overlap between the student's current activity trajectory and a certain historical activity trajectory is higher than the preset overlap threshold, it means that the student's current activity trajectory is that historical activity trajectory. If the degree of overlap between the student's current activity trajectory and a certain historical activity trajectory is less than the preset overlap threshold, it means that the student's current activity trajectory is not that historical activity trajectory.

[0057] Among the above, the process of analyzing whether the student is safe is as follows: When the average distance between the positions of the student at each activity moment is less than the preset distance threshold, obtain the activity video of the student, input the activity video into the deep learning model to output the posture of the student, and compare the posture of the student with each dangerous posture in the database. If the posture of the student is different from each dangerous posture in the database, it means that the student is safe. If the posture of the student is the same as a certain dangerous posture in the database, it means that the student is not safe. At this time, give an early warning and notify the security personnel to check.

[0058] It should be noted that the postures include standing, bending, supporting the wall, squatting, and relying, etc. Among them, the dangerous postures include bending, supporting the wall, and squatting, etc.

[0059] When the average distance between the positions of a student at each activity moment is greater than a preset distance threshold, compare the student's action destination with each dangerous destination in the database. If the student's behavior destination is different from each dangerous destination in the database, it means the student is safe. If the student's behavior destination is the same as a certain dangerous destination in the database, it means the student is unsafe. At this time, give an alarm and notify the security personnel to check.

[0060] It should be noted that the activity destinations include classrooms, canteens, riversides, construction sites, dormitory buildings, etc. Among them, the dangerous destinations include riversides and construction sites, etc.

[0061] The grade division module is used to identify the behaviors of other students in the activity area when the activity area where the student is located is unsafe, and re-divide the danger grades of other students.

[0062] It should be noted that other students in the activity area are called each marked student. Compare the behaviors of each marked student with each dangerous behavior in the database. If the behavior of a certain marked student is the same as a certain dangerous behavior in the database, the number of dangerous behaviors of the marked student is 1. If the behavior of a certain marked student is different from each dangerous behavior in the database, the number of dangerous behaviors of the marked student is 0. Obtain the number of dangerous behaviors of each marked student in this way, incorporate the number of dangerous behaviors of each marked student into the danger coefficient of each marked dangerous student, and update the danger coefficient of each marked student. Then In the formula, γ′ d represents the updated danger coefficient of the d-th marked student, M d represents the number of dangerous behaviors of the d-th marked student. According to the updated danger coefficients of each marked student, conduct danger grade division.

[0063] Among them, student behaviors include walking, running, waving, applauding, stamping feet, and clenching fists, etc. Among them, dangerous behaviors include clenching fists and stamping feet, etc.

[0064] The warning terminal is used to give an alarm when the student is unsafe or the activity area is unsafe.

[0065] The database is used to store each danger identifier, each dangerous posture, each building area, each dangerous destination, each classroom area, the historical activity tracks in the activity area, the off-campus activity points of each student, each dangerous behavior, and each student information.

[0066] In the embodiment of the present invention, the intelligent terminal is used to obtain the real-time location of students, and determine whether the students are on campus according to their real-time locations. If the students are not on campus, it is analyzed whether the students are safe based on the total number of students in the activity area and various traffic information. If the students are on campus, it is analyzed whether the activity area of the students is safe. If it is not safe, a warning is issued. If it is safe, it is analyzed whether the behavior of the students is abnormal. If it is abnormal, it is analyzed whether the students are safe according to their real-time locations. If it is not safe, a warning is issued. At the same time, the danger levels of other students are reclassified, so as to know whether the off-campus activities of the students are safe and timely discover the dangers of the students, effectively ensuring the safety of the students.

[0067] The above content is only an example and illustration of the concept of the present invention. Those skilled in the art of the present technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should fall within the protection scope of the present invention.

Claims

1. A campus guardianship system applied to a smart campus, characterized in that It includes the following modules: The location acquisition module is used to equip students with smart terminals integrated with UWB positioning and obtain the real-time locations of students; The location analysis module is used to compare the real-time locations of students with the coverage area of the school to determine whether the students are on campus; The campus guardianship module includes an off-campus guardianship unit and an on-campus guardianship unit: The off-campus guardianship unit is used to analyze whether the potential safety hazards in the activity area of the student are high when the student is off campus. When the potential safety hazards in the activity area of the student are high, it means the student is not safe, and at this time, a warning is issued; The on-campus guardianship unit is used to analyze whether the activity area of the student is safe when the student is on campus. When the activity area of the student is not safe, a warning is issued. If it is safe, it analyzes whether the behavior of the student is abnormal. If it is abnormal, it analyzes whether the student is safe. If not safe, a warning is issued; The risk level classification module is used to identify the behaviors of other students in the activity area when the activity area of the student on campus is not safe and reclassify the risk levels of other students; The warning terminal is used to issue a warning when the student is not safe or the activity area is not safe; The database is used to store various danger signs, various dangerous postures, various building areas, various dangerous destinations, various classroom areas, historical activity tracks in the activity area, off-campus activity locations of each student, various dangerous behaviors, and information of each student.

2. The campus guardianship system applied to a smart campus according to claim 1, characterized in that, The specific process of the off-campus guardianship unit is as follows: When the student is off campus, obtain the off-campus activity locations of each student from the database, cluster the off-campus activity locations of each student to obtain each off-campus activity area, compare the real-time location of the student with each off-campus activity area. If the real-time location of the student is not within each off-campus activity area, it means the potential safety hazards in the activity area of the student are high. At this time, a warning is issued to the student through the smart terminal and the student's counselor is notified. If the real-time location of the student is within a certain off-campus activity area, then this off-campus activity area is called the marked area, count the number of students in the marked area, and analyze whether the potential safety hazards in the marked area are high according to the number of students in the marked area and the traffic information of the marked area. If the potential safety hazards in the marked area are high, a warning is issued to the student through the smart terminal.

3. The campus guardianship system applied to the smart campus according to claim 2, characterized in that, The specific process of analyzing whether the potential safety hazards in the marked area are high is as follows: Normalize the number of students in the marked area and the traffic information in the marked area and input them into the safety analysis model to output the danger index of the marked area. If the danger index of the marked area is 1, it means the potential safety hazards in the marked area are high. If the danger index of the marked area is 0, it means the potential safety hazards in the marked area are low.

4. A campus guardianship system applied to a smart campus according to claim 1, characterized in that, The specific process of the on-campus guardianship unit is as follows: When the student is on campus, obtain the real-time location height of the student and compare it with the preset height threshold. If the real-time location height of the student is lower than the preset height threshold, it means the student is moving on the plane. If the real-time location height of the student is higher than the preset height threshold, it means the student is moving at a high place; Cluster the positions of the student at N activity times to obtain the activity area of the student, obtain the surveillance video of the activity area, and analyze whether the activity area of the student is safe based on the surveillance video of the activity area and the positions of the student at N activity times. If it is not safe, give a warning prompt to the student through the intelligent terminal. If it is safe, analyze whether the behavior of the student is abnormal. If it is abnormal, analyze whether the student is safe. If it is not safe, the intelligent terminal gives a warning prompt to the student and notifies the security personnel.

5. A campus guardianship system applied to a smart campus according to claim 4, characterized in that, When the student is active on the plane, analyze whether the activity area of the student is safe. The specific process is as follows: Obtain the surveillance video of the activity area and input it into the deep learning model to identify other students in the activity area, the total number of students in the activity area, each identifier in the activity area, and the student density. Call other students each marked student. Compare each identifier in the activity area with each dangerous identifier in the data. Call each identifier in the activity area that is the same as the dangerous identifier in the database each marked identifier. Count the total number of marked identifiers in the activity area. At the same time, divide the risk levels of each marked student and count the total number of marked students in each risk level. Determine the safety coefficient of the activity area based on the total number of marked students in each risk level, the total number of students in the activity area, the total number of marked identifiers in the activity area, and the student density. Compare the safety coefficient of the activity area with the preset first safety coefficient threshold. If the safety coefficient of the activity area is lower than the preset first safety coefficient threshold, it means that the activity area of the student is not safe. If the safety coefficient of the activity area is higher than the preset first safety coefficient threshold, it means that the activity area of the student is safe.

6. The campus guardianship system applied to a smart campus according to claim 5, characterized in that, The process of dividing the risk levels of each marked student is as follows: Obtain each piece of information of each marked student from the database, determine the risk coefficient of each marked student according to each piece of information of each marked student, and compare the risk coefficient of each marked student with the first risk coefficient and the second risk coefficient. If the risk coefficient of a certain marked student is less than the first risk coefficient, it means that the risk level of the marked student is the first-level risk. If the risk coefficient of a certain marked student is not less than the first risk coefficient and not higher than the second risk coefficient, it means that the risk level of the marked student is the second-level risk. If the risk coefficient of a certain marked student is higher than the second risk coefficient, it means that the risk level of the marked student is the third-level risk. Divide the risk levels of each marked student in this way.

7. The campus guardianship system applied to a smart campus according to claim 4, characterized in that, When the student is active at a height, analyze whether the activity area of the student is safe. The specific process is as follows: Obtain the total number of marked students in each danger level within the activity area, the total number of students in the activity area, and the student density. Obtain each building area on campus from the database, and compare the activity area with each building area on campus. If the activity area is within a certain building area on campus, it means that the students are conducting activities in that building, and this building is called the marked building. Calculate the distances between the positions of the students at N activity moments and the edge of the marked building area. Based on the position heights of the students at N activity moments, the distances between the positions of the N activity moments and the edge of the marked building area, the total number of students in each danger level within the activity area, the total number of students in the activity area, and the student density, determine the safety coefficient of the activity area. Compare the safety coefficient of the activity area with the preset second safety coefficient threshold. If the safety coefficient of the activity area is higher than the preset second safety coefficient threshold, it means that the activity area of the students is safe. If the safety coefficient of the activity area is lower than the preset second safety coefficient threshold, it means that the activity area of the students is unsafe.

8. A campus guardianship system applied to a smart campus according to claim 4, characterized in that, The specific process of analyzing whether the student's behavior is abnormal is as follows: Obtain the positions of the students at N activity moments, and calculate the average distance between the positions at each activity moment. Compare the average distance between the positions at each activity moment with the preset distance threshold. If the average distance between the positions at each activity moment is less than the preset distance threshold, it means that the student's behavior is abnormal. If the average distance between each activity moment is greater than the preset distance threshold, input the positions at each activity moment into the deep learning model, output the student's current activity trajectory and action destination, and analyze whether the student's current activity trajectory is the same as the historical activity trajectory. If the student's current activity trajectory is the same as the historical activity trajectory, it means that the student's behavior is normal. If the student's current activity trajectory is different from the historical activity trajectory, it means that the student's behavior is abnormal.

9. A campus guardianship system applied to a smart campus according to claim 8, characterized in that, The specific process of analyzing whether the student's current activity trajectory is the same as the historical activity trajectory is as follows: Obtain the positions of each point on each historical activity trajectory of the student within the activity area from the database. Based on the positions of each point on each historical activity trajectory of the student within the activity area and the positions of each point on the student's current activity trajectory, determine the coincidence degree between each historical activity trajectory and the current activity trajectory. Compare the coincidence degree between each historical activity trajectory and the current activity trajectory with the preset coincidence degree threshold. If the coincidence degree between a certain historical activity trajectory and the current activity trajectory is higher than the preset coincidence degree threshold, it means that the student's current activity trajectory is the same as the historical activity trajectory. If the coincidence degrees between each historical activity trajectory and the current activity trajectory are all less than the preset coincidence degree threshold, it means that the student's current activity trajectory is different from the historical activity trajectory.

10. A campus guardianship system applied to a smart campus according to claim 4, characterized in that, The specific process of analyzing whether the student is safe is as follows: When the average distance between the positions of the student at each activity moment is less than the preset distance threshold, obtain the activity video of the student, input the activity video into the deep learning model to output the posture of the student, compare the posture of the student with each dangerous posture in the database. If the posture of the student is different from each dangerous posture in the database, it means the student is safe. If the posture of the student is the same as a certain dangerous posture in the database, it means the student is not safe. At this time, give an early warning and notify the security personnel to check. When the average distance between the positions of the student at each activity moment is greater than the preset distance threshold, compare the destination of the student's action with each dangerous destination in the database. If the destination of the student's behavior is different from each dangerous destination in the database, it means the student is safe. If the destination of the student's behavior is the same as a certain dangerous destination in the database, it means the student is not safe. At this time, give an early warning and notify the security personnel to check.

Images