Intelligent building monitoring alarm positioning method and system and storage medium thereof
By combining image recognition and pressure detection data in the building for processing, the problem of insufficient number of surveillance cameras is solved, global monitoring and automatic risk judgment are achieved, and safety is improved.
Patent Information
- Application Number
- CN202411568873.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-05
- Publication Date
- 2025-02-21
AI Technical Summary
In the building, due to the insufficient number of surveillance cameras, it is difficult to achieve global monitoring and is easily used by people with bad intentions to hide props.
By acquiring image recognition data, pressure detection data and time data, and combining these data for processing, the item change data, pressure change data and item estimate data are determined, so as to monitor and item position estimate the parts outside the camera recognition range.
Global monitoring is achieved without adding a camera, which improves security and can automatically judge the dangers of people in the building, reminding users to protect buildings and personnel.
Smart Images
Figure CN119495154A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent security, and in particular to an intelligent building monitoring alarm positioning method, system and storage medium thereof. Background Art
[0002] Inside a building, due to structural problems, there are many load-bearing walls or decorative walls, which makes it difficult for a small number of surveillance cameras to monitor the overall situation. There are always some corners that are difficult to monitor, and these places are easily used by people with bad intentions to hide props. Summary of the invention
[0003] In order to improve the problem that a small number of surveillance cameras cannot monitor the overall situation, the present application provides a smart building monitoring alarm positioning method, system and storage medium thereof.
[0004] The present application provides a method for locating an intelligent building monitoring alarm, which adopts the following technical solution: A method for locating an intelligent building monitoring alarm, comprising: Obtain image recognition data, obtain pressure detection data, and obtain time data; Determine item change data using the image recognition data and the time data; Determine pressure change data by using the time data and the pressure detection data; Determine item estimation data through the item change data, the pressure change data and the time data; The global monitoring data is determined and outputted through the item estimation data and the image recognition data.
[0005] By adopting the above technical solution, through pressure detection, it is possible to monitor the part outside the recognition range of the camera's image recognition data, and at the same time estimate where the object is placed and display it to the user for viewing, so that global monitoring can be achieved without adding additional cameras and a small amount of additional cost, thereby greatly improving safety.
[0006] Optional, including: Determining the number of people by using the image recognition data and a preset recognition threshold; Determining personnel pressure state data according to the number of people, the pressure change data and a preset silent pressure threshold; Determine the action data of the person through the image recognition data and a preset action deliberateness threshold; Determine the danger data through the personnel stress state data, the personnel action data and a preset comprehensive judgment threshold; An alarm signal is determined and outputted based on the danger data and a preset danger threshold.
[0007] By adopting the above technical solution, the dangerousness of people moving in the building can be automatically judged. Once it is judged to be malicious, the user will be reminded, thereby protecting the building and the people in the building. It is often used in occasions such as buildings under construction or hotels.
[0008] Optional, including: Determining contact time data by using the image recognition data, a preset contact time threshold and the time data; Determining contact pressure acceleration data through the contact time data and the pressure change data; Determining the risk level of the shoe material by using the contact pressure acceleration data and a preset material flexibility threshold; The risk level of the shoe material is included in the calculation of the risk data.
[0009] By adopting the above technical solution, the material used for the shoe can be automatically identified through pressure changes, and the wearer's malicious intent can be judged by the material, making the judgment more accurate and reducing the probability of skilled behavior that is difficult to judge as malicious through actions.
[0010] Optional, including: Determining estimated contact time data using the image recognition data, a preset recognition accuracy threshold and the time data; Determine the sudden deceleration speed data by using the recognition accuracy threshold, the estimated contact time data and the contact time data; The person's action data is determined by the image recognition data, the time data, the sudden deceleration speed data and the action deliberateness threshold.
[0011] By adopting the above technical solution, it is possible to automatically determine whether a person suddenly reduces the speed of his feet falling when about to touch the ground, thereby avoiding making any sound, thereby concealing the camera's image recognition, making it more accurate to determine whether a person's actions are deliberate.
[0012] Optional, including: Determine clothing change data by using the image recognition data and the time data; Determining human body contour data through the clothing change data, the image recognition data and a preset human body contour threshold; Determining estimated human body mass data by using the human body contour data and a preset body mass threshold; Determine the mass data of the carried items by using the pressure detection data and the estimated human body mass data; Determine the shape data of the carried item by using the human body contour data and the clothing change data; Determine the dangerousness of the carried item by using the shape data of the carried item, the mass data of the carried item and a preset dangerous item type threshold; The risk level of the carried items and the risk level of the shoe materials are included in the calculation of the risk data.
[0013] By adopting the above technical solution, the body shape and mass of the human body can be automatically judged, thereby judging the mass and shape of the carried items, greatly improving the accuracy of the judgment of dangerous data.
[0014] Optional, including: Determining pressure data by using a plurality of the pressure detection data, a preset force conduction threshold and preset pressure position data; Determine the impact influence data by using the image recognition data, the time data, the recognition accuracy threshold, the sudden deceleration speed data, and a preset impact influence threshold; determining a pressure curve using the pressure data and the time data; Determine maximum pressure data through the pressure curve; Determine estimated mass data by using the maximum pressure data and the impact influence data; The estimated mass data replaces the pressure detection data and is used to participate in the calculation of the estimated human body mass data.
[0015] By adopting the above technical solution, the influence of the impact force brought by the foot landing on the pressure is reduced, so that the estimation of the human body mass is more accurate.
[0016] The present application provides an intelligent building monitoring alarm positioning system, which adopts the following technical solutions: An intelligent building monitoring alarm positioning system, comprising: An image recognition module, used for capturing and outputting the image recognition data; A pressure detection module, used to detect and obtain the pressure detection data and output it; The data processing module is used to receive the image recognition data and the pressure detection data, perform data processing and calculation, obtain the corresponding global monitoring data and the alarm signal, and output them.
[0017] By adopting the above technical solution, the monitoring range of the image recognition module is compensated by the pressure detection module.
[0018] The present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program that can be loaded by a processor and executes an intelligent building monitoring alarm locating method.
[0019] By adopting the above technical solution, the computer program is stored in a computer-readable storage medium.
[0020] In summary, the present application includes at least one of the following beneficial technical effects: 1. Global monitoring can be achieved without adding additional cameras and with a small additional cost, greatly improving security.
[0021] 2. Automatically determine the dangerousness of people in the building. Once it is determined to be malicious, it will alert the user, thereby protecting the building and the people in the building. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] Figure 1 It is a flow chart of an intelligent building monitoring alarm locating method in an embodiment of the present application.
[0023] Figure 2 It is a flowchart diagram of step S2 to step S24.
[0024] Figure 3 It is a flowchart diagram of step S3 to step S33.
[0025] Figure 4 It is a flowchart diagram of step S4 to step S42.
[0026] Figure 5 It is a flowchart diagram of step S5 to step S56.
[0027] Figure 6 It is a flowchart diagram of step S6 to step S65.
[0028] Figure 7 It is a module schematic diagram of an intelligent building monitoring alarm positioning system in an embodiment of the present application.
[0029] Explanation of the accompanying drawings: 1. Image recognition module; 2. Pressure detection module; 3. Data processing module. DETAILED DESCRIPTION
[0030] The following is combined with Figure 1-7 This application is described in further detail.
[0031] The present application embodiment discloses a method for locating an intelligent building monitoring alarm. Figure 1 ,The intelligent building monitoring alarm positioning method includes the following steps: S1, obtain image recognition data, obtain pressure detection data, and obtain time data; S11, determining item change data through image recognition data and time data; S12, determining pressure change data through time data and pressure detection data; S13, determining item estimated data through item change data, pressure change data and time data; S14. Determine and output global monitoring data through item estimation data and image recognition data.
[0032] In detail: the image recognition data is obtained by taking pictures with a camera, and the pressure detection data is obtained by, for example, a pressure sensor. Generally speaking, the pressure detection data is the mass of the person stepping on it, which is generally the total mass of the entire person including the items carried. The image recognition data and time data represent the video of the person's actions according to the time axis in the image. If the time data is 1 minute, if the image recognition data is a person carrying a bag, if the time data is 2 minutes or later, if the image recognition data is a person not carrying a bag, then the item change data can be obtained as one bag is missing, and if the time data is If the time data is 1min and the pressure detection data is 0N, if the time data is 2min and the pressure detection data is 10000N, the pressure change data is 10000N, and the position where the pressure increases is outside the recognition range of the image recognition data, and the person in the image recognition has passed this pressure position, it means that the person has dropped the bag at this position. If this pressure detection data is position No. 1, the item estimation data can be obtained as (position No. 1, 10000N), and it is outside the range of the image recognition data, it can be combined with the image recognition data to obtain the global monitoring data, thereby realizing global monitoring.
[0033] Reference Figure 2 , further comprising the following steps: S2. Determine the number of people through image recognition data and a preset recognition threshold; S21, determining personnel pressure status data based on the number of people, pressure change data and a preset silent pressure threshold; S22, determining the action data of the person through the image recognition data and a preset action deliberateness threshold; S23, determining the danger data through personnel stress status data, personnel action data and a preset comprehensive judgment threshold; S24. Determine an alarm signal based on the danger data and a preset danger threshold and output it.
[0034] In detail: the recognition threshold is the data for identifying moving objects, such as people, dogs, carts, etc., and then the people in the image recognition data are recognized and the number of people is obtained. If the pressure change data is 10000N / s, that is, when a person steps on and walks away, the ground is subjected to a pressure change of 10000N per second, that is, it first increases by 10000N per second and then decreases by 10000N per second. Assuming the silent pressure threshold is 5000N / s, if the number of people is 3, the pressure change data of each person is 10000 / 3=3333N / s, which is less than 5000N / s, indicating that the person walking at this time is walking in a silent state, that is, the footsteps are light, which is the personnel pressure. The force state data, the closer to the silent state, the more dangerous it is; the deliberate action threshold is the threshold data of unnatural movements when a person walks. For example, when a person walks naturally, the arms, shoulders, and hips will swing and move naturally as they walk. For example, sneaky actions are unnatural walking. The image recognition data is identified to determine how much of the person's behavior is unnatural at this time. This is the personnel action data. The more unnatural it is, the more dangerous it is. Finally, the personnel action data and the personnel pressure state data are combined with the comprehensive judgment threshold to calculate the danger data. If the danger data exceeds the danger threshold, it means that the person in the building is likely to have malicious intentions at this time, and the corresponding alarm signal is output to the security room.
[0035] Reference Figure 3 , the calculation of hazard data also includes the following steps: S3, determining contact time data through image recognition data, a preset contact time threshold and time data; S31, determining contact pressure acceleration data through contact time data and pressure change data; S32, determining the risk level of the shoe material through contact pressure acceleration data and a preset material flexibility threshold; S33, including the risk level of the shoe material in the calculation of the risk data.
[0036] Detailed: When the image recognition data determines that the person's feet touch the ground, the time data of this time point is recorded, and the time data of a period of time after this time point is used as the contact time data according to the contact time threshold. For example, when the time data touches the ground at 1 minute, the contact time threshold is set to 0.5 seconds, and the contact time data is the time data between 1 minute and 1 minute 0.5 seconds. Then, combined with the pressure change data, the pressure change data is 1000N. In order to facilitate the calculation of the distance, uniform acceleration is used, which means that the pressure detection data is from 0 to 1000N within 0.5 seconds, and the speed of pressure increase can be calculated to be 1000 / 0.5. 2 =4000N / s 2The material flexibility threshold is the acceleration generated when various flexible materials are impacted. After the material is impacted, there is usually a certain amount of buffering, and the better the buffering performance of the material, the smaller the acceleration it generates. Acceleration is not speed. Speed is related to the magnitude of the impact force. Acceleration is related to the buffering performance of the material to a certain extent. By comparing the acceleration with the material flexibility threshold, the material used for the shoes can be found, and the convenience of this material for malicious use is the shoe material hazard. For example, the shoe material with a characteristic foam sole material that allows silent walking has a higher hazard, while the shoe material with a metal sole that is difficult to walk silently has a very low hazard. Then the shoe material hazard is comprehensively calculated with the personnel pressure state data, personnel action data, and comprehensive judgment threshold calculated in the previous step.
[0037] Reference Figure 4 , including the following steps: S4, determining estimated contact time data through image recognition data, a preset recognition accuracy threshold and time data; S41, determining the sudden deceleration speed data by identifying the accuracy threshold, the estimated contact time data and the contact time data; S42. Determine the action data of the person through the image recognition data, time data, sudden deceleration speed data and action deliberateness threshold.
[0038] In detail: when the image recognition data is used to determine that a person is about to step on the ground, for example, the recognition accuracy threshold of the image recognition data is 1 cm, then when the sole is 1-2 cm away from the ground, the time data at this time is recorded as the estimated contact time data, and then when the pressure change data begins to change, the time data at this time is recorded again. If the estimated contact time data is 1 min, if the contact time data is 1 min 2 s, then the deceleration time can be obtained as 1 min 2 s-1 min = 2 s, and the sudden deceleration speed data can be calculated as [1 / 2 2 =0.25cm / s 2 ,2 / 2 2 =0.5cm / s 2 ], and then use this data in the calculation of personnel action data to make the calculation of personnel action data more accurate.
[0039] Reference Figure 5 , including the following steps: S5, determining clothing change data through image recognition data and time data; S51, determining human body contour data through clothing change data, image recognition data and a preset human body contour threshold; S52, determining estimated human body mass data based on the human body contour data and a preset body mass threshold; S53, determining the mass data of the carried items through the pressure detection data and the estimated human body mass data; S54, determining the shape data of the carried item through the human body contour data and the clothing change data; S55, determining the danger level of the carried item based on the shape data of the carried item, the mass data of the carried item, and a preset threshold value of the dangerous item type; S56. The danger level of the carried items and the danger level of the shoe materials are included in the calculation of the danger data.
[0040] In detail: the image recognition data of a period of time is recognized, and the changes of the clothes worn by the person during this period of time are recorded, such as the shaking with walking, the swaying caused by the wind or the wind caused by walking. This is the clothing change data. Combined with the human body contour threshold, the human body contour data can be obtained. For example, a person wears a loose sweater and a raised bracelet inside the sweater. As he walks, the sweater will be sunken, wrinkled or bulged in different places. At this time, the most sunken position of the sweater can be obtained, which is the position closest to the human skin at this time. After a long period of time, the position of most of the human skin can be obtained. Finally, combined with the human body contour threshold, for example, if the curve of most of the forearm has been obtained, the curve of the remaining forearm can be estimated, and the interference of the raised bracelet can be directly eliminated. This is the role of the human body contour threshold, so as to obtain the final human body contour data; the body mass threshold is the corresponding relationship between body shape and weight. For example, a person with a thin figure of 175cm has a corresponding weight of 70kg. However, if a person exercises and has more muscles, his weight may reach 80kg even if he is thin. Therefore, the obtained human body mass data is an interval (figure, [weight with less muscle, weight with more muscle]); then, the weight of all items carried by the person can be obtained by subtracting the estimated human body mass data from the pressure detection data, such as clothes, mobile phones, bags, etc., which is the mass data of the carried items. At the same time, the shape data of the items carried by the person can be obtained by excluding the previously estimated human body contour data in the image recognition data, which is the shape data of the carried items. Finally, combined with the mass data of the carried items, the danger of the item can be obtained by looking up the table in the dangerous goods type threshold, which is the danger degree of the carried items. Then, the danger degree of the carried items is combined with the calculation of the danger data in the previous step.
[0041] Reference Figure 6 , including the following steps: S6. Determine pressure data through a plurality of pressure detection data, a preset force conduction threshold and preset pressure position data; S61, determining impact influence data through the image recognition data, the time data, the recognition accuracy threshold, the sudden decrease speed data, and a preset impact influence threshold; S62, determining a pressure curve through pressure data and time data; S63, determining maximum pressure data through the pressure curve; S64, determining estimated quality data through maximum pressure data and impact influence data; S65. Substituting the pressure detection data with the estimated mass data to participate in the calculation of the estimated human body mass data.
[0042] Detailed: When a person steps on the ground, it is usually difficult for him to step on the detection position exactly. He usually steps between two detection positions. If the pressure detection data of position 1 is 1000N, the pressure detection data of position 2 is 1000N, the pressure detection data of position 3 is 1000N, and the pressure detection data of position 4 is 1000N, it means that the person steps on the position in the middle of the four detection positions 1, 2, 3, and 4. If position 1, 2, 3, and 4 are respectively at the four corners of a square with a side length of 1m, if the force conduction threshold is 10% / m, it can be calculated that the force in the middle should be This is the pressure data, and position 1, position 2, position 3, position 4 and the four corners of a square with a side length of 1m are the pressure position data; then the speed of the person's foot when entering the recognition accuracy threshold can be calculated through the image recognition data and time data, and the speed of the shoe just touching the ground can be estimated through the sudden speed reduction data, and then the impact influence data can be obtained by combining the impact influence threshold, and the timing of the pressure data with the time data, that is, from the time the person's foot steps on the ground until the foot leaves the ground, the pressure data will change accordingly, and the pressure data and time data are used as xy axis data to form a pressure curve, and then the maximum pressure data is taken as the maximum pressure data, and then the impact influence data is excluded from the maximum pressure data. If the impact influence data is 0.1, if the maximum pressure data is 77000N, the estimated mass data is 77000*(1-0.1)=70000N, which is about 70kg, which is the estimated human mass data.
[0043] The present application embodiment discloses an intelligent building monitoring alarm positioning system. Figure 7The intelligent building monitoring alarm positioning system includes an image recognition module 1, a pressure detection module 2 and a data processing module 3. The image recognition module 1 can use multiple cameras, which are distributed in the passages of the building according to actual needs, and the image recognition data is obtained by shooting. The pressure detection module 2 can use multiple pressure sensors and components that conduct pressure, such as metal plates covered by carpets. The pressure sensors can be buried in the recognition range of the image recognition module 1 and outside the range according to needs, such as aisles within the recognition range, and pits outside the recognition range (trapped in the wall, with a very small area, and no separate camera will be installed for it), etc. A carpet that enhances the experience can be laid on the floor of the building, and then components for conducting pressure, such as metal plates, cement boards, tiles, etc., can be laid under the carpet, so that when a person steps on any position, the pressure will be transmitted to the pressure sensor, so that the pressure and position caused by the person are calculated to obtain pressure detection data.
[0044] The data processing module 3 includes a database and a processor. The database is used to store various threshold data, such as silent pressure threshold, action deliberateness threshold, comprehensive judgment threshold, material flexibility threshold, human body contour threshold, body mass threshold, dangerous goods type threshold, force conduction threshold, impact impact threshold and other threshold data. The processor is used to receive image recognition data and pressure detection data, and then call the corresponding threshold data from the database. After processing and calculating the data, the corresponding alarm signal is obtained and output. At the same time, the estimated data of objects outside the recognition range of the image recognition module 1 is combined with the image recognition data within the recognition range to form global monitoring data and display it to the user for viewing.
[0045] The processor may include a central processing unit such as a CPU or MPU or a host system built around a CPU or MPU, including hardware or software. After the meter has a processor, people can freely control the metering instrument through programming to make it run according to people's wishes. The processor can control local measurement transmission, remote measurement transmission, remote communication, etc. through internal protocols. Internal protocols refer to all protocols that achieve mutual communication or links within the same metering instrument or the same system, including: human-computer interaction protocols, software / hardware (interface) protocols, chip bus (C-Bus) protocols, internal bus (I-Bus) protocols, etc. Part or all of the protocols. With the development of integrated circuit technology, some protocols that belong to the external bus (E-Bus) protocol are also classified as internal protocols after the external bus (E-Bus) is integrated into the chip.
[0046] The embodiment of the present application discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program that can be loaded by a processor and executes a method for locating an intelligent building monitoring alarm.
[0047] Computer-readable storage media include, for example, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
[0048] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.
Claims
1. A method for locating an intelligent building monitoring alarm, characterized in that: include: Obtain image recognition data, obtain pressure detection data, and obtain time data; Determine item change data using the image recognition data and the time data; Determine pressure change data by using the time data and the pressure detection data; Determine item estimation data through the item change data, the pressure change data and the time data; The global monitoring data is determined and outputted through the item estimation data and the image recognition data.
2. The intelligent building monitoring alarm positioning method according to claim 1 is characterized in that: include: Determining the number of people by using the image recognition data and a preset recognition threshold; Determining personnel pressure state data according to the number of people, the pressure change data and a preset silent pressure threshold; Determine the action data of the person through the image recognition data and a preset action deliberateness threshold; Determine the danger data through the personnel stress state data, the personnel action data and a preset comprehensive judgment threshold; An alarm signal is determined and outputted based on the danger data and a preset danger threshold.
3. The intelligent building monitoring alarm positioning method according to claim 2 is characterized in that: include: Determining contact time data by using the image recognition data, a preset contact time threshold and the time data; Determining contact pressure acceleration data through the contact time data and the pressure change data; Determining the risk level of the shoe material by using the contact pressure acceleration data and a preset material flexibility threshold; The risk level of the shoe material is included in the calculation of the risk data.
4. The intelligent building monitoring alarm positioning method according to claim 3 is characterized in that: include: Determining estimated contact time data using the image recognition data, a preset recognition accuracy threshold and the time data; Determine the sudden deceleration speed data by using the recognition accuracy threshold, the estimated contact time data and the contact time data; The person's action data is determined by the image recognition data, the time data, the sudden deceleration speed data and the action deliberateness threshold.
5. The intelligent building monitoring alarm positioning method according to claim 4 is characterized in that: include: Determine clothing change data by using the image recognition data and the time data; Determining human body contour data through the clothing change data, the image recognition data and a preset human body contour threshold; Determining estimated human body mass data by using the human body contour data and a preset body mass threshold; Determine the mass data of the carried items by using the pressure detection data and the estimated human body mass data; Determine the shape data of the carried item by using the human body contour data and the clothing change data; Determine the dangerousness of the carried item by using the shape data of the carried item, the mass data of the carried item and a preset dangerous item type threshold; The risk level of the carried items and the risk level of the shoe materials are included in the calculation of the risk data.
6. The intelligent building monitoring alarm positioning method according to claim 5, characterized in that: include: Determining pressure data by using a plurality of the pressure detection data, a preset force conduction threshold and preset pressure position data; Determine the impact influence data by using the image recognition data, the time data, the recognition accuracy threshold, the sudden deceleration speed data, and a preset impact influence threshold; determining a pressure curve using the pressure data and the time data; Determine maximum pressure data through the pressure curve; Determine estimated mass data by using the maximum pressure data and the impact influence data; The estimated mass data replaces the pressure detection data and is used to participate in the calculation of the estimated human body mass data.
7. An intelligent building monitoring alarm positioning system, using the intelligent building monitoring alarm positioning method according to claim 1, characterized in that: include: An image recognition module (1) is used to capture and obtain the image recognition data and output it; A pressure detection module (2), used for detecting and obtaining the pressure detection data and outputting it; The data processing module (3) is used to receive the image recognition data and the pressure detection data, perform data processing and calculation, obtain the corresponding global monitoring data and the alarm signal, and output them.
8. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and executes the method for locating an intelligent building monitoring alarm according to any one of claims 1 to 6.