Automatic lighting adjustment system based on scene analysis

Abstract

The present invention relates to the field of big data analysis technology, and more specifically, to a lighting automatic adjustment system based on scene analysis. It includes a scene perception module, a central control module, a lighting adjustment module, and a storage and management module. The scene perception module includes multiple human body sensors, and the central control module includes a microprocessor. The scene perception module is used to collect environmental information of the area where the lamp is located, and the central control module processes the environmental information transmitted by the scene perception module. Through a long-short-term memory model, the number of triggering times and duration of the human body sensor when a person passes through an area are recorded, and whether it is triggered again within the interval time, so as to determine whether the person passes through the area briefly or has a tendency to stay in the area for a long time. On the basis of the above, a buffer time zone is inserted (the duration of the human body sensor does not exceed 20% of its threshold value) to eliminate the influence of the walking speed difference of different people on the duration.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to an automatic lighting adjustment system based on scene analysis. Background Art

[0002] By analyzing scene information such as human activity, the brightness and on / off status of lamps can be precisely controlled. For example, in areas with frequent movement, a rule can be set: dim the lights when people are present and dim them when no one is around. This prevents lamps from being left on unnecessarily in areas with low foot traffic, thereby reducing energy consumption. However, if people move around too frequently, the light brightness will change frequently, which not only shortens the lamp's lifespan but also creates a negative visual experience for those in the area.

[0003] Currently, to prevent the lighting brightness from being adjusted too frequently, an adjustment model is usually established to learn the relationship between people's movement patterns and lighting adjustments, so as to identify the difference between people staying in the area for a short time and those who are active for a long time. For example, within a period of time (recorded as the window time), the number of times a person triggers the human body sensor in the area and the duration of the human body sensor being triggered are recorded. If the number of triggers reaches a certain amount, the duration does not exceed a pre-set threshold, and the human body sensor is not triggered again within a certain period of time after the window time ends, it can be determined that the person is passing through the area briefly and the light brightness does not need to be adjusted; otherwise, the light will remain on.

[0004] However, in areas of a shopping mall where certain products are sold, due to the varying walking speeds of different people, if the duration of a human presence sensor trigger is slightly longer than a pre-set threshold (e.g., exceeding 10%), the original control model will cause the lights to remain on. This will cause the model to interpret the person as inclined to purchase the product, when in reality, the person is simply walking slower or taking a different path. This affects the accuracy of the control model, making it difficult to issue accurate control instructions to the lights based on the differences in walking speed and path.

[0005] In order to address the above problems, there is an urgent need for an automatic lighting adjustment system based on scene analysis to improve the shortcomings of the existing technology. Summary of the Invention

[0006] The object of the present invention is to provide a lighting automatic adjustment system based on scene analysis to solve the problems raised in the above background technology.

[0007] The present invention provides an automatic lighting adjustment system based on scene analysis. In a shopping mall where certain products are sold, due to the differences in walking speeds of different people, if the duration of the human body sensor trigger is slightly longer than a preset threshold, according to the original adjustment model settings, the lights will be adjusted to be constantly on. This is because the adjustment model will judge the person as having a tendency to buy products, when in fact the person is just walking slowly or taking a different path through the area. This affects the accuracy of the adjustment model, namely:

[0008] It is difficult to issue correct adjustment instructions to the lamps based on the differences in people's walking speeds and paths.

[0009] To achieve the above objectives, a scene analysis-based automatic lighting adjustment system is provided, comprising a scene perception module, a central control module, a lighting adjustment module, and a storage and management module. The scene perception module includes a plurality of human body sensors, and the central control module includes a microprocessor.

[0010] The scene perception module is used to collect environmental information about the area where the lamp is located. The central control module processes the environmental information transmitted by the scene perception module and determines the lamp adjustment strategy based on the long short-term memory model. The steps of the adjustment strategy are as follows:

[0011] S1. Count the number of triggers and duration of the human body sensor within a period of time, and set thresholds for the number of triggers and duration of the human body sensor, denoted as M and T1 respectively;

[0012] S2. Compare the triggering times and duration of the time window with their respective thresholds, and record whether the human body sensor is triggered within a period of time after the end of the time window;

[0013] S3. Determine whether the following conditions are met: the number of triggers ≥ M, the duration ≤ T1, and the human body sensor is not triggered again within T2;

[0014] S4. Set a time buffer t and determine whether the duration satisfies the following conditions: T1 < duration ≤ (1 + t) T1;

[0015] S5. Determine whether the path of the person passing through the area where the human body sensor is located satisfies the following conditions: Passing path ≤ maximum path in the area;

[0016] S6. Calculate the speed variance when the person passes through every two adjacent human body sensors, and compare whether the speed variance is greater than a preset value.

[0017] In S3, if all conditions are met, it means that the person has briefly passed through the area where the lamp is located, and the lamp is not adjusted; otherwise, the judgment in S4 is carried out. If the judgment conditions in S3 are not met, it is simply defined as the person has a tendency to purchase goods in the area, and the lamp is adjusted to a constant state. This actually has a certain deviation. In the mall, if the person is in the area simply because of mobility difficulties, the duration of the human body sensor is slightly longer than T1 (for example, more than 10%). According to the original long short-term memory model (LSTM model) settings, the LSTM model will assume that the person has a tendency to view goods in the area and purchase them, so it will adjust the lamp to a constant state, which makes the LSTM model inaccurate.

[0018] In S4, if the condition T1 < duration ≤ (1+t)T1 is not met, the lights are set to constant on; otherwise, the decision in S5 is made. After inserting a time buffer t (depending on the walking speed of different people, the value of t is usually between 0-20%), if the duration of the human sensor is no longer than (1+t)T1 when the person is in the area, it can be assumed that the duration of the human sensor is slightly longer than T1 due to the person's slow walking speed, and the lights still do not need to be adjusted. If the duration of the human sensor is longer than (1+t)T1 when the person is in the area, it can be preliminarily determined that the person may be purchasing goods in the area.

[0019] In S5, if the condition is not met: the passing path ≤ the maximum path in the area, the lights are adjusted to be always on; otherwise, the judgment in S6 is carried out. While S4 eliminates the difference in walking speed of people, since there can be multiple paths through a certain area of ​​the mall, the distances of different paths are different, which may also cause different times for people to pass through the area, and thus cause different durations of human body sensors. If the path of the person through the area (the distances between the triggered human body sensors are added in sequence) is not greater than the maximum path in the area, it is considered that the duration of the human body sensor is slightly greater than T1 due to the different passing paths, and the lights are still not adjusted. On the contrary, it is considered that there are repeated paths when people pass through the area, so the lights are adjusted to be always on to meet the needs of people to purchase goods.

[0020] In S6, if the speed variance is greater than the preset value, the light is adjusted to be constantly on; otherwise, it represents that the person briefly passes through the area where the light is located, and the light is not adjusted.

[0021] In S6, the steps of velocity variance are as follows:

[0022] First, calculate the speed at which people pass between each group of adjacent human body sensors. Assume that the time it takes for a person to pass through the i-th group of adjacent human body sensors is t i , the distance between two adjacent human body sensors is di , then the speed at which people pass through this group of human sensors is

[0023] Then, calculate the average value v0 of the speed. Assuming there are n sets of adjacent human body sensors, then

[0024] Finally, the speed variance is calculated according to the variance formula

[0025] Among them, S 2 The variance of speed reflects the degree of dispersion of the speed of people passing through different groups of adjacent human body sensors. The larger the variance, the greater the fluctuation of speed, and the smaller the variance, the more stable the speed.

[0026] After the judgment of S3-S5, after considering the factors of people's walking speed and the difference in the passing path, for those people who meet the conditions of T1 < duration ≤ (1+t)T1 and the passing path ≤ the maximum path of the area, which still includes those who pass through briefly and those who have a tendency to shop. At this time, by comparing the speed fluctuations of people passing through the area, the activity status of people passing through the area is analyzed. If S 2 > preset variance, it means that the speed of people passing through the area varies greatly. It can be considered that when the speed of people in the area is slow, they are staying in the area to view product information and have a tendency to shop. At this time, the lights are adjusted to be always on. When the speed of people is fast, they are walking through the area and have no tendency to shop. If S 2 < preset variance, it means that the speed of people passing through the area changes little, and they are always walking through the area. At this time, the lamps do not need to be adjusted.

[0027] In this solution, the environmental information transmitted by the scene perception module includes human activity information and light intensity information. The scene perception module also includes a light sensor. The human body sensor detects the presence and movement of people through infrared or microwave technology, and determines whether there is human activity in the area where the human body sensor is located. The light sensor senses the light intensity in the area where the lamp is located and converts the light signal into an electrical signal.

[0028] On this basis, light sensors sense ambient light intensity and convert light signals into electrical signals, allowing the system to understand the current brightness level. Human sensors use infrared or microwave technologies to detect the presence and movement of people, determining whether someone is active in a specific area.

[0029] In this technical solution, the microprocessor calculates the speed v between two adjacent human body sensors based on the human activity information transmitted by the human body sensor. i , calculate the speed v between two adjacent human body sensors i, where i is 1, 2, 3...n), v i The calculation formula is:

[0030]

[0031] The storage unit stores preset parameters and historical data of the lamp adjustment system, wherein the preset parameters include a time window, thresholds for the number of human sensor triggers and trigger time, an interval time, a time buffer, a maximum path through the area where the lamp is located, a time taken to pass through the area where the lamp is located, and a preset value of a speed variance;

[0032] The communication unit uses a wired or wireless method to achieve data transmission between different modules.

[0033] As a further improvement to the above technical solution, a microprocessor serves as the system's "brain," running the control program, analyzing, processing, and making decisions based on input data, and coordinating the operations of various modules. The system's preset parameters, historical data, and operating programs are stored in a storage unit, ensuring proper operation and a certain level of memory and learning capabilities. The communication unit utilizes either wired (e.g., RS485, CAN bus) or wireless (e.g., Zigbee, Wi-Fi, Bluetooth) communication to transfer data between modules.

[0034] In this solution, the dimming unit adjusts the brightness and color temperature of the lamp according to the control instructions issued by the microprocessor. The dimming unit adjusts the brightness of the lamp, and the color adjustment unit adjusts the color temperature of the lamp by mixing light sources of different colors.

[0035] The driver circuit drives the lamp's light source according to the control signal, achieving functions such as brightness adjustment and color temperature switching. Different types of light sources (such as LEDs and fluorescent lamps) require different driver circuits. The dimming unit specifically performs brightness adjustment, changing the input current or voltage of the light source through technologies such as PWM (pulse width modulation), thereby achieving stepless brightness adjustment. For lamps with color adjustment functions, the color adjustment unit is responsible for adjusting the color temperature, typically by mixing light sources of different colors (such as warm white and cool white LEDs).

[0036] Preferably, the database stores all data generated by the operation of the lamp adjustment system, and the analysis unit analyzes the data stored in the database and provides a basis for optimizing the lamp adjustment strategy.

[0037] In this solution, the database uses a relational database (such as MySQL) or a non-relational database (such as MongoDB) to store scene perception data, lighting adjustment records, user settings, and other information. The analysis unit analyzes this stored data, uncovering patterns and trends. For example, it analyzes lighting usage habits across different time periods and scenarios to provide a basis for optimizing lighting strategies.

[0038] Based on the above description, it can be seen that compared with the prior art, the beneficial effects of the present invention are:

[0039] The long-short-term memory model records the number of times and duration of human body sensor triggering when a person passes through an area, and whether it is triggered again within the interval time, to determine whether the person passes through the area briefly or has a tendency to stay in the area for a long time (for example, with a tendency to view goods or shop).

[0040] On the basis of the above, a buffer time zone is inserted (the duration of the human body sensor does not exceed 20% of its threshold) to eliminate the impact of different people's walking speed differences on the duration.

[0041] At the same time, by comparing the path a person takes through the area with the maximum path, the interference caused by different paths through the area on the human sensor duration is eliminated. The speed fluctuations of people passing through the area are then calculated to determine whether they are walking continuously or stopping to view products. This improves the accuracy of the long-short-term memory model when issuing adjustment commands to the lighting by analyzing the activity status of people in the scene. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is the overall structural diagram of the present invention;

[0043] Figure 2 This is a flow chart of the steps of the central control module of the present invention identifying human behavior and issuing instructions to the lamps.

[0044] The meaning of each number in the figure is:

[0045] 100, scene perception module; 110, human body sensor; 120, light sensor;

[0046] 200, central control module; 210, microprocessor; 220, storage unit; 230, communication unit;

[0047] 300, lamp adjustment module; 310, driving circuit; 320, dimming unit; 330, color adjustment unit;

[0048] 400, storage and management module; 410, database; 420, analysis unit. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0050] See also Figure 1 and Figure 2 As shown, a lighting automatic adjustment system based on scene analysis is provided, including a scene perception module 100, a central control module 200, a lighting adjustment module 300 and a storage and management module 400. The scene perception module 100 includes a plurality of human sensors 110, and the central control module 200 includes a microprocessor 210;

[0051] The scene perception module 100 is used to collect environmental information about the area where the lamp is located. The central control module 200 processes the environmental information transmitted by the scene perception module 100 and determines the lamp adjustment strategy based on the long short-term memory model. The steps of the adjustment strategy are as follows:

[0052] S1. Count the number of triggers and duration of the human body sensor 110 within a period of time (recorded as a time window), and set thresholds for the number of triggers and duration of the human body sensor 110, denoted as M and T1 respectively;

[0053] S2. Compare the triggering times and duration of the time window with respective thresholds, and record whether the human body sensor 110 is triggered within a period of time after the end of the time window as interval time T2;

[0054] S3. Determine whether the following conditions are met: the number of triggers ≥ M, the duration ≤ T1, and the human body sensor 110 is not triggered again within T2;

[0055] S4. Set a time buffer t and determine whether the duration satisfies the following conditions: T1 < duration ≤ (1 + t) T1;

[0056] S5. Determine whether the path of the person passing through the area where the human body sensor 110 is located satisfies: passing path ≤ maximum path in the area;

[0057] S6. Calculate the speed variance when the person passes through every two adjacent human body sensors 110, and compare whether the speed variance is greater than a preset value.

[0058] It should be disclosed that in S3, if all conditions are met, it means that the person briefly passes through the area where the lamp is located and the lamp is not adjusted; otherwise, the determination in S4 is performed.

[0059] In S4 , if the condition: T1 < duration ≤ (1+t)T1 is not satisfied, the lamp is adjusted to be constantly on; otherwise, the determination in S5 is performed.

[0060] In S5 , if the condition is not satisfied: the passing path ≤ the maximum path in the area, the lamp is set to be constantly on; otherwise, the determination in S6 is performed.

[0061] In S6, if the speed variance is greater than the preset value, the light is adjusted to be constantly on; otherwise, it represents that the person briefly passes through the area where the light is located, and the light is not adjusted.

[0062] In S6, the steps of velocity variance are as follows:

[0063] First, calculate the speed of people passing between each group of adjacent human sensors 110. Assume that the time it takes for a person to pass through the i-th group of adjacent human sensors 110 is t i The distance between two adjacent human body sensors 110 is d i , then the speed at which a person passes through the group of human sensors 110

[0064] Then, the average value v0 of the speed is calculated. Assuming there are n groups of adjacent human body sensors 110, then

[0065] Finally, the speed variance is calculated according to the variance formula

[0066] Among them, S 2 The variance of the speed reflects the degree of dispersion of the speed of a person passing through different groups of adjacent human body sensors 110. The larger the variance, the greater the fluctuation of the speed, and the smaller the variance, the more stable the speed.

[0067] In order to collect environmental information in real time and provide a basis for adjusting lamps; the environmental information transmitted by the scene perception module 100 includes personnel activity information and light intensity information. The scene perception module 100 also includes a light sensor 120. The human body sensor 110 detects the presence and movement of people through infrared or microwave technology, and determines whether there is human activity in the area where the human body sensor 110 is located. The light sensor 120 senses the light intensity in the area where the lamp is located and converts the light signal into an electrical signal.

[0068] The light sensor 120 senses the ambient light intensity and converts the light signal into an electrical signal so that the system can understand the brightness of the current environment. The human sensor 110 detects the presence and movement of human bodies through infrared or microwave technology to determine whether there are people in a specific area.

[0069] Considering the need to process the environmental information data transmitted by the scene perception module 100, the adjustment strategy of the lamp is determined according to the LSTM model; the central control module 200 also includes a storage unit 220 and a communication unit 230. The microprocessor 210 calculates the speed v between two adjacent human sensors 110 based on the human activity information transmitted by the human sensor 110. i , calculate the speed v between two adjacent human body sensors 110 i , where i is 1, 2, 3...n (n means a person has passed through n sets of adjacent human sensors 110), v i The calculation formula is:

[0070]

[0071] The storage unit 220 stores preset parameters and historical data of the lamp adjustment system, wherein the preset parameters include a time window, thresholds for the number of triggers and trigger time of the human sensor 110, an interval time, a time buffer, a maximum path through the area where the lamp is located, the time it takes to pass through the area where the lamp is located, and a preset value of the speed variance;

[0072] The communication unit 230 uses a wired or wireless method to achieve data transmission between different modules.

[0073] Microprocessor 210 serves as the system's "brain," running the control program, analyzing, processing, and making decisions based on input data, and coordinating the operations of the various modules. The system's preset parameters, historical data, and operating programs are stored in storage unit 220, ensuring the system's normal operation and possessing a certain level of memory and learning capabilities. Communication unit 230 utilizes wired (e.g., RS485, CAN bus) or wireless (e.g., Zigbee, Wi-Fi, Bluetooth) communication to transfer data between modules.

[0074] After receiving the instructions from the communication unit 230, the brightness, color temperature, switch status, etc. of the lamp are adjusted next; the lamp adjustment module 300 includes a driving circuit 310, a dimming unit 320 and a color adjustment unit 330. The dimming unit 320 adjusts the brightness and color temperature of the lamp according to the control instructions issued by the microprocessor 210. The dimming unit 320 adjusts the brightness of the lamp, and the color adjustment unit 330 adjusts the color temperature of the lamp by mixing light sources of different colors.

[0075] The driver circuit 310 drives the lamp's light source according to the control signal, achieving functions such as brightness adjustment and color temperature switching. Different types of light sources (such as LEDs and fluorescent lamps) require different driver circuits 310. The dimming unit 320 specifically performs the brightness adjustment function, changing the input current or voltage of the light source through technologies such as PWM (pulse width modulation), thereby achieving stepless brightness adjustment. For lamps with color adjustment functions, the color adjustment unit 330 is responsible for adjusting the color temperature, typically by mixing light sources of different colors (such as warm white and cool white LEDs).

[0076] In order to store various data during the operation of the system, manage and analyze them, and provide support for system optimization and decision-making; the storage and management module 400 includes a database 410 and an analysis unit 420. The database 410 stores all data generated by the operation of the lighting adjustment system, and the analysis unit 420 analyzes the data stored in the database 410 and provides a basis for optimizing the lighting adjustment strategy.

[0077] Database 410 uses a relational database (e.g., MySQL) or a non-relational database (e.g., MongoDB) to store scene perception data, lighting adjustment records, user settings, and other information. Analysis unit 420 analyzes the stored data, uncovering patterns and trends within the data. For example, it analyzes lighting usage habits across different time periods and scenarios to provide a basis for optimizing lighting strategies.

[0078] In summary, the working principle of the present invention is as follows:

[0079] First, count the number of times human sensor 110 is triggered and its duration within a period of time (referred to as a time window). Thresholds are set for the number of times human sensor 110 is triggered, and for its duration, respectively, denoted as M and T1. The number of times human sensor 110 is triggered and its duration within the time window are compared with their respective thresholds. A period of time after the end of the time window, referred to as interval time T2, is recorded to determine whether the following conditions are met: number of triggers ≥ M, duration ≤ T1, and human sensor 110 is not triggered again within T2.

[0080] If the judgment condition in S3 is not met, the user is simply defined as having a tendency to purchase goods in the area, and the lights are adjusted to a constant state. This actually has a certain deviation. In the mall, if the user is in the area simply because of mobility difficulties, the duration of the human sensor 110 is slightly longer than T1. According to the original long short-term memory model settings, the LSTM model will assume that the person has a tendency to view goods in the area and purchase them, and will therefore adjust the lights to a constant state, which makes the LSTM model inaccurate. Therefore, a time buffer t is set to determine whether the duration satisfies the following: T1 < duration ≤ (1 + t) T1.

[0081] After inserting a time buffer t, if the duration of the presence sensor 110's presence detection while the person is in the area is no longer than (1 + t) T1, it can be assumed that the person's slow walking speed caused the presence sensor 110's presence detection duration to be slightly longer than T1, and in this case, the lighting adjustment is still unnecessary. If the duration of the presence sensor 110's presence detection duration while the person is in the area is longer than (1 + t) T1, it can be preliminarily determined that the person may be purchasing goods in the area. At this point, the person's path through the area where the presence sensor 110 is located satisfies the following condition: Passing Path ≤ Maximum Path in the Area.

[0082] While eliminating the factor of varying walking speeds, since a certain area of ​​the mall can have multiple paths with varying lengths, this can also cause people to take different amounts of time to pass through the area, leading to different durations of motion on human sensors 110. If the path a person takes through the area is no longer than the maximum path in the area, the duration of motion on human sensors 110 slightly exceeding T1 is assumed to be due to different paths, and the lighting is not adjusted. Conversely, it is assumed that the person has passed through the area repeatedly, so the lighting is set to constant on to satisfy the person's shopping needs. The speed variance of each person passing through two adjacent human sensors 110 is then calculated and compared to see if the speed variance is greater than a preset value.

[0083] After the above steps, and considering the factors of people's walking speed and the difference in the paths they pass through, for those people who meet the conditions of T1 < duration ≤ (1 + t) T1 and the paths they pass through ≤ the maximum path in the area, including those who pass through briefly and those who have a tendency to shop, the activity status of people passing through the area is analyzed by comparing the speed fluctuations of people passing through the area. If S 2 > preset variance, it means that the speed of people passing through the area varies greatly. It can be considered that when the speed of people in the area is slow, they are staying in the area to view product information and have a tendency to shop. At this time, the lights are adjusted to be always on. When the speed of people is fast, they are walking through the area and have no tendency to shop. If S 2 < preset variance, it means that the speed of people passing through the area changes little, and they are always walking through the area. At this time, the lamps do not need to be adjusted.

[0084] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A lighting automatic adjustment system based on scene analysis, comprising a scene perception module (100), a central control module (200), a lighting adjustment module (300) and a storage and management module (400), characterized in that: The scene perception module (100) includes a plurality of human body sensors (110), and the central control module (200) includes a microprocessor (210); The central control module (200) determines the adjustment strategy of the lamp based on the long short-term memory model, and the steps are as follows: S1, counting the number of triggering times and duration of the human body sensor (110) within the time window, and setting thresholds for the number of triggering times and duration, respectively, denoted as M and T1; S2, comparing the number of triggering times and duration of the time window with respective thresholds, and recording whether the human body sensor (110) is triggered within the interval T2 after the end of the time window; S3, judging whether the following conditions are met: the number of triggers ≥ M, the duration ≤ T1, and the human body sensor (110) is not triggered again within T2; S4. Set the time buffer t and determine whether the following conditions are met: T1 < duration ≤ (1 + t) T1; S5, determining whether the passing path of the person in the area where the human body sensor (110) is located satisfies: passing path ≤ maximum path in the area; S6, calculating the speed variance when a person passes through every two adjacent human body sensors (110), and comparing whether it is greater than a preset value; In S3, if all conditions are met, it means that the person briefly passes through the area where the lamp is located, and the lamp is not adjusted; Otherwise, proceed to S4; In S4, if the condition is not met: T1 < duration ≤ (1 + t) T1, the lamp is turned to be constantly on; otherwise, the judgment of S5 is carried out; In S5, if the condition is not met: the passing path ≤ the maximum path in the area, the lamp is set to be constantly on; otherwise, the judgment of S6 is carried out; In S6, if the speed variance is greater than the preset value, the light is adjusted to be constantly on; otherwise, it represents that the person briefly passes through the area where the light is located, and the light is not adjusted.

2. The automatic lighting adjustment system based on scene analysis according to claim 1, characterized in that: The environmental information transmitted by the scene perception module (100) includes human activity information and light intensity information. The scene perception module (100) further includes a light sensor (120). The human body sensor (110) detects the presence and movement of a person through infrared or microwave technology, and determines whether there is human activity in the area where the human body sensor (110) is located. The light sensor (120) senses the light intensity in the area where the lamp is located, and converts the light signal into an electrical signal.

3. The automatic lighting adjustment system based on scene analysis according to claim 1, characterized in that: The central control module (200) further includes a storage unit (220) and a communication unit (230). The microprocessor (210) calculates the speed of a person passing through two adjacent human body sensors (110) based on the human activity information transmitted from the human body sensors (110). , calculate the speed of a person passing between two adjacent human body sensors (110) , where i is 1, 2, 3...n (n means that the person has passed through n sets of adjacent human body sensors (110)); The storage unit (220) stores preset parameters and historical data of the lamp adjustment system, wherein the preset parameters include a time window, a threshold value of the number of triggering times and triggering time of the human body sensor (110), an interval time, a time buffer, a maximum path through the area where the lamp is located, a time spent passing through the area where the lamp is located, and a preset value of a speed variance; The communication unit (230) uses a wired or wireless method to achieve data transmission between different modules.

4. The automatic lighting adjustment system based on scene analysis according to claim 1, characterized in that: The lamp adjustment module (300) comprises a driving circuit (310), a dimming unit (320) and a color adjustment unit (330). The dimming unit (320) adjusts the brightness and color temperature of the lamp according to a control instruction issued by the microprocessor (210). The dimming unit (320) adjusts the brightness of the lamp, and the color adjustment unit (330) adjusts the color temperature of the lamp by mixing light sources of different colors.

5. The automatic lighting adjustment system based on scene analysis according to claim 1, characterized in that: The storage and management module (400) comprises a database (410) and an analysis unit (420), wherein the database (410) stores all data generated by the operation of the lamp adjustment system, and the analysis unit (420) analyzes the data stored in the database (410) and provides a basis for optimizing the lamp adjustment strategy.

6. The automatic lighting adjustment system based on scene analysis according to claim 1, characterized in that: In S6, the steps of velocity variance are as follows: First, the speed at which people pass between each set of adjacent human sensors (110) is calculated; Then, calculate the average speed ; Finally, the speed variance is calculated according to the variance formula.

Images