A lighting control system based on lighting scene recognition

Abstract

The present invention relates to the field of intelligent lighting control technology, and in particular to a lighting control system based on lighting scene recognition, including a scene recognition module, a scene analysis module, a light source analysis module, a control analysis module and a lighting control module, which first analyzes the position information and orientation information of each character in the scene to obtain a target lighting area, and then establishes a solution space based on the position information of each light source, the lighting parameters of the light source and the light source adjustment parameters, and based on an optimization algorithm, finds the optimal point in the solution space and controls each light source based on the optimal point. Compared with the prior art, the present invention describes the lighting needs of multiple characters in the scene by establishing a target light source area, and obtains the best lighting control strategy that can take into account both lighting effects and shadow effects by means of an optimization algorithm, solving the problem that the lighting control technology in the prior art is not optimized for multi-person scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent lighting control, and in particular to a lighting control system based on lighting scene recognition. Background Art

[0002] Modern lighting systems need to adapt to different environments and scenarios, such as homes, offices, commercial spaces, etc. Each scenario has its own specific lighting requirements. Although there are some intelligent lighting control systems on the market, they often rely on preset modes or manual control and lack the ability to intelligently identify actual scenarios, limiting the flexibility and efficiency of lighting systems.

[0003] Although there are some technologies that can adjust lighting according to scene conditions, most of the existing technologies are still overall lighting adjustment, and the flexibility is still unsatisfactory. On the one hand, the overall adjustment is not optimized for the specific position of each character in the scene, and the lighting utilization rate is not optimal. On the other hand, the overall adjustment does not take into account the mutual influence of shadows between multiple characters, which affects the user's lighting experience.

[0004] Therefore, people need a lighting control system that can be optimized for multi-person scenarios. Summary of the invention

[0005] Therefore, the present invention provides a lighting control system based on lighting scene recognition to solve the problem that the lighting control technology in the prior art is not optimized for multi-person scenes.

[0006] The present invention provides a lighting control system based on lighting scene recognition, comprising:

[0007] The scene recognition module is used to obtain and analyze scene data to obtain the position and orientation information of each character in the scene;

[0008] A scene analysis module is used to obtain the target lighting area based on the position information and orientation information of each character;

[0009] A light source analysis module, used to obtain light source position information, light source lighting parameters and light source adjustment parameters of each light source, and establish a solution space according to each light source position information, light source lighting parameters and light source adjustment parameters, wherein each point in the solution space is used to represent a lighting control strategy;

[0010] The control analysis module is used to find the optimal point in the solution space based on the optimization algorithm. The lighting control strategy corresponding to the optimal point can achieve the highest lighting comfort and shadow coverage while minimizing energy consumption. The shadow coverage indicates the degree of coverage of the target lighting area by the shadows of all characters in the scene illuminated by all light sources.

[0011] The lighting control module is used to control each light source based on the optimal point.

[0012] In a preferred implementation, based on an optimization algorithm, finding the optimal point in the solution space includes:

[0013] According to the target point, calculate the shadow feature value of the target person relative to each light source, where the target point is a point to be analyzed in the solution space, the target person is a person to be analyzed, and the shadow feature value represents the coverage degree of the target lighting area by the shadow of the target person when illuminated by a light source;

[0014] Count the number of light sources whose shadow characteristic values ​​are higher than a preset threshold to obtain the number of key light sources;

[0015] According to the sum of multiple shadow eigenvalues ​​and the number of key light sources, the shadow coverage eigenvalue of the target person is obtained, wherein the shadow coverage eigenvalue is positively correlated with the sum of multiple shadow eigenvalues, and the shadow coverage eigenvalue is negatively correlated with the number of key light sources. The sum of the shadow coverage eigenvalues ​​of all characters constitutes the shadow coverage rate of the target point.

[0016] In a preferred implementation, calculating the shadow feature value of the target person relative to each light source according to the target point includes:

[0017] According to the target illumination area, the target point and the person position information, a projection area of ​​the target illumination area, a person projection point indicating the target person position, a light source projection point indicating the target light source position, and a direction projection unit vector indicating the target light source illumination direction are established in a preset reference plane, wherein the target light source is a light source to be analyzed currently;

[0018] In a preset reference plane, the light source projection point is mirrored based on the person projection point to obtain a mirror point, and the length of the line between the mirror point and the person projection point in the projection area is calculated to obtain a shadow length coefficient;

[0019] Calculate the angle between the unit mass of the directional projection and the line connecting the mirror point and the projection point of the person to obtain the illumination angle, and obtain the directional coefficient according to the illumination angle, wherein the illumination angle is negatively correlated with the directional coefficient;

[0020] Get the intensity value of the target light source according to the target point;

[0021] The shadow characteristic value of the target person relative to the target light source is obtained according to the product of the shadow length coefficient, the direction coefficient and the intensity value.

[0022] In a preferred implementation, the illumination angle ranges from zero to one hundred and eighty degrees. If the illumination angle is greater than or equal to ninety degrees, the corresponding directivity coefficient is zero.

[0023] In a preferred implementation, obtaining a target illumination area according to the position information and orientation information of each character includes:

[0024] According to the position and orientation information of each character, establish the lighting requirement area for each character;

[0025] Summarize multiple lighting demand areas to obtain the target lighting area.

[0026] In a preferred implementation, according to the position information and orientation information of each character, a lighting requirement area for each character is established, including:

[0027] According to the character position information and character orientation information of each character, a orientation vector of each character is established, wherein the starting point of the orientation vector is the position of the character and the length is a preset length value;

[0028] With the end point of each orientation vector as the center, an area of ​​a preset shape is established as the lighting requirement area for each character.

[0029] In a preferred implementation, a plurality of lighting demand areas are aggregated to obtain a target lighting area, including:

[0030] Obtaining light source density, and expanding multiple lighting demand areas according to the light source density, wherein the extent of expansion of the lighting demand area is negatively correlated with the light source density;

[0031] The multiple enlarged lighting demand areas are aggregated to obtain the target lighting area.

[0032] In a preferred implementation, a solution space is established according to each light source position information, light source illumination parameters, and light source adjustment parameters, including:

[0033] According to the light source illumination parameters and the light source adjustment parameters, a plurality of adjustable features are obtained;

[0034] Standardizing actual value ranges of multiple adjustable features to obtain a standard value range for each adjustable feature;

[0035] A solution space is established by taking an adjustable feature of a light as a dimension and a standard value range of the adjustable feature as the value domain of the dimension corresponding to the adjustable feature.

[0036] In a preferred implementation, a solution space is established according to the position information of each light source, the illumination parameters of the light source and the adjustment parameters of the light source, and further includes:

[0037] Count invalid light sources;

[0038] In the solution space, the scales corresponding to the invalid light sources are trimmed to obtain an optimized solution space.

[0039] In a preferred implementation, based on an optimization algorithm, finding the optimal point in the solution space includes:

[0040] Obtain multiple initial points in the solution space;

[0041] Analyze the lighting comfort, shadow coverage and energy consumption of each initial point to obtain the fitness of each initial point;

[0042] Based on the fitness of each initial point, the positions of multiple initial points in the solution space are optimized to obtain multiple optimization points;

[0043] The optimization point is used as the new starting point, and the optimization is repeated until the preset end condition is reached, and the optimization point with the best fitness obtained last time is used as the optimal point.

[0044] The beneficial effects of adopting the above embodiment are:

[0045] The present invention provides a lighting control system based on lighting scene recognition, including a scene recognition module, a scene analysis module, a light source analysis module, a control analysis module and a lighting control module, which first obtains and analyzes scene data to obtain the position information and orientation information of each character in the scene, and then obtains the target lighting area according to the position information and orientation information of each character, and then obtains the light source position information, light source lighting parameters and light source adjustment parameters of each light source, and establishes a solution space according to each light source position information, light source lighting parameters and light source adjustment parameters, and based on the optimization algorithm, finds the optimal point in the solution space and controls each light source based on the optimal point. Compared with the prior art, the present invention describes the lighting needs of multiple characters in the scene by establishing a target light source area, and realizes multi-objective optimization of the highest lighting comfort and shadow coverage rate and the lowest energy consumption by means of an optimization algorithm, and obtains the best lighting control strategy that can take into account both lighting effects and shadow influences, solving the problem that the lighting control technology in the prior art is not optimized for multi-person scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A system architecture diagram of a lighting control system based on lighting scene recognition provided by the present invention;

[0047] Figure 2 for Figure 1 The method steps of each module are as follows:

[0048] Figure 3 for Figure 2 Specific step diagram of step S202;

[0049] Figure 4 for Figure 2 Specific steps of calculating the shadow coverage rate in step S204;

[0050] Figure 5 for Figure 4 Specific step diagram of step S401;

[0051] Figure 6 for Figure 4 Schematic diagram of calculation of step S401 in FIG. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] Combination Figure 1-2 As shown, a specific embodiment of the present invention discloses a lighting control system based on lighting scene recognition, including a scene recognition module 110, a scene analysis module 120, a light source analysis module 130, a control analysis module 140 and a lighting control module 150, and the five modules are respectively used to perform the following five steps:

[0054] S201, acquiring and analyzing scene data to obtain position information and orientation information of each character in the scene;

[0055] S202, obtaining a target illumination area according to the position information and orientation information of each character;

[0056] S203, obtaining light source position information, light source lighting parameters and light source adjustment parameters of each light source, and establishing a solution space according to each light source position information, light source lighting parameters and light source adjustment parameters, wherein each point in the solution space is used to represent a lighting control strategy;

[0057] S204, based on the optimization algorithm, searching for the optimal point in the solution space, wherein the lighting control strategy corresponding to the optimal point can achieve the highest lighting comfort and shadow coverage while minimizing energy consumption, and the shadow coverage indicates the degree of coverage of the target lighting area by the shadows of all characters in the scene illuminated by all light sources;

[0058] S205. Control each light source based on the optimal point.

[0059] In the above process, step S201 can be implemented by existing image recognition technology, which will not be described in detail herein. In practice, in most scenes that require lighting control, the orientation of people is regular, such as restaurants, classrooms, offices, etc., and the existing technology performs well in detecting the position and orientation of people. The target lighting area is the area that needs to be illuminated obtained by analyzing the position information and orientation information of the person.

[0060] The light source position information refers to the position of each light source. The light source lighting parameters are the parameters of the light source, such as color temperature, brightness, power, etc. The light source adjustment parameters are the types and adjustable ranges of the parameters that can be adjusted by the light source, such as brightness, color temperature, illumination angle, etc.

[0061] Lighting comfort is an indicator to measure whether the lighting control effect is appropriate. Its specific standards can be flexibly set according to actual needs, such as the illuminance, uniformity, and softness in the target lighting area. Similarly, the specific calculation method of the energy consumption and shadow coverage of a lighting control strategy can also be flexibly set according to actual conditions. It is understandable that when the parameters such as the position of the person, the direction of the person, and the position of the light source are known, the specific calculation method of the parameters mentioned in the above content is also the existing technology, so this article will not explain it in detail.

[0062] It can be seen that the present invention needs to optimize multiple light sources at the same time, and also needs to consider multiple objectives such as lighting comfort, shadow coverage and energy consumption during optimization. This is a typical multi-objective optimization problem. Therefore, the present invention uses an optimization algorithm to find the optimal solution to the above problem.

[0063] Compared with the prior art, the present invention describes the lighting needs of multiple characters in the scene by establishing a target light source area, and achieves multi-objective optimization of lighting comfort, maximum shadow coverage, and minimum energy consumption through an optimization algorithm, thereby obtaining the best lighting control strategy that can take into account both lighting effects and shadow influences, solving the problem that the lighting control technology in the prior art is not optimized for multi-person scenes.

[0064] Further, combined with Figure 3 As shown, in a preferred embodiment, step S202, obtaining a target illumination area according to the position information and orientation information of each character, specifically includes:

[0065] S301, establishing a lighting requirement area for each character according to the character position information and character orientation information of each character;

[0066] S302: Summarize multiple lighting demand areas to obtain a target lighting area.

[0067] In the above process, the lighting requirement area is the target lighting area for each character. The above process can merge overlapping lighting requirement areas to obtain a reasonable target lighting area.

[0068] The strategy for establishing the lighting demand area can be flexibly designed according to the actual situation. For example, taking the position information of each character as the starting point, a cone-shaped area is established along the direction of the character.

[0069] In a preferred embodiment, the above step: S301, establishing the lighting requirement area for each character according to the character position information and character orientation information of each character, specifically includes:

[0070] According to the character position information and character orientation information of each character, a orientation vector of each character is established, wherein the starting point of the orientation vector is the position of the character and the length is a preset length value;

[0071] With the end point of each orientation vector as the center, an area of ​​a preset shape is established as the lighting requirement area for each character.

[0072] The preset shape in the above process can be any suitable shape such as sphere, square, etc. The significance of this embodiment is to establish a region of a preset shape with the end point of the orientation vector as the center, because in reality in classrooms, meeting rooms, etc., the focus of a person's gaze may not be in front of them, but may be far away.

[0073] Further, in a preferred embodiment, the above step: S302, summarizing multiple lighting demand areas to obtain a target lighting area, specifically includes:

[0074] Obtaining light source density, and expanding multiple lighting demand areas according to the light source density, wherein the extent of expansion of the lighting demand area is negatively correlated with the light source density;

[0075] The multiple enlarged lighting demand areas are aggregated to obtain the target lighting area.

[0076] Light source density refers to parameters such as the number and spacing between light sources that can indicate the density of light sources. Obviously, in practice, light sources generally do not have obvious boundaries, and the shapes illuminated by multiple light sources are most likely inconsistent with the edges of the target illumination area. Therefore, in this embodiment, the target illumination area is further adjusted according to the light source density to improve the effect of subsequent illumination planning. When the light sources are denser, multiple light sources can illuminate a shape that fits the target illumination area more closely. Therefore, at this time, only a slight expansion of the target illumination area is required, which can achieve precise control, reduce the number of light sources that need to be turned on, and reduce energy consumption. On the contrary, when the light sources are sparse, the coverage of multiple illuminations is larger. At this time, the target illumination area should also be expanded to facilitate subsequent optimization.

[0077] Further, in a preferred embodiment, the above step S203, obtaining the light source position information, light source lighting parameters and light source adjustment parameters of each light source, and establishing a solution space according to each light source position information, light source lighting parameters and light source adjustment parameters, specifically includes:

[0078] According to the light source illumination parameters and the light source adjustment parameters, a plurality of adjustable features are obtained;

[0079] Standardizing actual value ranges of multiple adjustable features to obtain a standard value range for each adjustable feature;

[0080] A solution space is established by taking an adjustable feature of a light as a dimension and a standard value range of the adjustable feature as the value domain of the dimension corresponding to the adjustable feature.

[0081] A point in the solution space can be represented as a one-dimensional vector. Each element in the one-dimensional vector corresponds to an adjustable feature (such as color temperature, rotation angle, pitch angle, brightness, etc.), and the position of the pixel represents the corresponding relationship with each light source.

[0082] Furthermore, in a preferred embodiment, the above step S203, establishing a solution space according to each light source position information, light source illumination parameters and light source adjustment parameters, further includes:

[0083] Count invalid light sources;

[0084] In the solution space, the scales corresponding to the invalid light sources are trimmed to obtain an optimized solution space.

[0085] The significance of the above process is that, in cases where the scene is large or there are few characters, the light sources in some corners have little effect on the lighting experience of the characters, that is, invalid light sources. Therefore, in this embodiment, the solutions corresponding to these invalid light sources are eliminated to reduce the amount of calculation in the subsequent optimization process.

[0086] It is understandable that invalid light sources may be flexibly designated in any suitable existing manner, for example, some light sources that are far from the target illumination area or have a smaller impact may be regarded as invalid light sources.

[0087] Furthermore, in a preferred embodiment, the above step S204, based on the optimization algorithm, searching for the optimal point in the solution space, specifically includes:

[0088] Obtain multiple initial points in the solution space;

[0089] Analyze the lighting comfort, shadow coverage and energy consumption of each initial point to obtain the fitness of each initial point;

[0090] Based on the fitness of each initial point, the positions of multiple initial points in the solution space are optimized to obtain multiple optimization points;

[0091] The optimization point is used as the new starting point, and the optimization is repeated until the preset end condition is reached, and the optimization point with the best fitness obtained last time is used as the optimal point.

[0092] The above process is a common optimization algorithm process in the prior art, which is basically a process of repeatedly iterating the position of the point in the solution space according to the fitness. The specific method of optimizing the position of multiple initial points in the solution space in the above process varies according to the optimization algorithm adopted. For example, for a genetic algorithm, optimizing the position of multiple initial points in the solution space is equivalent to performing operations such as crossover, mutation, and inheritance on the chromosomes corresponding to the multiple initial points. For a particle swarm algorithm, optimizing the position of multiple initial points in the solution space is equivalent to adjusting all initial points to the position of the current better point.

[0093] It is understandable that in this embodiment, the fitness of an initial point needs to calculate three values: lighting comfort, shadow coverage, and energy consumption. How to calculate lighting comfort and energy consumption is a prior art, and the present invention does not explain it in detail. Shadow coverage is also a prior art. For example, assuming that the body shape of each character is the same and a preset outline is set, the shape of its shadow can be calculated based on the existing physical model, and then the shadow coverage can be calculated.

[0094] However, it is foreseeable that the method of calculating the shadow coverage ratio in the prior art is extremely computationally intensive, which will seriously slow down the operation speed of the system and affect the actual operation effect. Therefore, the present invention further improves the problem of overly complex calculation of the shadow coverage ratio.

[0095] Specifically, combined Figure 4 As shown, in a preferred embodiment, in the above step S204, based on the optimization algorithm, in the process of finding the optimal point in the solution space, when calculating the fitness, the process of calculating the shadow coverage rate includes:

[0096] S401, calculating the shadow feature value of the target person relative to each light source according to the target point, wherein the target point is a point to be analyzed in the solution space, the target person is a person to be analyzed, and the shadow feature value represents the coverage degree of the target lighting area by the shadow generated by the target person when illuminated by a light source;

[0097] S402, counting the number of light sources whose shadow characteristic values ​​are higher than a preset threshold, and obtaining the number of key light sources;

[0098] S403. Obtain a shadow coverage characteristic value of the target person based on the sum of multiple shadow characteristic values ​​and the number of key light sources, wherein the shadow coverage characteristic value is positively correlated with the sum of multiple shadow characteristic values, and the shadow coverage characteristic value is negatively correlated with the number of key light sources, and the sum of the shadow coverage characteristic values ​​of all characters constitutes the shadow coverage rate of the target point.

[0099] The shadow coverage feature value is obtained according to the sum of multiple shadow feature values, which indicates the shadow effect produced by multiple light sources acting on the target person. In the above process, the shadow feature value can be calculated by the method described above, and the area of ​​the target person's shadow covering the target illumination area is used as the shadow feature value. Although the present embodiment still uses the prior art to calculate the shadow area produced by a light source on a person, the improvement of the present invention lies in the process of summarizing multiple shadow feature values ​​to obtain the shadow coverage feature value.

[0100] In practice, when multiple light sources illuminate a target at the same time, multiple shadows will be generated. These shadows will not be formed independently, but will also be affected by other light sources. In practice, it is obviously too complicated to simulate this point when calculating the shadow coverage. Therefore, this embodiment simplifies the above steps, that is, the number of key light sources is used to indicate that the shadow is affected by other light sources, and only the sum of multiple shadow eigenvalues ​​and the number of key light sources are used to replace the complex simulation process to reduce the computational complexity. Among them, the key light source is the light source corresponding to the higher shadow eigenvalue, and the higher the shadow eigenvalue, the more "powerful" the light source is, and the stronger its ability to reduce the influence of other shadows. Therefore, the more key light sources there are, the lighter the shadow may be, and the lower the shadow coverage eigenvalue. In this way, this embodiment reduces the amount of calculation and improves the computational efficiency while ensuring the accuracy.

[0101] However, it can be seen that the above embodiment still does not solve the problem that the process of calculating shadows is relatively complicated. Figure 5 As shown, in a preferred embodiment, in a preferred embodiment, the above step S401, calculating the shadow feature value of the target person relative to each light source according to the target point, specifically includes:

[0102] S501, according to the target illumination area, the target point and the person position information, establish in a preset reference plane a projection area of ​​the target illumination area, a person projection point indicating the target person position, a light source projection point indicating the target light source position, and a direction projection unit vector indicating the target light source illumination direction, wherein the target light source is a light source to be analyzed currently;

[0103] S502, in a preset reference plane, mirroring the light source projection point based on the person projection point to obtain a mirror point, and calculating the length of a line between the mirror point and the person projection point within the projection area to obtain a shadow length coefficient;

[0104] S503, calculating the angle between the directional projection unit mass and the line connecting the mirror point and the person projection point to obtain an illumination angle, and obtaining a directional coefficient according to the illumination angle, wherein the illumination angle is negatively correlated with the directional coefficient;

[0105] S504, obtaining the intensity value of the target light source according to the target point;

[0106] S505 . Obtain a shadow characteristic value of the target person relative to the target light source according to the product of the shadow length coefficient, the direction coefficient and the intensity value.

[0107] In the above process, the preset reference plane can generally be regarded as a horizontal plane such as the ground. The key to this embodiment is to use three parameters, namely the shadow length coefficient, the direction coefficient and the intensity value, to fuzzily indicate the size of the shadow, instead of the complicated process of calculating the shadow shape. Among them, the shadow length coefficient indicates the length of the shadow and the degree of coverage of the target lighting area, while the direction coefficient and the intensity value indicate the strength characteristics of the shadow. Through this embodiment, the amount of calculation for calculating the shadow coverage rate can be greatly reduced, and the calculation fitness and the running speed of the optimization algorithm can be significantly improved, making the present invention more practical. The process of calculating the characteristic value of the shadow of the target person relative to the target light source in this embodiment is as follows: Figure 6 As shown, the bold part on the line connecting the character's projection point and the mirror point is the part that affects the shadow length coefficient.

[0108] It is understandable that the above process itself is an embodiment of the idea of ​​sacrificing accuracy for speed. In addition, the present invention is based on the background of light control, and the scattering effect of light in practice cannot form an accurate shape. Therefore, if there are special scenes that do not conform to the rules described in the above process, the error is acceptable. In addition, the above process is a logical step described for the convenience of understanding, and does not mean that the computer running this system will actually draw a picture like Figure 6 Based on the operating principle of computers, under the condition of consistent basic logic, the method of calculating the above shadow characteristic value only by algebraic calculation and other means should also fall within the protection scope of the present invention.

[0109] Furthermore, in a preferred embodiment, the range of the illumination angle is from zero to one hundred and eighty degrees. If the illumination angle is greater than or equal to ninety degrees, the corresponding directional coefficient is zero. It can be understood that for a light source with adjustable direction, when the illumination angle defined above is greater than 90 degrees, it means that the direction of the light source is actually away from the target person. Therefore, the directional coefficient can be directly set to zero at this time, so that the shadow characteristic value of the light source for the target person is also zero. This further reduces the amount of calculation and achieves a better effect.

[0110] The present invention provides a lighting control system based on lighting scene recognition, including a scene recognition module, a scene analysis module, a light source analysis module, a control analysis module and a lighting control module, which first obtains and analyzes scene data to obtain the position information and orientation information of each character in the scene, and then obtains the target lighting area according to the position information and orientation information of each character, and then obtains the light source position information, light source lighting parameters and light source adjustment parameters of each light source, and establishes a solution space according to each light source position information, light source lighting parameters and light source adjustment parameters, and based on the optimization algorithm, finds the optimal point in the solution space and controls each light source based on the optimal point. Compared with the prior art, the present invention describes the lighting needs of multiple characters in the scene by establishing a target light source area, and realizes multi-objective optimization of the highest lighting comfort and shadow coverage rate and the lowest energy consumption by means of an optimization algorithm, and obtains the best lighting control strategy that can take into account both lighting effects and shadow influences, solving the problem that the lighting control technology in the prior art is not optimized for multi-person scenes.

[0111] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0112] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A lighting control system based on lighting scene recognition, characterized in that: include: The scene recognition module is used to obtain and analyze scene data to obtain the position and orientation information of each character in the scene; A scene analysis module is used to obtain the target lighting area based on the position information and orientation information of each character; A light source analysis module, used to obtain light source position information, light source lighting parameters and light source adjustment parameters of each light source, and establish a solution space according to each light source position information, light source lighting parameters and light source adjustment parameters, wherein each point in the solution space is used to represent a lighting control strategy; The control analysis module is used to find the optimal point in the solution space based on the optimization algorithm. The lighting control strategy corresponding to the optimal point can achieve the highest lighting comfort and shadow coverage while minimizing energy consumption. The shadow coverage indicates the degree of coverage of the target lighting area by the shadows of all characters in the scene illuminated by all light sources. A lighting control module, used to control each light source based on the optimal point; Among them, based on the optimization algorithm, the optimal point is found in the solution space, including: According to the target point, calculate the shadow feature value of the target person relative to each light source, where the target point is a point to be analyzed in the solution space, the target person is a person to be analyzed, and the shadow feature value represents the coverage degree of the target lighting area by the shadow of the target person when illuminated by a light source; Count the number of light sources whose shadow characteristic values ​​are higher than a preset threshold to obtain the number of key light sources; According to the sum of multiple shadow feature values ​​and the number of key light sources, the shadow coverage feature value of the target person is obtained, wherein the shadow coverage feature value is positively correlated with the sum of multiple shadow feature values, the shadow coverage feature value is negatively correlated with the number of key light sources, and the sum of the shadow coverage feature values ​​of all characters constitutes the shadow coverage rate of the target point; Obtain multiple initial points in the solution space; Analyze the lighting comfort, shadow coverage and energy consumption of each initial point to obtain the fitness of each initial point; Based on the fitness of each initial point, the positions of multiple initial points in the solution space are optimized to obtain multiple optimization points; The optimization point is used as the new starting point, and the optimization is repeated until the preset end condition is reached, and the optimization point with the best fitness obtained last time is used as the optimal point.

2. The lighting control system based on lighting scene recognition according to claim 1, characterized in that: According to the target point, calculate the shadow feature value of the target person relative to each light source, including: According to the target illumination area, the target point and the person position information, a projection area of ​​the target illumination area, a person projection point indicating the target person position, a light source projection point indicating the target light source position, and a direction projection unit vector indicating the target light source illumination direction are established in a preset reference plane, wherein the target light source is a light source to be analyzed currently; In a preset reference plane, the light source projection point is mirrored based on the person projection point to obtain a mirror point, and the length of the line between the mirror point and the person projection point in the projection area is calculated to obtain a shadow length coefficient; Calculate the angle between the unit mass of the directional projection and the line connecting the mirror point and the projection point of the person to obtain the illumination angle, and obtain the directional coefficient according to the illumination angle, wherein the illumination angle is negatively correlated with the directional coefficient; Get the intensity value of the target light source according to the target point; The shadow characteristic value of the target person relative to the target light source is obtained according to the product of the shadow length coefficient, the direction coefficient and the intensity value.

3. The lighting control system based on lighting scene recognition according to claim 2 is characterized in that: The range of the illumination angle is from zero to one hundred and eighty degrees. If the illumination angle is greater than or equal to ninety degrees, the corresponding directivity coefficient is zero.

4. The lighting control system based on lighting scene recognition according to claim 1, characterized in that: According to the position information and orientation information of each character, the target lighting area is obtained, including: According to the position and orientation information of each character, establish the lighting requirement area for each character; Summarize multiple lighting demand areas to obtain the target lighting area.

5. The lighting control system based on lighting scene recognition according to claim 4 is characterized in that: According to the position and orientation information of each character, establish the lighting requirement area for each character, including: According to the character position information and character orientation information of each character, a orientation vector of each character is established, wherein the starting point of the orientation vector is the position of the character and the length is a preset length value; With the end point of each orientation vector as the center, an area of ​​a preset shape is established as the lighting requirement area for each character.

6. The lighting control system based on lighting scene recognition according to claim 4, characterized in that: Summarize multiple lighting demand areas to obtain the target lighting area, including: Obtaining light source density, and expanding multiple lighting demand areas according to the light source density, wherein the extent of expansion of the lighting demand area is negatively correlated with the light source density; The multiple enlarged lighting demand areas are aggregated to obtain the target lighting area.

7. The lighting control system based on lighting scene recognition according to claim 1, characterized in that: The solution space is established based on the position information of each light source, the lighting parameters of the light source, and the adjustment parameters of the light source, including: According to the light source illumination parameters and the light source adjustment parameters, a plurality of adjustable features are obtained; Standardizing actual value ranges of multiple adjustable features to obtain a standard value range for each adjustable feature; A solution space is established by taking an adjustable feature of a light as a dimension and a standard value range of the adjustable feature as the value domain of the dimension corresponding to the adjustable feature.

8. The lighting control system based on lighting scene recognition according to claim 7, characterized in that: The solution space is established based on the position information of each light source, the lighting parameters of the light source and the adjustment parameters of the light source, and also includes: Count invalid light sources; In the solution space, the scales corresponding to the invalid light sources are trimmed to obtain an optimized solution space.

Images