Landscape lamp control method based on visual detection, storage medium, electronic equipment and landscape lamp

Abstract

The invention relates to the technical field of illumination, in particular to a landscape lamp control method based on visual detection, a storage medium, electronic equipment and a landscape lamp. The method comprises the following steps: acquiring a dressing color group of a tourist in an irradiation range of a landscape lamp; counting the dressing color group to obtain first color data; inputting the first color data into a lamp adjusting large model to obtain a first lamp control parameter; wherein the lamp adjusting large model comprises a first feature layer, a second feature layer and a third feature layer, the first feature layer is used for outputting tourist favorite color data according to the first color data, the second feature layer is used for outputting a first light parameter according to the tourist favorite color, and the third feature layer is used for outputting a first lamp control parameter according to the first light parameter; and controlling the landscape lamp according to the first lamp control parameter. According to the scheme provided by the invention, real-time light adjustment can be carried out according to user preferences, and the adaptive ability of the landscape lamp is enhanced.

Description

Technical Field

[0001] The present application relates to the field of lighting technology, and in particular to a landscape light control method based on visual detection, a storage medium, an electronic device, and a landscape light. Background Art

[0002] Landscape lights are a type of outdoor decorative lighting fixture. Their core function is to enhance the beauty of the environment through light effect design while taking into account basic lighting needs. Therefore, they are not only required to have high ornamental value, but also emphasize the coordination and unity of the landscape of the artistic lights with the historical culture of the scenic area and the surrounding environment. Landscape lights have a wide range of applications, including squares, residential areas, public green spaces and other landscape places. Landscape lights usually adopt a modular design and are composed of multiple light-emitting units. By adjusting parameters such as the brightness, color temperature and light-emitting angle of each unit, a variety of lighting performance effects can be achieved to meet the visual needs of landscape aesthetics. Traditional landscape lights can only operate according to preset programs. Users cannot interact with the landscape lights and can only passively accept the preset lighting options. They cannot meet the differentiated aesthetic needs of individuals, resulting in a single user experience.

[0003] In the existing technology, the solution to the problem of single change of landscape lights is mainly to set multiple modes for landscape lights, and switch them manually or through programs at a fixed rhythm according to different scenes and time. Although this method improves the variability of landscape lights to a certain extent and enhances the user experience to a certain extent, it is still a mechanical preset and cannot be changed according to the preferences of tourists, and cannot meet dynamic personalized needs.

[0004] Therefore, the lack of adaptive capabilities of landscape lights and the inability to make real-time adjustments based on user preferences are technical issues that need to be urgently addressed. Summary of the Invention

[0005] In order to overcome the problems existing in the relevant technology, the present application provides a landscape light control method, storage medium, electronic device and landscape light based on visual detection, which solves the technical problem that the existing technology has insufficient adaptability of landscape lights and cannot make real-time adjustments based on user preferences, which is an urgent problem that needs to be solved.

[0006] A first aspect of the present invention provides a landscape lighting control method based on visual detection, comprising: S1: Obtain the clothing color groups of tourists within the illumination range of landscape lights; S2: Count the clothing color group to obtain first color data; S3: Input the first color data into the light adjustment model to obtain the first light control parameter; The lighting adjustment model includes a first feature layer, a second feature layer, and a third feature layer. The first feature layer is used to output the tourist's favorite color data according to the first color data, the second feature layer is used to output the first lighting parameter according to the tourist's favorite color, and the third feature layer is used to output the first lighting control parameter according to the first lighting parameter. S4: Control the landscape light according to the first light control parameter.

[0007] In a first possible implementation method of the first aspect, the method before S3 further includes: S11: Obtaining the primary color of the environment in which the landscape light is located; S12: Counting the primary colors of the environment to obtain second color data; The second feature layer is used to output a second lighting parameter according to the second color data and the tourist's favorite color data; S4 is replaced by: controlling the landscape light according to the second light control parameter.

[0008] In a second possible implementation method of the first aspect, the step before S3 further includes: S21: Obtain the location coordinates of tourists within the illumination range of the landscape light; S22: Divide the tourist into N groups according to the location coordinates, where N is an integer greater than or equal to 2; S23: Dividing the illumination range of the landscape light into N light zones according to the position coordinates of the N groups of people, and the same group of people is in the same light zone; S2 includes: counting the clothing color group of each group of people respectively to obtain N first color data; S3 includes: inputting N first color data into the light adjustment model to obtain N first light control parameters; S4 includes: controlling the corresponding N landscape lights in the lighting area according to the N first lighting control parameters.

[0009] In a third possible implementation method of the first aspect, the step before S3 further includes: S31: Obtain the movement frequency of tourists within the illumination range of the landscape light; S32: Count the movement frequency to obtain frequency data; The light adjustment model further includes a fourth feature layer, which is used to output stroboscopic parameters according to the frequency data; The third feature layer is used to output a fourth light control parameter according to the stroboscopic parameter and the first light parameter; S4 is replaced by: controlling the landscape light according to the fourth light control parameter.

[0010] In conjunction with the visual detection-based landscape lighting control method provided by the first aspect, the first possible implementation method of the first aspect, the second possible implementation method of the first aspect, or the third possible implementation method of the first aspect, in a fourth possible implementation method of the first aspect, the clothing color group is in the form of a color histogram, the abscissa of the color histogram represents the color type, and the ordinate represents the area of the color; S2 includes: when the number of the color histograms is greater than 1, summing up the multiple color histograms to obtain a total color histogram, and the total color histogram is the first color data.

[0011] In a fifth possible implementation method of the first aspect, S2 includes: The counting of the clothing color group is to calculate the mode of the color component values of the clothing color group of all tourists within the illumination range of the landscape light, to obtain the mode of the color component values, and the mode of the color component values is used as the first color data.

[0012] In combination with the landscape light control method based on visual detection provided in the first aspect, the first possible implementation method of the first aspect, the second possible implementation method of the first aspect, or the third possible implementation method of the first aspect, in a sixth possible implementation method of the first aspect, the large light adjustment model is a convolutional neural network (CNN) with deep learning capabilities; S4 and later also include: S5: Obtain the time the visitor stayed after the landscape light changed; S6: When the visitor's stay time is greater than the time threshold, the visitor's stay time, the corresponding light adjustment model input data, and the light adjustment model output data are used as training samples to train the light adjustment model.

[0013] A second aspect of the present invention provides a landscape lamp for implementing any possible implementation method provided in the first aspect, including: A lighting component for generating light with controllable color and frequency; An information acquisition component, configured to acquire lighting adjustment information, the lighting adjustment information including any one or more of an environment primary color, a position coordinate, and a motion frequency, and a clothing color group; An information processing component, configured to process the light adjustment information and output control parameters for controlling the light assembly according to the processing result; The control component is used to control the lighting component according to the control parameters.

[0014] A third aspect of the present application provides an electronic device, including: processor; and A memory stores executable code thereon, which, when executed by a processor, causes the processor to execute any one of the possible implementation methods provided in the first aspect.

[0015] The fourth aspect of the present application provides a non-temporary machine-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor executes any possible implementation method provided in the first aspect.

[0016] The technical solution provided by this application may have the following beneficial effects: 1. This application first obtains the clothing color group of tourists viewing landscape lights, then performs statistical processing on the obtained clothing color group, and then inputs the first color data obtained by the statistical processing into a pre-trained lighting adjustment model. The lighting adjustment model predicts the tourists' favorite color combinations based on the tourists' clothing colors, and then adjusts lights that can meet the preferences of most tourists based on the favorite color combinations. Finally, the first lighting control parameters are obtained based on the first lighting parameters corresponding to the favorite lights. The landscape lights can be controlled to emit lights that match the color preferences of most tourists according to the first lighting control parameters, thereby realizing real-time lighting adjustment based on user preferences and enhancing the adaptive ability of landscape lights.

[0017] 2. Improve the diversity of landscape lights, and landscape lights can be flexibly controlled by parameters instead of changing them through inherent patterns.

[0018] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and other objects, features and advantages of the present application will become more apparent through a more detailed description of exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0020] Figure 1 1 is a flow chart of a landscape light control method based on visual detection according to an embodiment of the present application; Figure 2 is another flow chart of a landscape light control method based on visual detection according to an embodiment of the present application; Figure 3 is another flow chart of a landscape light control method based on visual detection according to an embodiment of the present application; Figure 4 is another flow chart of a landscape light control method based on visual detection according to an embodiment of the present application; Figure 5 It is a structural diagram of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The preferred embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Instead, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0022] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this application and the appended claims, the singular forms "a," "an," "the," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0023] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0024] In the existing technology, the solution to the problem of single change of landscape lights is mainly to set multiple modes for landscape lights, and switch them manually or through programs at a fixed rhythm according to different scenes and time. Although this method improves the variability of landscape lights to a certain extent and enhances the user experience to a certain extent, it is still a mechanical preset and cannot be changed according to the preferences of tourists, and cannot meet dynamic personalized needs.

[0025] Example 1 The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0026] Figure 1 It is a flow chart of a landscape light control method based on visual detection shown in an embodiment of the present application.

[0027] This embodiment provides a landscape lighting control method, including: S1: Obtain the clothing color groups of tourists within the illumination range of landscape lights; The color group of a tourist's clothing is obtained by setting a sensor that can obtain color information. The sensor can be a multi-spectral camera or a photosensitive sensor. The illumination range of the landscape light varies depending on the type of lamp. It is usually a circular lighting area with a diameter of about 3-15 meters. When more than half of the tourist's body enters the circular lighting area, it is identified as a valid target object. The number of tourists can be one or more. People's clothing consists of multiple parts such as clothes, pants and shoes, so the color of clothing may be a mixture of multiple colors or a single color. Therefore, the color group of a tourist's clothing can refer to a combination of multiple colors or a single color. However, modern clothing is generally a combination of multiple colors, so this embodiment is described as a combination of multiple colors. Each tourist's clothing color corresponds to a clothing color group.

[0028] Preferably, the clothing color group is in the form of a color histogram, where the horizontal axis of the color histogram is the color type and the vertical axis is the color area. Since the clothing color group contains multiple colors and the area occupied by each color is different, a histogram is more intuitive. More specifically, the color histogram is a two-dimensional coordinate system, where the horizontal axis is the color type and the vertical axis is the color area. For example, in the color histogram of a tourist's clothing color group, the horizontal axis is red, blue, and white, and the vertical axis is the area of red, which is 3500cm 2 、The blue area is 2700cm 2 、The white area is 4500cm 2 .

[0029] S2: Count the clothing color groups to obtain first color data; Since the clothing colors of individual tourists under landscape lights are usually mixed with multiple colors, and the number of tourists may be one or more, the clothing color group obtained in S1 may be one or more, and the overall color situation is complicated. In order to obtain the overall situation of the clothing colors of tourists, the obtained clothing color groups are counted in this step to obtain how many colors exist in total, the area of each color, and the proportion of each color.

[0030] Preferably, when a certain color accounts for a small proportion in a clothing color group, its influence on the determination of the favorite color of the tourist corresponding to the clothing color group is very small and can be ignored. Therefore, in order to reduce the amount of calculation, this color can be eliminated from the clothing color group. In specific implementation, a color ratio threshold can be set. Of course, after completing the statistics of all clothing color groups, colors with a small proportion can be eliminated from the statistical results and not included in the prediction of the favorite color. This can also be done by setting a ratio threshold.

[0031] Correspondingly, when the clothing color group is in the form of a color bar graph, S2 is optimized as follows: when the number of color bar graphs is greater than 1, multiple color bar graphs are summed to obtain a total color bar graph; the total color bar graph is the first color data. Because when there is more than one visitor to the landscape light, the parameters of the landscape light need to be determined based on the situation of multiple people. Therefore, when the number of color bar graphs is greater than 1, multiple bar graphs are summed to obtain a total color bar graph. For example, when there are two visitors to the landscape light, the color bar graph of one of them is red and the area is 3500cm 2 、The blue area is 2700cm 2 、The white area is 4500cm 2 The other color bar is blue and the area is 3500cm 2 、The golden area is 2700cm 2 、The gray area is 4500cm 2 The sum of the color bars is red, and the area is 3500cm 2 、The blue area is 6200cm 2 、The white area is 4500cm 2 、The golden area is 2700cm 2 、The gray area is 4500cm 2 The color histogram is the first color data used to input the lighting adjustment model.

[0032] Optionally, when there are a large number of people observing the landscape lights, there will be more types of colors, resulting in a more chaotic first color data, which in turn increases the amount of computation required for the large model for adjusting the lights. Based on cost factors, S2 is optimized to reduce the amount of computation: the color group of clothing is counted to calculate the mode of the color component values of the clothing color group of all tourists within the illumination range of the landscape lights, and the mode of the color component values is obtained; the mode of the color component values is used as the first color data. Specifically, by finding the mode of the component values of the clothing colors of all tourists within the illumination range of the landscape lights, the component value can be the color area. For example, if the area occupied by blue is the largest, blue is used as the mode. The component value can also be the number of tourists corresponding to each color. For example, when the number of tourists is 15, it is found that 12 of the 15 people are wearing blue, 8 are wearing red, and 4 are wearing green. At this time, the area of the color is not considered, and only the number of tourists wearing the color is used as the basis, so the mode is blue. This data is input into the large model for adjusting the lights as the first color data. By calculating the mode of the color component values of the clothing color groups of all tourists within the illumination range of the landscape lights, the mode of the color component values is obtained. The mode of the color component values represents the main color of all clothing color groups and represents the color preferences of most people. The mode of the color component values is used as the first color data to realize color adjustment according to the main color of the clothing color group, which solves the problem of large computational complexity and reduces the training and operation costs of the large lighting adjustment model.

[0033] S3: Input the first color data into the lighting adjustment model to obtain the first lighting control parameter; wherein, the lighting adjustment model includes a first feature layer, a second feature layer and a third feature layer, the first feature layer is used to output the tourists' favorite color data according to the first color data, the second feature layer is used to output the first lighting parameter according to the tourists' favorite color, and the third feature layer is used to output the first lighting control parameter according to the first lighting parameter.

[0034] The lighting adjustment model is a trained artificial intelligence model. Based on the functions of its different computing units, the model is divided into different feature layers. In this embodiment, the model is divided into a first feature layer, a second feature layer, and a third feature layer. The first feature layer converts the first color data (the color of the tourist's clothing) into the tourist's preferred color data. This first feature layer is designed because the color of a tourist's clothing isn't always identical to their preferred color. However, clothing color often has a certain correlation with preferred colors. For example, tourists wearing red tops and blue pants tend to prefer highly saturated and contrasting color combinations, which can be extended to include a favorite color group of red, blue, and white. The second feature layer converts visitors' preferred colors into first lighting parameters. The purpose of the second feature layer is to infer visitors' preferred lighting based on their preferred colors. The visual perception of lighting is determined by multiple factors, such as color, color temperature, brightness, color combination scheme, and change strategy. These factors constitute the first lighting parameters. For example, if a visitor's favorite red, white, and blue color combination is input into the second feature layer, the resulting color combination will be red with blue, with white as the dividing line. The next color combination will be white with red, with blue as the dividing line. The third feature layer converts the first lighting parameters into first control parameters. These are executable instructions (such as voltage and current) for specific landscape lights. Landscape lights are influenced by factors other than color, such as flicker frequency and light change rate. Furthermore, the third feature layer allows the first lighting parameters to be adapted to different landscape light states. Furthermore, while the generation of the first control parameters is based on the first lighting parameters, it also needs to consider the physical constraints of the specific lighting fixture, such as the color gamut, maximum brightness, and lower and upper specification limits.

[0035] The training samples of the lighting adjustment model are obtained through statistical surveys. For example, by statistically surveying people with different clothing and their respective favorite colors, we obtain training samples (clothing color, favorite color) for training the parameters of the first feature layer, namely (first color data, tourist favorite color data); the public's parameter adjustment behavior is collected through online platforms, such as developing a web-based lighting simulator. After the user enters the favorite color, the virtual lighting parameters can be freely adjusted, and then the parameter combination and adjustment process finally selected by the user are stored, so as to obtain training samples (favorite color, favorite lighting) for training the second feature layer, namely (tourist favorite color data, first lighting parameters). Specifically, the first lighting parameters can be understood as a combination of factors that affect the perception of lighting, including light color, color temperature, brightness, etc.

[0036] S4: Control the landscape lights according to the first light control parameters.

[0037] Since the first control parameter is an executable instruction for the landscape light, the control center of the landscape light controls the light changes according to the executable instruction. For example, when the landscape light receives the first control parameter, it changes the light to a red, white and blue color combination, and changes the combination every 10 seconds.

[0038] The beneficial effects of this embodiment are: This application first obtains the clothing color group of tourists watching the landscape lights, then performs statistical processing on the obtained clothing color group, and then inputs the first color data obtained by the statistical processing into a pre-trained lighting adjustment model. The lighting adjustment model predicts the tourists' favorite color combinations based on the tourists' clothing colors, and then adjusts the lights that can meet the preferences of most tourists based on the favorite color combinations. Finally, the first lighting control parameters are obtained based on the first lighting parameters corresponding to the favorite lights. The landscape lights can be controlled to emit lights that match the color preferences of most tourists according to the first lighting control parameters, thereby realizing real-time lighting adjustment according to user preferences and enhancing the adaptive ability of the landscape lights.

[0039] Example 2: In actual operation, landscape lights are generally placed outdoors in different places such as parks, squares, streets, and courtyards. The color tones of different places often vary. If the color of the landscape lights does not match the primary color of the environment, it will lead to a poor experience for tourists. Therefore, how to properly match the color of landscape lights with the primary color of the environment is a technical problem that needs to be solved urgently.

[0040] The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0041] Figure 2 This is another flow chart of a landscape light control method based on visual detection shown in an embodiment of the present application.

[0042] In order to solve this technical problem, this embodiment adds the following steps before S3 of the landscape light control method of the first embodiment: S11: Obtaining the primary color of the environment in which the landscape light is located; A sensor capable of acquiring color information is provided to obtain the primary color of the landscape light's surroundings. This sensor can be the same or different from the sensor used to acquire the clothing color in S1. The landscape light's surroundings include vegetation, roads, and buildings. For example, when the landscape light is in a park, the surrounding trees are green, the roads are gray, and the buildings are blue. The sensor collects these colors and their component values.

[0043] S12: Counting the primary colors of the environment to obtain second color data; Since there is usually more than one primary color under the landscape light, more than one primary color is obtained in S1. The purpose of statistically calculating the primary colors is to integrate multiple primary colors to obtain the total primary color of the landscape light's environment. The total primary color represents the overall condition of the landscape light's environment and is input into the color adjustment model as the second color data. There are many statistical methods, including summation method and weighted average method.

[0044] Correspondingly, the second feature layer of the lighting adjustment model is used to output the second lighting parameters according to the second color data and the tourists' favorite color data.

[0045] The second feature layer can be used to convert the second color data (the total ambient primary color) and the visitor's preferred color data into the second lighting adjustment parameters. For example, if the second color data is 3 green: 2 gray: 1 blue, and the visitor's preferred color data is 3 red: 2 white: 1 blue, although the visitor prefers red, this color clashes with the green primary color in the environment, making the combination glaring and potentially creating a pink color. Furthermore, the visitor's preferred blue color clashes with the ambient blue color. If the lighting color matches the ambient primary color, the lighting will be less noticeable and the viewing experience will be compromised.

[0046] Correspondingly, S4 is replaced by: controlling the landscape light according to the second light control parameter.

[0047] Since the second light control parameter is an executable instruction for the landscape light, the landscape light changes its light according to the executable instruction. For example, when the landscape light receives the second light control parameter, it changes its light to a pink, white, and light blue color combination, and changes the combination every 10 seconds.

[0048] This application first obtains the primary color of the environment in which the landscape lights are located, then statistically processes the obtained primary colors, and then inputs the second color data obtained by the statistical processing into the large light adjustment model. The large light adjustment model mixes lights that are adapted to the environment and meet the preferences of most tourists based on the tourists' favorite color combinations and the primary colors of the environment. Finally, the second light control parameter is obtained based on the second light parameter corresponding to the light switch. The landscape lights can be controlled according to the second light control parameter to emit lights that are adapted to the primary colors of the environment and meet the preferences of most tourists. The influence of the primary colors of the environment on the final presentation of the lights is taken into account, avoiding the incoordination between the lights and the primary colors of the environment, improving the harmony between the lights and the environment, thereby further improving the viewing experience and solving the problem of the incoordination between the lights and the surrounding environment.

[0049] Example 3: In actual operation, some types of landscape lights are large in scale. For example, the landscape lights exhibited in activities such as light festivals have a large number of viewers. If the colors are uniformly adjusted according to the preferences of all tourists, it will be difficult to satisfy everyone.

[0050] The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0051] Figure 3 This is another flow chart of a landscape light control method based on visual detection shown in an embodiment of the present application.

[0052] This embodiment adds the following steps before S3 of the landscape light control method of the first or second embodiment: S21: Obtaining the location coordinates of tourists within the illumination range of the landscape lights; The position coordinates of tourists within the illumination range of the landscape lights are obtained by setting up a sensor that can obtain position coordinates. The sensor can be an ultrasonic sensor or a visual sensor. The illumination range of landscape lights varies depending on the type of lamp. It is usually a circular lighting area with a diameter of about 3-15 meters. A two-dimensional or three-dimensional coordinate system is constructed based on the illumination range. When more than half of the tourist's body enters the circular lighting area, it is identified as a valid target object, and the coordinates of the tourist's location are determined. For example, the illumination range of the landscape light forms a circle with a diameter of 10 meters on the bottom surface. A two-dimensional coordinate system is constructed with the center of the circle as the origin. When the tourist is at the center of the illumination range, his coordinates are (0,0). When the tourist stands 1m to the right of the center of the circle and 1m upward, his coordinates are (1,1).

[0053] S22: Divide the tourists into N groups according to their location coordinates, where N is an integer greater than or equal to 2; The purpose of this step is to divide the people within the illumination range of the landscape lights into different groups according to their location or density. There are many ways to divide tourists into N groups according to location coordinates. You can divide tourists into 4 groups according to the 4 quadrants of the coordinates, or you can divide tourists who are closer to each other into one group. For example, choose the method of dividing tourists who are closer to each other into one group. First, set a division standard according to the size of the illumination range of the landscape lights. When the illumination range of the landscape lights is a circular area with a diameter of 10 meters, the reference value of proximity is less than or equal to 0.3 meters; in addition, set a threshold for the number of people in the group. Only when the number of tourists is greater than or equal to the threshold can they be divided into one group. For example, if the threshold is set to 5 people, then only when the number of people close to each other is greater than or equal to 5 can they be divided into one group. For example, there are 5 tourists with coordinates of (1,1), (0.7,1), (1,0.7), (1.3,1), and (1,1.3). These 5 tourists are divided into the first group. In addition, there are 5 tourists with coordinates of (-1,-1), (-0.7,-1), (-1,-0.7), (-1.3,-1), and (-1,-1.3). These 5 tourists are divided into the second group. It is worth noting that if the coordinates of a tourist (1.3,1.3) are greater than 0.3 from the coordinates (1,1), but are 0.3 from (1,1.3), they can also be divided into the same group. The above method divides the 10 tourists into 2 groups.

[0054] S23: Divide the illumination range of the landscape lights into N light zones according to the position coordinates of the N groups of people, and the same group of people are in the same light zone; In order to ensure that the lighting of the landscape lights in the area where the crowd is located matches the preferences of the majority of the people in the group, it is necessary to divide the illumination range of the landscape lights into light zones equal to the number of people according to the location of the crowd. For example, an edge algorithm can be used to determine the edge of the crowd, and then multiple landscape lights close to the edge are determined to obtain a corresponding landscape light group. The coordinates of each landscape light in the landscape light group are determined, and then the edge of the landscape light is determined based on the coordinates of the multiple landscape lights. This edge is used as the edge of the corresponding light zone to determine the light zone. In addition, if there are still landscape lights that are not divided into a certain light zone after dividing the light zone according to the previous method, the landscape lights are divided into the nearest light zone according to the principle of proximity. In this embodiment, two groups of tourists are divided according to their location coordinates, namely the first group and the second group. The first group corresponds to the first light zone. The light of the first light zone needs to cover everyone in the first group and be larger than the range where the first group is located. For example, the illumination range of the first light zone is a circular area with (1,1) as the center and a radius of 0.4.

[0055] Accordingly, S2 is adjusted as follows: counting the clothing color groups of each group of people respectively to obtain N first color data; To ensure that the lighting in each lighting zone is based on the preferences of the majority of people in the corresponding group, it is necessary to collect data on clothing color groups of different groups and obtain multiple first color data. For example, the clothing color group of the first group is collected to obtain the first color data of 6 parts red, 5 parts green, 4 parts yellow, and 3 parts cyan. The clothing color group of the second group is collected to obtain the second color data of 6 parts green, 5 parts yellow, 4 parts black, and 3 parts orange, where each color occupies an equal area on the clothing.

[0056] Accordingly, S3 is adjusted to: input N first color data into the light adjustment model to obtain N first light control parameters; The landscape lights in each lighting area need to be controlled independently to ensure that the landscape lights in each lighting area are adjusted according to the preferences of the majority of people in the lighting area. Therefore, the multiple first color data obtained based on the clothing color groups of the people in each lighting area are used as multiple inputs and respectively input into the first lighting adjustment model to obtain multiple first lighting control parameters corresponding to the multiple lighting areas. In this embodiment, there are two first color data. The two first color data are input into the lighting adjustment model to obtain two first lighting control parameters. The first color data of the first group of people are input into the lighting adjustment model to obtain the gold, gray and white matching lighting parameters, and the lighting parameters and the position information of the first lighting area are input into the third feature layer to obtain the lighting control parameters with specific position information. The first color data of the second group of people are input into the lighting adjustment model to obtain the orange, yellow and white matching lighting parameters, and the lighting parameters and the position information of the second lighting area are input into the third feature layer to obtain the lighting control parameters with specific position information.

[0057] S4 includes: controlling the landscape lights of the corresponding N light zones respectively according to the N first light control parameters.

[0058] Because the first light control parameters include the location information of the landscape lights, the landscape lights in the corresponding lighting areas are controlled separately by using multiple first light control parameters, so that the landscape lights in the lighting areas present lighting that matches the preferences of most tourists in the lighting areas. In this embodiment, the landscape lights in the corresponding two lighting areas are controlled separately by using two first light control parameters.

[0059] This application first obtains the position coordinates of tourists who are watching the landscape lights, divides the tourists into multiple groups according to their position coordinates, and divides the illumination range of the landscape lights into different light zones according to the areas where the groups are located. Then, by counting the clothing color data of tourists in different groups and inputting it into the light adjustment model for calculation, the lights in the corresponding light zones are controlled based on the preferences of different groups of people, realizing the function of adjusting lights in multiple areas separately. The lights in one area only serve the tourists in that area, making the lights more in line with the preferences of tourists, improving the viewing experience while increasing the diversity of lights, and solving the problem of difficulty in satisfying everyone's tastes due to too many people.

[0060] Example 4: In addition to the function of changing the color of the light, the landscape light also has the function of changing the flashing of the light. Currently, most landscape lights control the light to flash at a set frequency through a preset control program, and the flashing mode is fixed, resulting in a poor user experience. Therefore, this embodiment adds the following steps before S3 of the landscape light control method based on visual detection provided in the previous embodiment, so as to change the flashing frequency of the landscape light according to the movement status of the tourists, thereby further improving the adaptability of the landscape light.

[0061] The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0062] Figure 4 This is another flow chart of a landscape light control method based on visual detection shown in an embodiment of the present application.

[0063] S31: Obtaining the movement frequency of tourists within the illumination range of the landscape lights; Visitors' movement frequency is determined by using a sensor capable of detecting movement frequency. This sensor can be a multispectral camera or a photosensor. For example, a multispectral camera can continuously capture images at a 30-frame rate, analyze the amplitude of changes between consecutive frames, and determine movement frequency based on this amplitude. The illumination range of landscape lights varies depending on the type of lamp, but is typically a circular area with a diameter of approximately 3-15 meters. Visitors are considered valid targets when more than half of their body enters this circular area. The number of visitors can be one or more.

[0064] S32: Counting the movement frequency to obtain frequency data; Since there are typically more than one visitor under a landscape light, more than one motion frequency is obtained in S31. The purpose of statistically analyzing motion frequencies is to integrate multiple motion data to obtain an overall picture of the motion frequencies of all visitors within the illumination range of the landscape light, namely, frequency data. Specifically, frequency data includes frequency and the number of visitors corresponding to each frequency. Motion frequencies include cadence, gesture frequency, joint swing frequency, and blink frequency. Accordingly, a fourth feature layer is added to the large-scale light adjustment model. The fourth feature layer is used to output stroboscopic parameters based on the frequency data. Due to the difference between frequency data and color data, a fourth feature layer is also required in the large-scale light adjustment model to process and analyze the frequency data to determine the stroboscopic parameters used to control the flickering of the landscape lights. Specifically, the flickering frequency of the landscape lights can be determined based on the frequency mode or the frequency band in which the frequencies are concentrated. This is done to match the flickering frequency of the landscape lights with the motion frequencies of the majority of visitors, thereby providing a better viewing experience for visitors. Accordingly, the third feature layer is configured to output fourth light control parameters based on the stroboscopic parameters and the first light parameters. The stroboscopic parameters are added to the light control parameters, and the fourth light control parameters include both color information and stroboscopic information. In this way, the lights are adjusted according to the tourists' favorite colors and their movement frequency, which further enhances the connection between the lights and the tourists' personalities, meets the tourists' personalized needs to a greater extent, and further enhances the viewing experience of lights off.

[0065] S4 is replaced by: controlling the landscape lights according to the fourth light control parameter.

[0066] Since the fourth control parameter is an executable instruction for the landscape light, the landscape light can control the light changes according to the executable instruction, specifically controlling the light color and flashing frequency. For example, when the landscape light receives the first control parameter, it changes the light to a red, white and blue color combination and flashes at a frequency of 1.5Hz (90 times / minute).

[0067] This application first obtains the movement frequency of tourists watching the landscape lights, then performs statistical processing on the obtained movement frequency, and then inputs the frequency data obtained by the statistical processing into a pre-trained light adjustment model. The light adjustment model calculates the stroboscopic parameters that are compatible with the movement frequency of most tourists based on the frequency data, and then combines the stroboscopic parameters with the light parameters to output the fourth light control parameters. The landscape lights are then controlled by the fourth light control parameters to obtain lights that match the tourists' favorite colors and movement frequencies, that is, the flashing frequency of the landscape lights is changed according to the tourists' movement status, thereby enhancing the adaptive ability of the landscape lights.

[0068] Embodiment 5: In actual operation, different artificial intelligence models excel in different fields. The light adjustment model in this application can select different models according to actual conditions. Based on the above embodiment, this embodiment provides a preferred solution for the light adjustment model, which uses a convolutional neural network (CNN) with deep learning capabilities to construct the light adjustment model: Convolutional neural networks are a type of feedforward neural network that includes convolutional calculations and has a deep structure, and can perform deep learning through training data. Deep learning refers to learning the inherent laws and representation levels of sample data, so that machines can have analytical learning capabilities like humans and can recognize text, images, and sound data. Convolutional neural networks are good at processing graphic data. In this embodiment, the data that needs to be processed by the large model is basically graphic data. There are multiple convolutional layers in the convolutional neural network. Convolutional layers at different levels can achieve different functions. The convolutional layers can be divided into different feature layers according to their different functions.

[0069] Furthermore, in order to enable the large lighting control model to have the ability of continuous self-learning and evolution, the following steps are added after S4 in the landscape lighting control method based on visual detection provided in the above embodiment: S5: Obtain the tourist stay time after the landscape lights change; S6: When the tourist's stay time is greater than the time threshold, the tourist's stay time and the corresponding lighting adjustment model input data and lighting adjustment model output data are used as training samples to train the lighting adjustment model.

[0070] To ensure more accurate data output by the large-scale lighting adjustment model, a feedback learning mechanism is also implemented. This sensor, which can be a multispectral camera or a photosensor, measures the length of time visitors spend at a site after a lighting change. The length of time visitors spend at a site after a lighting change indicates their satisfaction with the change. If visitors stop to observe the change, it indicates they are generally satisfied with the change, and vice versa.

[0071] When a visitor's stay time exceeds a threshold, it indicates that the visitor is relatively satisfied with the dimming time. The threshold is a specific duration. For example, if the threshold is 10 seconds, then a visitor's stay time of 10 seconds indicates that the visitor is relatively satisfied. The visitor's stay time, the corresponding dimming model input data, and the output data obtained based on this input data are input into the dimming model to train the model. In short, the excellent output of the dimming model and the corresponding input constitute new training samples for training the dimming model to improve its accuracy.

[0072] Example 6 This embodiment further provides a landscape lamp for implementing the landscape lamp control method based on visual detection provided in the above embodiment, comprising: Lighting components are used to produce light with controllable color and frequency. Lighting components are composed of integrated lighting components. The lighting components can be LED lamp beads or fluorescent lamp beads. Each LED lamp bead and fluorescent lamp bead can produce light of different colors and emit light at different frequencies.

[0073] The information acquisition component is used to obtain lighting adjustment information, which includes any one or more of the following: ambient primary colors, location coordinates, and motion frequency, as well as a clothing color group. The information acquisition component is composed of sensors, which can be multispectral cameras or optical sensors. The number of sensors can be one or more, and can be of a single type or a combination of different types. The above sensors can be installed on or near the landscape lights.

[0074] The information processing component processes the dimming information and outputs control parameters for the lighting components based on the processing results. This component is used to run the large dimming model and typically includes a graphics processing unit (GPU), an application-specific integrated circuit (ASIC), or a field-programmable gate array (FPGA). The information processing component processes the dimming information and outputs control parameters for the lighting components based on the processing results.

[0075] The control component is configured to control the lighting component according to the control parameters. One end of the control component is connected to the information processing component, and the other end is connected to the lighting component. This connection can be electrical or communication. The control component is configured to control the lighting component using the control parameters output by the large lighting control model.

[0076] Embodiment seven: Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated again here.

[0077] Figure 5 It is a structural diagram of an electronic device shown in an embodiment of the present application.

[0078] See also Figure 5 , the electronic device 1000 includes a memory 1010 and a processor 1020.

[0079] The processor 1020 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0080] Memory 1010 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage. ROM may store static data or instructions required by processor 1020 or other computer modules. Permanent storage may be a readable and writable storage device. Permanent storage may be a non-volatile storage device that retains stored instructions and data even when the computer is powered off. In some embodiments, the permanent storage device utilizes a mass storage device (e.g., a magnetic or optical disk, flash memory). In other embodiments, the permanent storage device may be a removable storage device (e.g., a floppy disk, optical drive). System memory may be a readable and writable storage device or a volatile readable and writable storage device, such as dynamic random access memory (DRAM). System memory may store some or all instructions and data required by the processor during operation. Furthermore, memory 1010 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), as well as magnetic disks and / or optical disks. In some embodiments, the memory 1010 may include a readable and / or writable removable storage device, such as a compact disc (CD), a read-only digital versatile disc (e.g., DVD-ROM, double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (e.g., SD card, mini SD card, Micro-SD card, etc.), a magnetic floppy disk, etc. Computer-readable storage media do not include carrier waves and transient electronic signals transmitted wirelessly or wired.

[0081] The memory 1010 stores executable codes. When the executable codes are processed by the processor 1020 , the processor 1020 may execute part or all of the above-mentioned methods.

[0082] The scheme of the present application has been described in detail above with reference to the accompanying drawings. In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. Those skilled in the art should also be aware that the actions and modules involved in the description are not necessarily required for this application. In addition, it is understood that the steps in the method of the embodiment of the present application can be adjusted in order, combined, and deleted according to actual needs, and the modules in the device of the embodiment of the present application can be combined, divided, and deleted according to actual needs.

[0083] In addition, the method according to the present application may also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing some or all of the steps in the above method of the present application.

[0084] Alternatively, the present application can also be implemented as a non-transitory machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) on which executable code (or computer program, or computer instruction code) is stored. When the executable code (or computer program, or computer instruction code) is executed by a processor of an electronic device (or electronic device, server, etc.), the processor executes part or all of the steps of the above-mentioned method according to the present application.

[0085] Those skilled in the art will further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the application herein may be implemented as electronic hardware, computer software, or combinations of both.

[0086] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems and methods according to multiple embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a part of a module, program segment or code, and the part of the module, program segment or code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0087] The embodiments of the present application have been described above. The above description is illustrative and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to the technology in the market, or to enable other persons skilled in the art to understand the embodiments disclosed herein.

Claims

1. A landscape lighting control method based on visual detection, characterized in that: include: S1: Obtain the clothing color groups of tourists within the illumination range of landscape lights; S2: Counting the clothing color groups to obtain first color data; S3: Inputting the first color data into the light adjustment model to obtain first light control parameters; The lighting adjustment model includes a first feature layer, a second feature layer, and a third feature layer. The first feature layer is used to output the tourist's favorite color data based on the first color data. The second feature layer is used to output the first lighting parameter based on the tourist's favorite color. The third feature layer is used to output the first lighting control parameter based on the first lighting parameter. S4: Control the landscape lights according to the first light control parameters.

2. A landscape light control method based on visual detection according to claim 1, characterized in that: S3 previously also included: S11: Obtaining the primary color of the environment in which the landscape light is located; S12: Counting the ambient primary colors to obtain second color data; The second feature layer is used to output second lighting parameters according to the second color data and the tourist's favorite color data; S4 is replaced by: controlling the landscape light according to the second light control parameter.

3. The landscape lighting control method based on visual detection according to claim 1, characterized in that: After S2 and before S3, it also includes: S21: Obtaining the location coordinates of tourists within the illumination range of the landscape lights; S22: Divide the tourists into N groups according to the location coordinates, where N is an integer greater than or equal to 2; S23: Dividing the illumination range of the landscape lights into N lighting zones according to the position coordinates of the N groups of people, with the same group of people being in the same lighting zone; S2 includes: counting the clothing color groups of each group of people to obtain N first color data; S3 includes: inputting N first color data into the light adjustment model to obtain N first light control parameters; S4 includes: controlling the corresponding N landscape lights in the light zones according to the N first light control parameters.

4. The landscape lighting control method based on visual detection according to claim 1, characterized in that: S3 previously also included: S31: Obtaining the movement frequency of tourists within the illumination range of the landscape lights; S32: Counting the movement frequency to obtain frequency data; The light adjustment model further includes a fourth feature layer, and the fourth feature layer is used to output stroboscopic parameters according to the frequency data; The third feature layer is used to output a fourth light control parameter according to the stroboscopic parameter and the first light parameter; S4 is replaced by: controlling the landscape light according to the fourth light control parameter.

5. A landscape lighting control method based on visual detection according to any one of claims 1 to 4, characterized in that: The clothing color group is in the form of a color bar graph, wherein the abscissa of the color bar graph is the color type, and the ordinate is the area of the color; S2 includes: when the number of the color histograms is greater than 1, summing up the multiple color histograms to obtain a total color histogram, where the total color histogram is the first color data.

6. The landscape light control method based on visual detection according to claim 1, characterized in that S2 include: The counting of the clothing color groups is to calculate the mode of the color component values of the clothing color groups of all tourists within the illumination range of the landscape lights, to obtain the mode of the color component values, and the mode of the color component values is used as the first color data.

7. A landscape lighting control method based on visual detection according to any one of claims 1 to 4, characterized in that: The large light adjustment model is a convolutional neural network (CNN) with deep learning capabilities; S4 and later also include: S5: Obtaining the visitor's stay time after the landscape light changes; S6: When the visitor's stay time is greater than a time threshold, the visitor's stay time and the corresponding input data and output data of the lighting adjustment model are used as training samples to train the lighting adjustment model.

8. A landscape lamp, characterized in that: The method for controlling a landscape light based on visual detection according to any one of claims 1 to 7 comprises: A lighting component for generating light with controllable color and frequency; An information acquisition component, configured to acquire lighting adjustment information, wherein the lighting adjustment information includes any one or more of an environment primary color, a position coordinate, and a motion frequency, and a clothing color group; An information processing component, configured to process the lighting adjustment information and output control parameters for controlling the lighting component according to the processing result; A control component is used to control the lighting component according to the control parameters.

9. An electronic device, characterized in that: include: processor; as well as A memory having executable code stored thereon, which, when executed by the processor, causes the processor to execute the landscape lamp control method based on visual detection as described in any one of claims 1 to 7.

10. A non-transitory machine-readable storage medium, characterized in that Executable codes are stored thereon, and when the executable codes are executed by a processor of an electronic device, the processor is caused to execute the landscape light control method based on visual detection as described in any one of claims 1 to 7.

Images