Lamp effect configuration method and device, equipment and medium

Abstract

The invention relates to a light effect configuration method and device, equipment and a medium, and the method comprises the steps: obtaining a target image, carrying out the pixel clustering of the target image, obtaining a plurality of pixel clusters and the representative color of each pixel cluster, and enabling the representative color to be the color of the center pixel of the corresponding pixel cluster; determining the area proportion of the total pixel amount of each pixel cluster relative to the total pixel amount of the target image; determining a data distance between each representative color and an average color of the target image to quantify a significance degree of the representative color; and selecting a part of representative color with a relatively high area ratio as a background color of the lamp effect, selecting a part of representative color with a relatively high significance degree from the remaining representative colors as an embellishment color of the lamp effect, and constructing a lamp effect color matching scheme according to the background color and the embellishment color. The problem of extracting the color matching scheme required by the lamp effect from the image is solved by applying a new thought, and the diversified requirements on the lighting system in the modern smart home environment can be met.

Description

Technical Field

[0001] The present application relates to the field of lighting effect control technology, and in particular to a lighting effect configuration method and device, equipment, and medium. Background Art

[0002] In modern smart home systems, lighting equipment has gone beyond the simple lighting function and has become a key factor in creating a comfortable environment and enhancing the living experience. Users' demands for lighting are becoming increasingly diverse. They not only require lights to provide basic lighting, but also hope that lights can adjust colors and dynamic effects according to occasions, atmospheres, and even personal preferences. In order to meet these needs, many smart lighting systems have appeared on the market, which can adjust lighting effects according to preset modes or users' immediate instructions.

[0003] Although these systems have achieved the adjustment of lighting effects to a certain extent, there are still challenges in how to translate the user's environmental or emotional needs into specific lighting effects. In particular, when users want to define lighting effects through specific images, existing systems often lack effective mechanisms to parse the image content and convert it into a coordinated and aesthetically pleasing lighting color scheme. For example, a user may want to set the lighting atmosphere of a room based on a landscape photo or an art painting, but it is difficult for current lighting systems to automatically extract key colors from these images and generate harmonious lighting effects based on them.

[0004] In addition, existing lighting control systems often ignore the psychological impact and visual aesthetics of color when processing images to generate lighting effects, resulting in the generated lighting effects may deviate from user expectations. Users need a more intuitive and intelligent method that can automatically extract key visual elements from images and transform them into lighting color schemes that are emotionally resonant and visually appealing. Summary of the invention

[0005] The primary purpose of the present application is to solve at least one of the above problems and provide a lighting effect configuration method and device, equipment, and medium.

[0006] In order to meet the various objectives of this application, this application adopts the following technical solutions: A lighting effect configuration method, device, equipment, and medium provided to meet one of the purposes of this application include the following steps: Acquire a target image, perform pixel clustering on the target image, obtain multiple pixel clusters and a representative color of each pixel cluster, wherein the representative color is the color of a central pixel of the corresponding pixel cluster; Determine the area ratio of the total number of pixels of each pixel cluster relative to the total number of pixels of the target image; Determine the data distance between each representative color and the average color of the target image to quantify the prominence of the representative color; Select representative colors with a relatively high area ratio as the background color of the lighting effect, select representative colors with a relatively high degree of significance from the remaining representative colors as the decorative colors of the lighting effect, and construct a lighting effect color scheme based on the background color and the decorative color.

[0007] A lighting effect configuration device is proposed to meet one of the purposes of the present application, and includes: A color clustering module is configured to obtain a target image, perform pixel clustering on the target image, obtain multiple pixel clusters and a representative color of each pixel cluster, wherein the representative color is the color of a central pixel of the corresponding pixel cluster; A weight determination module is configured to determine an area ratio of a total number of pixels of each pixel cluster relative to a total number of pixels of a target image; a saliency determination module configured to determine a data distance between each representative color and an average color of the target image to quantify the saliency of the representative color; The scheme construction module is configured to select representative colors with a relatively high area ratio as the background color of the lighting effect, select representative colors with a relatively high degree of significance from the remaining representative colors as the accent colors of the lighting effect, and construct a lighting effect color scheme based on the background color and the accent color.

[0008] On the other hand, a lighting effect configuration device provided to meet one of the purposes of the present application includes a controller, which is communicatively connected to a target lamp, and is used to execute the steps of the lighting effect configuration method to control the target lamp to display the lighting effect.

[0009] On the other hand, a computer-readable storage medium is provided to meet another purpose of the present application, which stores a computer program implemented according to the lighting effect configuration method in the form of computer-readable instructions. When the computer program is called and executed by a computer, the steps included in the corresponding method are executed.

[0010] This application effectively solves the challenges faced by traditional technologies in converting the environment or emotional needs expressed by the target image provided by the user into specific lighting effects. Compared with traditional technologies, this application has obvious technical advantages, including: First, this application can directly extract key color information from images uploaded by users, including background color and accent color. These colors not only capture the visual essence of the image, but also convey the atmosphere and emotion that the image wants to express. In this way, this application can automatically convert the colors in the image into a coordinated and aesthetic lighting color scheme, greatly improving the accuracy and convenience of lighting effect configuration.

[0011] Secondly, the application also takes into account the psychological impact and visual aesthetics of color, ensuring that the generated lighting effects match the user's expectations and enhancing the personalization and dynamic adaptability of the lighting effects. By determining the area proportion and significance of each pixel cluster, the application can give priority to the dominant colors in the image as the background color, and select those colors that are visually attractive and expressive as the embellishment colors, thereby constructing a lighting color scheme with rich layers and visual impact.

[0012] In addition, this application reduces the complexity of user operations through intelligent color extraction and color scheme generation. Users can easily achieve personalized lighting effects without in-depth understanding of technical parameters and interface specifications. This not only improves the overall coordination of lighting effects and optimizes the use of lighting characteristics, but also simplifies the user operation process and enhances the intelligent correspondence of lighting effect semantics, providing users with a richer and more convenient lighting experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 It is a schematic diagram of the principle of an exemplary lighting effect joint control system of the present application, which includes a lighting effect configuration device and a plurality of lamps; Figure 2 A flowchart of a typical embodiment of the lighting effect configuration method of the present application; Figure 3 This is a principle block diagram of the lighting effect configuration device of this application; Figure 4 A schematic diagram of the structure of a computer device used in this application. DETAILED DESCRIPTION

[0014] like Figure 1 As shown, in the structure of the exemplary lighting effect joint control system of the present application, the lighting effect joint control system includes a lighting effect configuration device 80 and a plurality of lamps (82, 84), which include a planar lamp 82 and a non-planar lamp 84. The lighting effect configuration device 80 is responsible for managing and controlling all connected lamps to achieve unified lighting effect control and coordination.

[0015] The lighting effect configuration device 80 is connected to various lamps through wireless communication. This connection method provides flexibility and scalability, allowing the system to easily add or remove lamps without complicated wiring. Through this wireless connection, the lighting effect configuration device 80 can obtain data such as the lamp identification, lamp type, and lighting effect motion mode data set of each lamp. The lamp identification is a unique identifier for each lamp, which is used to distinguish different lamps in the system. The lamp type refers to the physical form and function of the lamp. For example, the planar lamp 82 can include any one or any combination of curtain lights, mesh lights, pixel screens, etc., while the non-planar lamp 84 can include any one or any combination of square lights, linear lights, eaves lights, wall lights, etc. The lighting effect motion mode data set contains one or more lighting effect motion modes preset for the corresponding lamps. These modes define how the lamps display changes in color and brightness.

[0016] The lighting effect configuration device 80 determines the lighting effect corresponding to each lamp according to the interface call information of each lamp and the type of lamp. The interface call information includes the interface name and its corresponding encapsulation format specification, which guides the lighting effect configuration device 80 on how to encapsulate the lighting effect motion mode and color scheme into instructions that the lamp can understand and execute. The lighting effect configuration device 80 can encapsulate the lighting effect motion mode and lighting effect color scheme into a message body, and then construct it into a lighting effect playback instruction, and send it to the corresponding lamp to control it to play the corresponding lighting effect.

[0017] In the lighting effect joint control system, the planar lamps 82 usually have a large light-emitting area and can display continuous images or color gradients, such as curtain lights and mesh lights. These lamps are suitable for creating a wide visual impact and atmosphere rendering. Non-planar lamps 84 usually have smaller light-emitting units, and the color and brightness of each unit can be independently controlled, such as square lights and linear lights. These lamps are suitable for creating rich and varied visual effects.

[0018] The lighting effect configuration device 80 determines the lighting effect of the non-planar lamp 84 by adapting, so as to ensure that the overall lighting effect atmosphere played by each lamp of the lighting effect joint control system is more harmonious and coordinated. This means that regardless of whether the system includes planar lamps 82, the lighting effect configuration device 80 can generate unified lighting effect action semantics and lighting effect color matching semantics according to the input material (such as text or image) and the characteristics of the lamp, so that the lighting effect performance of the entire system is both unified and diverse.

[0019] In some embodiments, the lighting effect configuration device 80 includes a control chip, a control panel, a communication component, and a display screen. The control chip can be implemented using various embedded chips, such as Bluetooth SoC (System on Chip), WiFi SoC, MCU (Micro Controller Unit), DSP (Digital Signal Processing) and other types of chips. The control chip usually includes a central processing unit and a memory, and the memory and the central processing unit are used to store and execute program instructions respectively to achieve corresponding functions. The control panel usually provides one or more buttons for implementing switch control of the controller 1, selecting various preset lighting effects, etc. The communication component is used to achieve wireless communication connection with each lamp. The display screen can be used to display various control information so as to cooperate with the buttons in the control panel to support the implementation of human-computer interaction functions. In some embodiments, the control panel and the display screen can be integrated into the same touch display screen.

[0020] In some embodiments, the lighting effect configuration device 80 of the present application can be implemented in an independent computer device, as long as the computer device is equipped with the corresponding hardware of the lighting effect configuration device 80, and the corresponding business logic of the lighting effect configuration device 80, including the business logic executed by the corresponding method of the present application, is implemented as a computer program installed in the computer device for operation. When the lighting effect configuration device 80 is implemented in a computer device, various resources inherent in the computer device can be shared to save the overall implementation cost. The computer device referred to here can be any terminal device for use by the user, such as a smart phone, a personal computer, a laptop computer, a tablet computer, etc.

[0021] In some embodiments, the lighting effect configuration device of the present application can be directly integrated with the lamp into the same product, that is, the same business logic of the lighting effect configuration device is implemented in the lamp, so that the lamp itself can determine its lighting effect color scheme after acquiring the target image.

[0022] In some embodiments, the lighting effect configuration device may be further equipped with a camera unit, which captures a target image and determines a corresponding lighting effect color scheme based on the target image.

[0023] According to the product architecture and working principle of the above-mentioned lighting effect configuration device 80, the lighting effect configuration method of the present application can be implemented as a computer program, stored in the storage medium of the controller of the lighting effect configuration device 80, and called from the storage medium by the control chip of the controller to run, so as to control each lamp in the entire lighting effect linkage control system to play the corresponding lighting effect.

[0024] The lighting effect joint control system of the present application realizes precise control of various types of lamps through wireless communication connection and intelligent adaptation, so that both planar lamps 82 and non-planar lamps 84 can work together in the same system to provide users with a unified and coordinated lighting effect experience.

[0025] See also Figure 2 In some embodiments, the lighting effect configuration method of the present application includes: Step S5100, acquiring a target image, performing pixel clustering on the target image, obtaining a plurality of pixel clusters and a representative color of each pixel cluster, wherein the representative color is the color of the central pixel of the corresponding pixel cluster; When acquiring the target image, a variety of implementation methods can be flexibly adopted according to different application scenarios. In the scenario where the lighting effect configuration device and the lamp are integrated into the same product, the target image can be captured directly through the built-in camera unit. This integrated design enables the lamp itself to have the ability to acquire images, so that images can be directly acquired from the real world and the lighting effects can be adjusted based on these images. For example, if the lamp is installed outdoors, the outdoor scenery can be directly photographed as the basis for the lighting effect. When the lighting effect configuration device is part of the lighting effect joint control system, the acquisition of the target image may come from the direct input of the user, such as selecting a picture through a smartphone or computer, and then wirelessly transmitting it to the lighting effect configuration device. In addition, the target image can also be acquired through an image library on the Internet or an image uploaded by the user. Of course, no matter what the scenario, the target image can be acquired through any of the methods listed above, as long as the relevant program technical support is implemented in advance.

[0026] After acquiring the target image, pixel clustering can be performed on the target image to achieve image analysis. The purpose is to divide the pixels in the image into multiple groups, namely pixel clusters, according to color similarity.

[0027] In different embodiments, one of the multiple clustering algorithms can be flexibly selected to implement this process. Optional clustering algorithms include but are not limited to K-means algorithm, DBSCAN algorithm, Gaussian mixture model clustering algorithm, etc. K-means clustering algorithm is a commonly used method, which randomly selects K points as initial cluster centers, then iteratively assigns each pixel to the nearest cluster center, and updates the position of the cluster center until the convergence condition is met or the preset number of iterations is reached. In addition, the density-based DBSCAN algorithm can also be used, which determines the clustering results based on the density of the pixel points rather than the preset number of clusters. This method is suitable for processing clustering problems of different shapes and sizes. There is also a model-based Gaussian mixture model (GMM) clustering algorithm, which assumes that the data is a mixture of multiple Gaussian distributions, and the clustering results are determined by estimating the parameters of these distributions.

[0028] In practical applications, the choice of clustering algorithm depends on the specific application scenario and performance requirements. For example, for scenarios that require fast processing of a large number of images, methods with higher computational efficiency can be preferred. For scenarios that require high-precision color analysis, algorithms that can provide more detailed color information can be selected.

[0029] Although these clustering algorithms work in different ways, their common goal is to divide the pixels into multiple clusters and determine the representative color of each cluster, that is, the color of the cluster center. This representative color can be the pixel value of the cluster center or the average color of all pixel values ​​in the cluster. In this way, key color information can be extracted from the target image, providing a basis for subsequent lighting effect color schemes. For example, if the target image is a landscape photo, the clustering algorithm can identify the main colors of different elements such as the sky, grass, and flowers, and use these colors as representative colors for the configuration of lighting effects.

[0030] Step S5200, determining the area ratio of the total number of pixels of each pixel cluster relative to the total number of pixels of the target image; By determining the area ratio of the total number of pixels in each pixel cluster relative to the total number of pixels in the target image, the distribution of each color in the image can be identified, thus providing a basis for the subsequent selection of background and accent colors. Specifically, the area ratio reflects the relative importance of each color in the target image, helps to distinguish between primary and secondary colors, and thus affects the overall atmosphere of the lighting effect.

[0031] In order to calculate the area ratio, we first need to count the number of pixels in each pixel cluster. This can be done by traversing the clustering results and counting the pixels in each cluster. After obtaining the number of pixels in each pixel cluster, divide it by the total number of pixels in the target image to get the area ratio of each pixel cluster.

[0032] In different embodiments, different methods can be used to process pixel data and calculate area percentage. An optional embodiment is to perform calculations directly at the original resolution of the image, so that all detail information of the image can be retained. Another embodiment is to first downsample the image, that is, reduce the resolution of the image to reduce the amount of calculation and increase the processing speed. In the process of downsampling, different algorithms can be used to achieve, such as nearest neighbor interpolation, bilinear interpolation or cubic spline interpolation.

[0033] In some embodiments, pixel clusters corresponding to representative colors close to black in the target image may be additionally excluded, thereby filtering out colors that contribute less to the lighting effect color scheme, because pixel clusters close to black will not have a significant visual impact on the final lighting effect, while also reducing the complexity of subsequent processing.

[0034] Specifically, a preset threshold is introduced. This threshold is used to define what is a "close to black" color. In image processing, black is usually defined as (0, 0, 0) in the RGB color space, but in practical applications, since images may contain shadows or other dark tones, "close to black" colors can be defined as a range, such as colors with RGB values ​​between (0, 0,0) and (20, 20, 20). It should be pointed out that if the pixel cluster is represented based on the LAB color space, then this threshold can also be represented according to the LAB color mode.

[0035] Before calculating the area ratio, all pixel clusters whose representative colors fall within this preset range are identified. Then, the total number of pixels in these pixel clusters is excluded from the total number of pixels to avoid them affecting the final area ratio calculation. In this way, only those pixel clusters with higher brightness and color saturation are taken into account, ensuring that the lighting effect color scheme is more focused on the visually significant parts of the image.

[0036] For example, if the representative color of a pixel cluster is very close to black, and the total number of pixels in the cluster accounts for 5% of the total number of pixels in the entire image, but according to the preset threshold, these pixels are identified as non-visually significant parts, so when calculating the area share, this 5% will not be included. The final area share will be calculated only based on the remaining pixel clusters with higher brightness and color saturation, which can more accurately reflect the main color distribution in the image and provide more meaningful reference data for selecting background colors and accent colors.

[0037] Through this embodiment, the lighting effect color scheme can more accurately reflect the color characteristics of the target image, while avoiding the problem of the color scheme being too dim or monotonous due to the inclusion of a large number of black or dark-tone pixels, thereby improving the overall coordination and visual appeal of the lighting effect.

[0038] By accurately determining the area ratio of each pixel cluster, key data support is provided for building lighting effect color schemes. This is the basis for achieving the conversion from image to lighting effect color, ensuring that the lighting effect can truly reflect the color characteristics of the image and enhance the visual effect and user experience of the lighting effect.

[0039] Step S5300: determining the data distance between each representative color and the average color of the target image to quantify the prominence of the representative color; Determining the data distance between each representative color, especially each representative color in the feature color set and the average color of the target image can quantify the prominence of these representative colors in the color effect of the target image. Therefore, it involves the conversion of color space, the calculation of average color and the quantification of data distance. The purpose is to identify the prominence of each representative color in the image, thereby providing a scientific basis for selecting background color and accent color.

[0040] First, the average color of the target image is determined by counting the color values ​​of all pixels in the image. In the LAB color space, the average color is determined by calculating the average of the L, a, and b values ​​of all pixels in the image. The L value represents brightness, while the a and b values ​​represent the color's counterpart channel, where the a channel ranges from green to red and the b channel ranges from blue to yellow. By calculating the average of these values, a LAB value is obtained that represents the average color of the entire target image.

[0041] Next, the data distance between each representative color and the average color is calculated. This distance can be calculated using a preset data distance formula, and commonly used data distance formulas include but are not limited to Euclidean distance, Manhattan distance, cosine similarity, or CIEDE2000 color difference formula. These formulas can quantify the degree of difference between two color values ​​and can be applied at will. For example, Euclidean distance is obtained by calculating the straight-line distance between two points in the LAB color space, while the CIEDE2000 color difference formula is specifically used to measure the human eye's perception of color differences.

[0042] After calculating the data distance between each representative color and the average color, these distances are normalized to ensure that they are in the same dimension so that the prominence of different colors can be compared. Normalization can be done by dividing the data distance by the maximum value of all data distances or a preset benchmark value, so that the prominence of each color can be represented by a value between 0 and 1. The larger the value, the greater the difference between the color and the average color, and the higher the prominence.

[0043] Based on this, a characteristic color set can be constructed, in which each representative color is accompanied by its area share and prominence. This set provides important information for determining the lighting effect color scheme of the target lamp. For example, if a representative color occupies a large area in the target image and has a high degree of prominence compared to the average color, then this color may be selected as the background color; while those colors with a small area share but high prominence may be selected as accent colors to increase the visual impact of the lighting effect.

[0044] In specific implementations, different algorithms and parameter settings can be used to implement this step. For example, different color spaces (such as RGB, LAB, or HSV) can be selected for calculation. In addition, the normalization method and the threshold of the significance level can be adjusted according to the needs of actual applications to adapt to different images and lighting scenarios. Through these specific implementations, technicians in this field can achieve precise control of the conversion from image to lighting effect color according to specific technical requirements and conditions.

[0045] Step S5400: select representative colors with a relatively high area ratio as the background color of the lighting effect, select representative colors with a relatively high significance from the remaining representative colors as the decorative colors of the lighting effect, and construct a lighting effect color scheme based on the background color and the decorative color.

[0046] Considering that the target lamps usually need to be assigned background colors and accent colors to finely express the lighting effects, the background colors and accent colors can be selected according to the area ratio and prominence of the identified pixel clusters, and then the lighting effect color scheme can be constructed. The key representative colors in the image can be converted into the lighting effects of the lamps, so as to achieve the coordination and unity of the image colors and lighting effects.

[0047] First, from all the representative colors in the characteristic color set, select the representative colors with the highest area as the background color of the lighting effect. The background color plays a dominant role in the lighting effect, and it determines the basic hue of the lighting effect. For example, if the blue sky occupies most of the area in an image, the color representing the blue sky may be selected as the background color. The background color is selected based on its wide distribution in the image, which can provide a harmonious basic color for the lighting effect. The number of background colors can be set as needed.

[0048] Next, from the remaining representative colors of the feature color set except the selected background color, the accent color is selected according to the degree of prominence. Although the distribution area of ​​the accent color in the image is not large, it can provide visual focus and increase color levels in the lighting effect because of its large difference from the average color. For example, a red flower in the image is very conspicuous among the green grass, and its representative color may be selected as the accent color. The selection of the accent color is not only based on its difference from the average color of the image, but also considers its role in enhancing the visual impact of the lighting effect.

[0049] In specific implementations, different strategies can be used to select background colors and accent colors. One possible strategy is to set a fixed area ratio threshold, and representative colors exceeding the threshold are considered as background color candidates, while colors below the threshold but with high significance are considered as accent color candidates. Another strategy may be based on the psychological effects of color, selecting colors that can trigger specific emotional responses as background colors or accent colors.

[0050] In addition, the selection of background and accent colors can be optimized by calculating the complementarity of colors. For example, if the background color is blue, then red or orange as an accent color can provide a stronger visual contrast and enhance the attractiveness of the lighting effect.

[0051] After the background color and the accent color are determined, the lighting effect color scheme can be directly constructed. In some embodiments of constructing the lighting effect color scheme, the permutations and combinations of all representative colors formed by the selected background color and accent color can also be considered. Multiple candidate color sequences can be obtained through permutations and combinations, and then the sequence with the largest sum of color differences between adjacent colors is selected as the final lighting effect color scheme. The purpose of this is to ensure that the transition between colors in the lighting effect is natural while maintaining sufficient visual impact.

[0052] For example, if there are three main colors in an image: blue, yellow, and red, blue occupies the largest area and has a lower degree of significance compared to the average color, so it may be selected as the background color; although yellow and red occupy a smaller area, they are more significant than the average color, so they may be selected as accent colors. When constructing a color scheme, you can calculate the color difference between blue and yellow, blue and red, and yellow and red, and then select the color sequence with the largest total color difference, such as blue-yellow-red, as the color scheme for the lighting effect.

[0053] In one embodiment, after selecting the top three representative colors with the highest area percentage as the background colors, the top five representative colors with the highest significance are selected from the remaining representative colors as the embellishment colors, thereby forming an eight-color color matching table. The construction of a lighting effect color matching scheme based on an eight-color color matching table has unique advantages. This is mainly because such a color matching relationship not only makes the lighting effect show a colorful effect, but also enables each representative color to be effectively distinguished by the user, and therefore is more unique. In this color matching relationship, for the arrangement and combination of 8 colors, each background color can be further placed in a candidate color matching sequence between two embellishment colors, so that each background color and each embellishment color are alternately arranged. For candidate color matching sequences that are not arranged according to this relationship, it is not necessary to calculate the sum of their color differences, and they can be directly deleted.

[0054] Through this meticulous color selection and color scheme construction process, a precise conversion from image to lighting effect can be achieved, so that the lighting effects of the lamps can truly reflect the color characteristics of the image while providing a rich and coordinated visual experience.

[0055] The lighting effect color matching scheme constructed in the present application can be used for various types of lamps. When it is applied to non-planar lamps, it can better match the distribution characteristics of point light sources in non-planar lamps and provide a more effective color matching solution.

[0056] The above embodiments effectively solve the challenges faced by traditional technologies in converting the environment or emotional needs expressed by the target image provided by the user into specific lighting effects. Compared with traditional technologies, the technical advantages of this application are obvious, including but not limited to: First, this application can directly extract key color information from images uploaded by users, including background color and accent color. These colors not only capture the visual essence of the image, but also convey the atmosphere and emotion that the image wants to express. In this way, this application can automatically convert the colors in the image into a coordinated and aesthetic lighting color scheme, greatly improving the accuracy and convenience of lighting effect configuration.

[0057] Secondly, the application also takes into account the psychological impact and visual aesthetics of color, ensuring that the generated lighting effects match the user's expectations and enhancing the personalization and dynamic adaptability of the lighting effects. By determining the area proportion and significance of each pixel cluster, the application can give priority to the dominant colors in the image as the background color, and select those colors that are visually attractive and expressive as the embellishment colors, thereby constructing a lighting color scheme with rich layers and visual impact.

[0058] In addition, this application reduces the complexity of user operations through intelligent color extraction and color scheme generation. Users can easily achieve personalized lighting effects without in-depth understanding of technical parameters and interface specifications. This not only improves the overall coordination of lighting effects and optimizes the use of lighting characteristics, but also simplifies the user operation process and enhances the intelligent correspondence of lighting effect semantics, providing users with a richer and more convenient lighting experience.

[0059] Based on any embodiment of the method of the present application, pixel clustering is performed on the target image, including: Step S5101, converting the target image from the RGB color space to the LAB color space to obtain a human eye field image; The target image is usually represented in RGB color mode. In this case, the target image can be converted from RGB color space to LAB color space so that the LAB color space can better reflect the human eye's perception of color. It divides color information into brightness (L) and two color channels (a and b), where the a channel represents the range from green to red and the b channel represents the range from blue to yellow. This color space conversion is for more accurate color clustering because the LAB color space is closer to the way the human eye perceives color differences.

[0060] In specific implementation, a variety of existing color conversion algorithms can be used to achieve the conversion from RGB to LAB. These algorithms are based on the principles of color science and convert RGB values ​​into LAB values ​​through a series of predefined mathematical formulas. For example, a commonly used conversion method is to complete this process through linear transformation and nonlinear compression, where the linear transformation involves the use of a basis matrix, and the nonlinear compression is to simulate the nonlinear response of the human eye to brightness.

[0061] In the specific implementation, the method defined in the ITU-R BT.2100-1 standard can be used, which provides a detailed conversion process from RGB to LAB color space. In addition, many image processing libraries, such as OpenCV or MATLAB's color conversion toolbox, provide direct function call support, making this conversion process easily integrated into the lighting effect configuration method.

[0062] After conversion to the LAB color space, the target image is represented in the form of brightness and color opponent channels, becoming a human eye visual field image, which helps to more accurately identify and distinguish the colors in the image in subsequent steps.

[0063] Step S5102, randomly determining the number of clusters, applying a preset clustering algorithm to perform multiple iterative pixel clustering on the pixel data in the human eye domain image according to the number of clusters, each clustering forming a corresponding plurality of pixel clusters and a central pixel of each pixel cluster; Determining the appropriate number of clusters in the process of pixel clustering directly affects the accuracy and practicality of the clustering results. If the K value is too small, important color information in the image may be merged and details may be lost; if the K value is too large, it may lead to over-segmentation, increase computational complexity, and may cause unstable clustering results. Therefore, determining the appropriate K value is the key to achieving effective clustering.

[0064] To this end, in one embodiment, the number of clusters K can be determined randomly, allowing the algorithm to automatically adjust the number of clusters according to the complexity and color distribution of the image to obtain the best clustering effect. Specifically, a range of the number of clusters can be preset, such as from 2 to 10, and then a number is randomly selected as the initial number of clusters. This random selection process can be implemented by programming, such as using a random number generator in a computer program to determine the number of clusters.

[0065] In another embodiment, the number of pixel clusters K can also be randomly set according to actual conditions. In practice, the K value can be determined by a variety of methods, such as using the Elbow Method to observe the change in the sum of squared errors within the cluster, and selecting the point where the error decreases slowly as the K value; or using the Silhouette Coefficient to evaluate the clustering effect under different K values, and selecting the K value with the largest Silhouette Coefficient.

[0066] Once the number of clusters is determined, the pixel data in the image can be processed using a preset clustering algorithm. In this embodiment, a K-means clustering algorithm is used to perform this process.

[0067] The K-means algorithm is a widely used clustering algorithm that iteratively assigns data points (pixels) to different clusters. Specifically, the algorithm randomly selects K points as the initial cluster centers at the beginning, and each point represents a cluster. Then, the algorithm assigns each pixel to the nearest cluster center to form K pixel clusters. Subsequently, the algorithm recalculates the center point of each cluster, that is, the mean of all pixels in each cluster, and this mean will be used as the new cluster center. This process is repeated until a specific stopping condition is met, such as the change in the cluster center is less than a certain threshold, or the preset number of iterations is reached.

[0068] In this embodiment, the K-means algorithm is applied to the human visual field image, that is, the target image converted to the LAB color space. In the LAB color space, each pixel is represented by three values, L, a, and b, corresponding to the brightness and two color opponent channels respectively. The algorithm calculates the distance between pixels based on these values ​​and assigns each pixel to the nearest cluster center. This process is repeated until the center point of each cluster no longer changes significantly, or the preset number of iterations is reached, at which time the algorithm stops iterating.

[0069] In practice, the K-means algorithm can be implemented in a variety of programming languages ​​and image processing libraries. For example, in Python, you can use the KMeans function in the SciPy library, or use the kmeans function in MATLAB. These functions provide a direct interface to set the number of clusters K, perform clustering operations, and obtain the cluster label of each pixel and the center point of each cluster.

[0070] By applying the K-means clustering algorithm, multiple pixel clusters and their representative colors can be extracted from the target image. This information will be used to construct the subsequent lighting effect color scheme, thereby achieving accurate mapping of image color to lighting effects.

[0071] Step S5103: selecting the version with the smallest square error in each pixel cluster among all the iterative versions as the valid clustering version, determining each pixel cluster and its central pixel in the valid clustering version, and determining the representative color of the corresponding pixel cluster by the color value of the central pixel; In the pixel clustering process, multiple iterations are performed, so multiple versions of clustering results are generated. Therefore, the quality of the clustering results after each iteration can be evaluated, and the version with the smallest error can be selected as the final clustering result.

[0072] In order to determine the most effective clustering version, it is necessary to calculate the within-cluster sum of squares (WCSS) within each pixel cluster. The within-cluster sum of squares refers to the sum of the squares of the distances from each pixel to its cluster center. The smaller this value is, the tighter the clustering result is, that is, the color of the cluster center can better represent the colors of all pixels in the cluster. Specifically, for each pixel cluster, calculate the square of the Euclidean distance between the LAB value of all pixels in the cluster and the LAB value of the cluster center, and then sum these square values ​​to obtain the within-cluster sum of squares for the cluster. For each iteration, calculate the within-cluster sum of squares for all pixel clusters, and then select the clustering result of the iteration with the smallest within-cluster sum of squares as the effective clustering version.

[0073] After selecting the iteration version with the smallest sum of squared errors as the valid clustering version, the central pixel of each pixel cluster, i.e., the cluster center, can be determined. The color value of this central pixel represents the representative color of the entire pixel cluster, which will be used to determine the background color and accent color of the lighting effect in subsequent steps. For example, if the value of the central pixel of a pixel cluster in the LAB color space is (L*, a*, b*), then this value directly reflects the dominant color of the cluster. This selection mechanism based on minimizing the sum of squared errors ensures the quality and practicality of the clustering results, so that the final lighting effect color scheme more accurately reflects the color characteristics of the target image.

[0074] Step S5104: convert the representation value of each representative color from the LAB color space to the RGB color space.

[0075] After completing pixel clustering and determining the center pixel of each pixel cluster, the color values ​​of these center pixels can be converted from the LAB color space back to the RGB color space. This is because although the LAB color space is conducive to color clustering and analysis, the RGB color space is more universal, and most display devices and lighting systems work based on the RGB model. Therefore, in order to apply the clustering results to the actual lighting effect configuration, it is necessary to apply the corresponding conversion algorithm to convert the color value back to the RGB color space.

[0076] In actual applications, the conversion algorithm receives the LAB value of the center point of each pixel cluster as input and outputs the corresponding RGB value. For example, if the value of the center pixel of a pixel cluster in the LAB color space is (L*, a*, b*), the conversion algorithm will calculate and output the corresponding RGB value (R, G, B). This RGB value can be directly used to control the color of the lamp to achieve a lighting effect that matches the color of the target image.

[0077] This embodiment achieves significant and unique technical advantages compared to other embodiments by converting the target image from RGB color space to LAB color space for pixel clustering.

[0078] First, the LAB color space is closer to the way the human eye perceives color, so that the clustering results can more accurately reflect the human eye's perception of color differences, thereby improving the accuracy and practicality of color clustering. This conversion makes color clustering not only based on the surface differences of color, but also takes into account the psychological visual impact of color, providing more accurate color information for subsequent lighting effect color matching schemes.

[0079] Secondly, by randomly determining the number of clusters and applying a preset clustering algorithm to perform multiple iterative clustering, this embodiment can adapt to images of different complexity and color distribution, and automatically adjust the number of clusters to obtain the best clustering effect. This method avoids the problem of detail loss or over-segmentation that may be caused by a fixed number of clusters, and improves the flexibility and adaptability of clustering. By evaluating the clustering results after each iteration and selecting the version with the smallest error as the final result, this embodiment further ensures the compactness and representativeness of the clustering results, so that the central pixel of each pixel cluster can better represent the color of all pixels in the cluster.

[0080] In addition, the clustering results are converted from the LAB color space back to the RGB color space. This embodiment ensures the practicality and compatibility of the clustering results. This conversion enables the representative colors obtained by clustering to be directly applied to display devices and lighting systems based on the RGB model, achieving seamless connection from image analysis to actual lighting effect configuration. In this way, the lighting effects of the lamps can truly reflect the color characteristics of the target image while providing a rich and coordinated visual experience.

[0081] It can be seen that the technical advantage of this embodiment is that it can provide more accurate, flexible and practical pixel clustering results, which directly support the precise conversion from images to lighting effects, enhance the authenticity and visual effects of lighting effect color schemes, and provide users with a richer and more coordinated lighting experience.

[0082] Based on any embodiment of the method of the present application, determining the data distance between each representative color and the average color of the target image to quantify the prominence of the representative color includes: Step S5210, statistically determine the average color of the target image; Determining the average color of the target image is the basis for quantifying the degree of color significance. In specific implementation, the color values ​​of all pixels in the target image are first counted and the average of these values ​​is calculated. In the LAB color space, determining the average color includes calculating the average of the L, a, and b values ​​of all pixels respectively, and then reconstructing these averages into a representation of the average color.

[0083] In the specific implementation, we first need to traverse each pixel in the target image and extract its L, a, and b values ​​in the LAB color space. Then, we sum the L values ​​of all pixels and divide them by the total number of pixels to get the average brightness value; the same method is used to calculate the average a value and the average b value. In this way, the final average color is defined by these three values, namely (L average, a average, b average).

[0084] Step S5220: Based on the color representation of the LAB color space, calculate the data distance between each representative color and the average color, and normalize all data distances to the same dimension to represent the prominence of each representative color; Determining the data distance between the representative color and the average color of the target image can quantify the color saliency. In the LAB color space, the data distance can be used to represent the difference between each representative color (i.e., the center pixel color of each pixel cluster) and the average color of the target image. Optional data distance algorithms include but are not limited to Euclidean distance, Manhattan distance, cosine similarity, and CIEDE2000 color difference formula.

[0085] Taking Euclidean distance as an example, its calculation formula is the straight-line distance between two points in n-dimensional space. For two points P1(L1, a1, b1) and P2(L2, a2, b2) in LAB color space, the Euclidean distance D can be calculated by the following formula:

[0086] This formula calculates the straight-line distance between two points in the LAB color space, reflecting the degree of difference between the two colors. The CIEDE2000 color difference formula is specifically used to measure the human eye's perception of color differences and is suitable for scenes that require accurate simulation of human eye perception.

[0087] In actual calculation, based on the average color of the target image, for each representative color, the data distance between it and the average color is calculated using the above formula D These distance values ​​reflect the degree of difference between each representative color and the average color of the entire image, thereby quantifying the saliency of each color.

[0088] In order to make these distance values ​​comparable in the same dimension, normalization is required. Normalization is to scale the data so that it falls into a small specified interval. Optional but not limited normalization methods include minimum-maximum normalization, whose formula is:

[0089] in, and The normalized data distance value will be between 0 and 1, which is convenient for comparing the significance of different representative colors.

[0090] Step S5230: construct each representative color and its area ratio and prominence into a characteristic color set for determining a lighting effect color scheme of a target lamp.

[0091] After determining the area proportion and prominence of each representative color, a characteristic color set can be constructed to achieve a comprehensive combination of representative colors, their area proportions and prominence, which is used to ultimately determine the lighting effect color scheme of the target lamp. The characteristic color set combines the spatial distribution and visual prominence of colors in the image, providing a set of representative colors for the lamp, which can reflect the main visual features and emotional atmosphere of the image.

[0092] Each entry in the feature color set contains three key pieces of information: the LAB value representing the color, the area ratio of the color in the image, and the data distance from the average color (i.e., the degree of significance). For example, if the central color of a pixel cluster is (L*, a*, b*), the color occupies 15% of the area in the image, and the Euclidean distance from the average color is 5, then this color will be recorded as an entry in the feature color set.

[0093] When constructing a feature color set, there are many ways to process and select colors. One method is to directly use the center colors of all pixel clusters as representative colors, which can reflect the main color distribution in the image. Another method is to select only those colors whose area ratio exceeds a certain threshold, or select those colors whose significance is higher than a certain standard, to ensure that the selected colors are sufficiently representative and visually influential in the image.

[0094] In the process of constructing the characteristic color set, some optimization strategies can be used to further improve the effect of color selection. For example, the concept of color harmony can be introduced to select colors that are coordinated with each other on the color wheel to enhance the aesthetic effect of the lighting effect. In addition, the psychological effect of color can be considered, and colors that can trigger specific emotional responses can be given priority, such as warm colors for creating a comfortable atmosphere and cool colors for creating a calm atmosphere.

[0095] In short, the construction of the characteristic color set is a process that comprehensively considers color distribution, visual prominence and psychological effects. It provides a scientific basis for the determination of the lighting effect color scheme, ensuring that the lighting effect of the lamp can truly reflect the color characteristics of the target image while providing a rich and coordinated visual experience. In this way, accurate conversion from image to lighting effect can be achieved, enhancing the visual effect and user experience of the lighting effect.

[0096] This embodiment provides a more detailed and scientific method to determine the importance and visual impact of colors in an image by accurately counting the average color of the target image and quantifying the prominence of each representative color. Compared with other embodiments, the unique technical advantage of this method is that it can more realistically reflect the color distribution and visual focus of the image, thereby improving the accuracy and expressiveness of the lighting effect color scheme.

[0097] First, by statistically averaging the LAB values ​​of all pixels in the target image, this embodiment can obtain an average color that fully represents the overall color characteristics of the image. Compared with methods that rely only on individual pixels or preset color schemes, this color statistics based on the entire image can more accurately capture the color center trend of the image, providing a solid foundation for subsequent color comparison and selection.

[0098] Secondly, this embodiment quantifies the prominence of the color by calculating the data distance between each representative color and the average color. This quantification process not only takes into account the absolute difference between colors, but also, through normalization, allows the prominence of different colors to be compared on the same dimension. Such processing makes the color selection process more objective and standardized, and helps to identify colors with higher visual prominence in the image, so that when these colors are reproduced in the lighting effect, they can produce a stronger visual impact.

[0099] Finally, the step of constructing a characteristic color set further enhances the technical advantage of this embodiment. By comprehensively considering the area proportion and prominence of the colors, this embodiment can screen out the most representative and influential colors in the image, which are not only dominant in quantity but also the most prominent visually. Applying these colors to the lighting effect color scheme can ensure that the lighting effect of the lamp is highly consistent with the target image in color, while also highlighting the visual focus of the image and enhancing the emotional expression and aesthetic value of the lighting effect.

[0100] On the basis of any embodiment of the method of the present application, constructing a lighting effect color scheme according to the background color and the accent color includes: Step S5410, determining the data distance between each pair of colors in each representative color as the color difference between the two colors; Based on the characteristic color set consisting of background color and accent color obtained by color feature extraction based on the target image, the color difference between them is determined by analyzing the colors in the characteristic color set. The colors in the characteristic color set can accurately reflect the user's intention and preference for color, thereby constructing a lighting effect color scheme that meets the user's expectations.

[0101] Specifically, the feature color set contains a set of selected representative colors that represent the main color features of the target image. In order to quantify the relationship between these colors, the concept of data distance is further introduced to represent the color difference between any two representative colors. Color difference refers to the degree of difference between two colors, which can be determined by calculating the data distance between two color values. In color space, this distance can be calculated in a variety of ways, such as using Euclidean distance, Minkowski distance, cosine similarity, CIELAB color difference formula, etc. Preferably, the calculation of the data distance corresponding to the color difference can be performed in the LAB color space to more accurately quantify the color difference effect of the human eye.

[0102] In order to more intuitively represent the color difference between these colors, we can use a distance matrix to show it. The distance matrix is ​​a two-dimensional array, where the rows and columns represent different colors in the feature color set, and each element in the matrix represents the color difference between the corresponding two colors. By calculating the color difference between each pair of colors and filling it into the matrix, we get a complete distance matrix, which describes in detail the relationship between all colors in the feature color set.

[0103] Step S5420: Arrange and combine the representative colors to obtain multiple candidate color matching sequences; Based on the characteristic color set and the color difference relationship between each representative color, these colors can be arranged and combined to generate multiple candidate color matching sequences. The candidate color matching sequences are used to explore the visual effects of different color combinations so as to select the best lighting effect color matching scheme.

[0104] In the specific implementation, we first need to determine the number of colors contained in the characteristic color set, and then use the method of permutation and combination to generate all possible candidate color sequences. For example, if the characteristic color set contains three colors A, B and C, then the possible color matching sequences include ABC, ACB, BAC, BCA, CAB, CBA. These sequences represent the order in which the colors are displayed on the lamps, and different sequences may bring different visual effects.

[0105] Step S5430: Calculate the sum of the color differences between each two adjacent colors in each candidate color matching sequence, and determine the candidate color matching sequence with the largest sum of color differences as the lighting effect color matching scheme of the non-planar lamp.

[0106] After determining the characteristic color set and its color difference relationship, we can continue to calculate the sum of the color differences between two adjacent representative colors in each candidate color sequence to determine the best lighting color scheme. The core of this step is to evaluate the visual impact and coordination of each candidate color sequence, and select the sequence with the largest sum of color differences as the final lighting color scheme, because a larger sum of color differences usually means a more vivid and dynamic visual effect.

[0107] Specifically, for each candidate color sequence, treat it as a color arrangement, such as the sequence ABC. In this sequence, it is necessary to calculate the color difference between colors A and B (denoted as d(A,B)), and the color difference between colors B and C (denoted as d(B,C)). These color difference values ​​can be directly obtained from the distance matrix constructed previously. Then, add these two color difference values ​​to get the total color difference of the sequence ABC.

[0108] For all candidate color matching sequences, the above process is repeated to calculate the sum of the color differences of each sequence. For example, for the sequence ACB, d(A,C) and d(C,B) are calculated and added to get the sum of the color differences of the sequence ACB. Similarly, the sum of the color differences of all other sequences such as BAC, BCA, CAB and CBA is also calculated.

[0109] After the color difference sum of all candidate color matching sequences is calculated, the color difference sum values ​​of these sequences are compared, and the candidate color matching sequence with the largest color difference sum is selected as the lighting effect color scheme of the non-planar lamp. The reason for this is that the sequence with a larger color difference sum provides a stronger contrast and change visually, which is more likely to attract the attention of the observer and create a more vivid and interesting lighting effect atmosphere.

[0110] For example, suppose the sum of color differences of the following candidate color matching sequences is obtained: the sum of ABC is 10, the sum of ACB is 12, the sum of BAC is 9, the sum of BCA is 11, the sum of CAB is 8, and the sum of CBA is 14. Among these sequences, CBA has the largest sum of color differences, so CBA is selected as the final lighting effect color scheme.

[0111] This embodiment achieves a series of unique technical advantages by accurately calculating the color difference between representative colors and determining the lighting effect color scheme of the lamp based on this. These advantages not only improve the visual effect of the lighting effect, but also enhance the user experience.

[0112] First, this embodiment quantifies the differences between colors by constructing a distance matrix, making the color matching more scientific and accurate. This data-based method ensures that the color matching scheme is not just based on subjective judgment, but has a clear mathematical basis, thereby improving the professionalism and reliability of the color matching scheme.

[0113] Secondly, this embodiment calculates the sum of the color differences of all possible color combinations and selects the sequence with the largest sum of color differences as the final lighting color scheme, which can maximize the visual effect. In this way, the lamp can show more vivid and dynamic color changes, providing users with a richer and more attractive visual experience.

[0114] Furthermore, this embodiment is technically highly flexible and adaptable. It is not limited to a specific color space or color difference calculation method, and the most suitable color space and color difference calculation formula can be selected according to different needs and conditions. This flexibility enables this embodiment to be widely used in different scenarios and needs, and has strong practicality.

[0115] In addition, this embodiment has obvious advantages in improving the overall coordination of lighting effects, optimizing the use of lighting fixture characteristics, simplifying user operation processes, and enhancing the intelligent correspondence between lighting effect semantics and target images. By intelligently generating a lighting effect color scheme corresponding to the target image input by the user, this embodiment not only improves the personalization of lighting effects, but also enables lighting effects to dynamically adapt to different scenes and user needs, providing users with a more flexible and intelligent lighting solution.

[0116] On the basis of any embodiment of the method of the present application, after constructing a lighting effect color scheme according to the background color and the accent color, the method includes: Step S6100: input the target image into a large language model to generate a content description text describing the image content of the target image; Based on the target image, a large language model can be used to generate text describing the content of the target image, i.e., content description text. The large language model is an artificial intelligence technology that can understand and generate natural language text. It predicts and generates text by analyzing input data (image data in this application).

[0117] In order to enable the large language model to generate the content description text required for this application with high quality based on the target image, it can be fine-tuned in advance. Fine-tuning training means further training with a specific data set based on the original pre-training of the model to adapt to specific tasks or application scenarios. In this application, the purpose of fine-tuning training is to enable the model to understand and describe the image content more accurately so as to generate description text that matches the image semantics.

[0118] In the specific implementation, we first collect and prepare a dataset containing images and their corresponding content description text. This dataset will be used to train and fine-tune the large language model so that it can understand and generate descriptions related to the image content. Through training, the model learns how to extract key information from the image and convert it into text descriptions. For example, if the input image is a beach at sunset, the trained model can generate a description text such as "golden beach at sunset".

[0119] After fine-tuning training is completed, the large language model can receive the target image as input and output text describing the content of the image. For example, if the target image is a landscape painting, the large language model that has been fine-tuned can accurately identify elements such as mountains, water, and trees in the image and generate descriptive text "A tranquil landscape painting with mountains and water." Such a descriptive text not only captures the main visual elements of the image, but also conveys the overall atmosphere and emotion of the image.

[0120] In this way, the large language model provides semantic supplement to the lighting effect color scheme, so that the lighting effects of the lamps not only match the target image in color, but also coordinate with the image content in dynamic performance, thus providing users with a richer and more coordinated lighting experience.

[0121] Step S6200: calling the description information in the lighting effect motion mode data set of the target lamp, and determining the lighting effect motion mode corresponding to the description information that is closest in semantics to the content description text; When this step is executed, the lighting effect color scheme has been constructed based on the background color and the accent color, and the content description text corresponding to the target image has been generated through the large language model. This content description text actually carries the semantics of the lighting effect action, so the lighting effect motion mode dataset of the target lamp can be called, and the lighting effect motion mode corresponding to the description information closest to the semantics of the content description text can be determined. Based on this, the semantic understanding of the image content is matched with the dynamic performance of the target lamp to achieve a more personalized and emotional lighting effect.

[0122] As mentioned above, each target lamp is associated with a lighting effect motion mode dataset, which contains multiple preset lighting effect motion modes, and each mode has corresponding description information. These description information is a textual expression of the lighting effect motion characteristics, such as "flowing", "flashing" or "stationary", which defines how the lamp displays color and brightness over time. Since the content description text generated by the large language model has a stronger expressive function, it is more effective to match the content description text with the description information.

[0123] In order to achieve this matching process, it is first necessary to compare the content description text generated by the large language model with the description information of each lighting effect motion mode in the lighting effect motion mode dataset. This comparison can be achieved through keyword matching, semantic similarity calculation or other natural language processing technologies. For example, if the content description text is "golden beach under the sunset", the system may look for lighting effect motion mode description information related to "sunset", "golden" and "beach", such as "warm", "gradient" or "tranquil".

[0124] During the matching process, the semantic similarity between the content description text and each lighting effect motion mode description information can be evaluated, and the lighting effect motion mode corresponding to the description information with the closest semantic similarity can be selected. This selection can be based on preset matching rules, such as selecting the mode with the highest keyword overlap or the highest semantic similarity score. For example, if the semantic similarity between "golden beach under sunset" and the description information of "warmth" and "gradual change" is the highest, the corresponding lighting effect motion mode is selected.

[0125] In addition, the selection of lighting effect motion mode can also take into account the physical characteristics of the lamp and the usage scenario. For example, for a smart lamp that can display complex color changes, a dynamically changing lighting effect motion mode can be selected; while for a simple ambient light, a more static mode may be selected.

[0126] Step S6300: Send the determined lighting effect motion mode and lighting effect color scheme to the target lamp, so that the target lamp applies the lighting effect motion mode and lighting effect color scheme to play the corresponding lighting effect.

[0127] After constructing a lighting effect color scheme based on the background color and the accent color and determining the lighting effect motion mode that is closest to the semantics of the image content description text, these determined settings can be sent to the target fixture so that the target fixture can play the corresponding lighting effect according to these settings.

[0128] In some embodiments, the signal may be sent to the target luminaire through the following specific process: First, in order to ensure that the determined lighting effect motion mode and lighting effect color scheme can be correctly understood and executed by the target lamp, the interface call information of the target lamp is obtained in advance. The interface call information includes the interface name and its corresponding encapsulation format specification, which guides how to encapsulate the lighting effect motion mode and color scheme into instructions that the lamp can understand and execute. The interface call information can be obtained through a pre-set protocol manual or identified through an automatic discovery process with the lamp.

[0129] After obtaining the interface call information, the lighting effect motion mode and color scheme are encapsulated according to the encapsulation format specification to form the corresponding message body. For example, if the lighting interface requires receiving data in JSON format, the parameters of the lighting effect motion mode and color scheme need to be organized into a JSON object containing all necessary key-value pairs. During the encapsulation process, the specific parameters of the lighting effect motion mode and color scheme, such as color value, brightness level, change speed, etc., are converted into the specified format and arranged according to the requirements of the interface.

[0130] After the message body is correctly encapsulated, the lighting effect playback instruction is constructed, combining the interface name and the message body to form a complete lighting effect playback instruction. This instruction contains all the information required for the target lamp to execute the lighting effect, such as the specific parameters of the motion mode, color value, brightness level, etc.

[0131] Finally, the lighting effect playback command is sent to the corresponding target lamps through the wireless communication network. After the target lamps receive the lighting effect playback command, they will execute the corresponding lighting effect according to the data in the command. This means that the lamps will display light according to the predetermined lighting effect motion mode and color scheme, thereby rendering the expected lighting effect atmosphere in the user's environment.

[0132] In different embodiments, the interface call information and encapsulation format specifications may vary. For example, some lamps may support RESTful APIs, in which case the instructions may need to be sent in the form of HTTP requests, including a specific URL and HTTP method (such as GET or POST). Other lamps may use the MQTT protocol for communication, in which case the encapsulation and sending of instructions need to follow the MQTT specification. Regardless of the specific communication protocol, the core goal is to ensure that the lighting effect motion mode and color scheme can be correctly understood and executed by the lamp to achieve precise lighting effect control.

[0133] By precisely docking the interface call information and packaging format specifications of the target lamps, highly customized lighting effect motion modes and color schemes are transmitted, ensuring seamless communication with the lamps and accurate execution of instructions. Compared with other embodiments, the unique technical advantage of this embodiment lies in its strong adaptability and compatibility. It can cope with the interface differences of lamps of different brands and models. Through flexible packaging and instruction construction mechanisms, the same lighting effect configuration method can be widely used in a variety of lamps, greatly expanding the scope of application. In addition, by ensuring the integrity and correctness of information, the accuracy and reliability of lighting effect control are improved, and the risk of lighting effect configuration failure due to communication errors or instruction misunderstandings is reduced, thereby providing users with a more stable and efficient intelligent lighting solution.

[0134] This embodiment combines the content description text generated by the large language model and the light effect motion pattern dataset to achieve a deep integration of the light effect color scheme and the light effect motion pattern, providing a richer and more dynamic lighting experience. Compared with other embodiments, the unique technical advantage of this method is that it can convert the semantic understanding of the image content into specific lighting effects, thereby achieving seamless integration of image content and lighting effects.

[0135] First, by using the content description text generated by the large language model, this embodiment can capture the main visual elements and overall atmosphere of the target image, and convert these abstract concepts into specific text descriptions. This text description not only contains the key information of the image, but also conveys the emotion and atmosphere of the image, providing rich semantic information for the dynamic performance of the lighting effect. In this way, the lighting effect of the lamp is no longer limited to static color display, but can be dynamically adjusted according to changes in image content, providing a more vivid and personalized lighting experience.

[0136] Secondly, this embodiment realizes intelligent selection of lighting effect motion mode by calling the lighting effect motion mode data set of the target lamp and determining the lighting effect motion mode corresponding to the description information that is closest to the semantics of the content description text. This selection is not only based on the semantic similarity of the image content, but also takes into account the physical characteristics and usage scenarios of the lamp, so that the selected lighting effect motion mode can match the characteristics of the lamp and the application environment, improving the adaptability and coordination of the lighting effect.

[0137] Finally, the determined lighting effect motion mode and lighting effect color scheme are sent to the target lamp. This embodiment ensures that the lamp can play the corresponding lighting effect according to these settings. This transmission process not only takes into account the compatibility of the lamp interface and the differences in communication protocols, but also realizes highly customized lighting effect control through precise encapsulation and instruction construction mechanisms. This precise control not only improves the execution efficiency of the lighting effect, but also reduces the risk of lighting effect configuration failure caused by communication errors or instruction misunderstandings, providing users with a more stable and efficient intelligent lighting solution.

[0138] Based on any embodiment of the method of the present application, acquiring a target image includes: Step S5111, responding to a user input event, determining the original input material of the user, where the original input material of the user is a target image; The user provides the original input materials through specific input events. These materials are the target images. They can be image files uploaded directly by the user or photos taken by the user through smart devices. When the user triggers the user input event, the lighting effect configuration device responds to the user input event and starts to receive and process the target image input by the user.

[0139] The target image provided by the user can be any form of static image, including but not limited to landscape photos, art paintings, user personal pictures, etc. The content of the target image will directly affect the generation of the subsequent lighting effect color scheme, so the quality, color and theme of the image are all factors that the system needs to consider.

[0140] In order to ensure that the system can accurately understand and process these images, the format of the target image usually needs to meet certain technical requirements. For example, the image file may need to be in a common format such as JPEG, PNG, etc. to ensure that the system can recognize and parse it. In addition, the image can also be pre-processed according to actual needs, such as adjusting the image size, format conversion, etc., to meet the needs of subsequent processing flows.

[0141] For example, if there are planar lamps in the system, the target image can be converted into a pixelated image and adapted according to the resolution and display capability of the planar lamp. The adapted pixelated image will be sent to the planar lamp application as a lighting effect to achieve direct display of the image content. For non-planar lamps, a specific lighting effect color scheme can be generated according to the content of the target image to control the lamp to display the corresponding color and dynamic effects.

[0142] Step S5112, identify the type of lamp in the lamp joint control system, convert the target image into a pixelated image, send the pixelated image as a lighting effect of a planar lamp type to the planar lamp application, and use the target image to generate lighting effects for a target lamp of a non-planar lamp type in the lamp joint control system.

[0143] In the lighting effect joint control system, after the target image is successfully acquired and preprocessed, it can also be processed accordingly according to the type of lamps in the lighting joint control system. The types of lamps in the lighting joint control system can include planar lamp types and non-planar lamp types. Planar lamps, such as curtain lights or mesh lights, have a large luminous area and can display continuous images or color gradients; while non-planar lamps, such as square lights or linear lights, are composed of smaller light-emitting units and are suitable for displaying discrete colors and dynamic effects.

[0144] For planar luminaires, the target image needs to be converted into a pixelated image to fit the display capabilities of the luminaire. To do this, each pixel of the target image can be mapped to the corresponding pixel of the luminaire to ensure that the image can be accurately displayed on the luminaire. During the conversion process, the resolution of the planar luminaire must first be identified, that is, the number of rows and columns of pixels that the luminaire can display. Then, the size of the target image is adjusted according to this resolution to match the display capabilities of the luminaire. If the resolution of the target image is higher than the resolution of the luminaire, the image needs to be downsampled; if it is lower, appropriate enlargement or padding operations may be required. After the size adjustment, each pixel of the image will directly correspond to the corresponding pixel of the planar luminaire, realizing the direct display of the image content.

[0145] For non-planar lamps, the target image is used to generate a specific lighting effect color scheme according to the embodiments disclosed above in this application to determine a lighting effect color scheme suitable for the non-planar lamps.

[0146] In specific implementations, a variety of image processing techniques can be used to achieve the conversion of the target image into a pixelated image and the generation of a lighting effect color scheme. For example, an image scaling algorithm can be used to adjust the image size, a color extraction algorithm (such as the clustering algorithm mentioned above) can be used to analyze the image color features, and a dynamic effect generation algorithm can be used to create dynamic lighting effects for non-planar lamps. The application of these technologies ensures that the target image can be effectively converted into lighting effects suitable for different types of lamps, thereby achieving a unified and coordinated lighting effect.

[0147] This embodiment obtains the target image by directly responding to the user's input event, and performs differentiated processing according to the type of lamps in the lamp joint control system, thereby achieving a more personalized and efficient lighting effect configuration. Compared with other embodiments, the unique advantage of this embodiment is that it can directly extract key visual elements from the image uploaded by the user, and generate a matching lighting effect color scheme and dynamic effect according to the specific type and display capability of the lamp. For planar lamps, by converting the target image into a pixelated image and adapting its resolution, the direct display of the image content is achieved, providing intuitive visual feedback to the user; while for non-planar lamps, the color and dynamic characteristics of the image are used to generate a specific lighting effect color scheme and determine the corresponding lighting effect motion mode, so that the colors and dynamic effects displayed by various types of lamps in the lighting effect joint control system are coordinated with the image content input by the user. This differentiated processing not only improves the flexibility and accuracy of the lighting effect configuration, but also enhances the user experience of the lighting system, so that the lighting effects of the lamps can truly reflect the user's environmental or emotional needs, while providing a rich and coordinated visual experience.

[0148] Based on any embodiment of the method of the present application, acquiring a target image includes: Step S5121, responding to a user input event, determining the original user input material, where the original user input material is a content description text of an image; The user can trigger a user input event by providing the content description text of the image he or she expects. The lighting effect configuration device can determine the user's original input material, that is, the content description text, in response to the user input event. This text can be a few simple sentences that outline the scene or theme corresponding to the lighting effect that the user wants the lamp to display. For example, the user may enter descriptive text such as "Summer Beach Party" or "Warm Family Gathering", which provides intuitive guidance for the lighting effect configuration of the lamp.

[0149] Step S5122: call the large language model to generate a corresponding target image based on the original input material of the user, use the target image to generate a lighting effect color scheme for the target lamp, use the content description text to determine the lighting effect motion mode of the target lamp, and define the lighting effect applied by the target lamp by the lighting effect color scheme and the lighting effect motion mode.

[0150] After the content description text is determined, a large language model can be used to generate a corresponding target image based on the content description text provided by the user, and the image can be used to define the lighting effect color scheme and lighting effect motion mode of the lamp.

[0151] First, after the user provides their original input material through content description text, the large language model is called to understand the semantic content of these texts. Through its pre-trained language understanding ability, the large language model can identify key words and concepts in the text. For example, "summer beach party" may be identified as a scene containing elements such as sunshine, beach, and waves. This understanding enables the model to generate an image based on the text description. This image becomes the target image, which can visually represent the scene or theme described in the text.

[0152] The generated target image is then used by the embodiments disclosed above in this application to determine the lighting effect color scheme of the lamp, thereby extracting the main colors that can represent the atmosphere and emotion of the image, and applying these colors to the lighting effect color scheme of the target lamp.

[0153] At the same time, the content description text is also used to determine the lighting effect motion mode of the target lamp. For example, "warm family gathering" may be matched with soft, gradual lighting effects, while "summer beach party" may be matched with bright, dynamic lighting effects. In this way, the appropriate lighting effect motion mode can be selected according to the emotion and atmosphere in the text description.

[0154] This embodiment provides an innovative method to meet the user's personalized needs for lighting effects by directly converting the user's text description into the lighting effect configuration of the lamp. Compared with other embodiments, the unique advantage of this embodiment is that it can directly extract intentions and emotions from the user's text description without the user providing a specific image file. Through the semantic understanding ability of the large language model, the abstract text description can be converted into a specific visual effect, a target image matching the description can be generated, and the lighting effect color scheme and motion mode of the lamp can be determined accordingly. This method not only simplifies the user's operation process and lowers the threshold for use, but also greatly expands the diversity and creativity of lighting effects. The user only needs to provide a simple text description to automatically generate the corresponding visual and dynamic effects, so that the lamp can show a richer and more coordinated lighting atmosphere, and improve the flexibility and interactivity of the user experience. In addition, this method also enhances the connection between the lighting effect and the user's emotion, making the lamp not only a lighting tool, but also an intelligent device that can express and enhance specific emotions and atmospheres.

[0155] See also Figure 3 A lighting effect configuration device provided to meet one of the purposes of the present application is a functional embodiment of the lighting effect configuration method of the present application, the device includes a color clustering module 5100, a weight determination module 5200, a dominance determination module 5300, and a scheme construction module 5400, wherein the color clustering module 5100 is configured to obtain a target image, perform pixel clustering on the target image, obtain multiple pixel clusters and a representative color of each pixel cluster, and the representative color is the color of the central pixel of the corresponding pixel cluster; the weight determination module 5200 is configured to determine the area ratio of the total number of pixels of each pixel cluster relative to the total number of pixels of the target image; the dominance determination module 5300 is configured to determine the data distance between each representative color and the average color of the target image to quantify the prominence of the representative color; the scheme construction module 5400 is configured to select a portion of representative colors with a relatively high area ratio as the background color of the lighting effect, select a portion of representative colors with a relatively high degree of prominence from the remaining representative colors as the decorative colors of the lighting effect, and construct a lighting effect color scheme according to the background color and the decorative color.

[0156] Based on any embodiment of the device of the present application, the color clustering module 5100 includes: a color conversion module, which is configured to convert the target image from the RGB color space to the LAB color space to obtain a human eye visual field image; a clustering iteration module, which is configured to randomly determine the number of clusters, and apply a preset clustering algorithm to perform multiple iterative pixel clustering on the pixel data in the human eye visual field image according to the number of clusters, each clustering forms a corresponding plurality of pixel clusters and a central pixel of each pixel cluster; a version selection module, which is configured to select the version with the smallest square error in each pixel cluster among all the iterative versions as the valid clustering version, determine each pixel cluster and its central pixel in the valid clustering version, and determine the representative color of the corresponding pixel cluster by the color value of the central pixel; a color restoration module, which is configured to convert the representation value of each representative color from the LAB color space to the RGB color space.

[0157] Based on any embodiment of the device of the present application, the prominence determination module 5300 includes: a uniform color determination module, configured to statistically determine the average color of the target image; a degree quantification module, configured to calculate the data distance between each representative color and the average color based on the color representation of the LAB color space, and normalize all data distances to the same dimension to represent the prominence of each representative color; a data association module, configured to construct each representative color and its area proportion and prominence as a characteristic color set for determining the lighting effect color scheme of the target lamp.

[0158] Based on any embodiment of the device of the present application, the scheme construction module 5400 includes: a distance calculation module, configured to determine the data distance between each pair of colors in each representative color as the color difference between the two colors; an arrangement and combination module, configured to arrange and combine each representative color to obtain multiple candidate color matching sequences; a color difference selection module, configured to calculate the sum of the color differences between each adjacent two colors in each candidate color matching sequence, and determine the candidate color matching sequence with the largest sum of color differences as the lighting effect color matching scheme of the non-planar lamp.

[0159] On the basis of any embodiment of the device of the present application, after the scheme construction module 5400, the device of the present application includes: a text generation module, configured to input the target image into a large language model to generate a content description text describing the image content of the target image; a mode determination module, configured to call the description information in the lighting effect motion mode data set of the target lamp, and determine the lighting effect motion mode corresponding to the description information that is closest in semantics to the content description text; a lighting effect application module, configured to send the determined lighting effect motion mode and lighting effect color scheme to the target lamp, so that the target lamp can apply the lighting effect motion mode and lighting effect color scheme to play the corresponding lighting effect.

[0160] Based on any embodiment of the device of the present application, the color aggregation module includes: an image acquisition module, which is configured to respond to user input events and determine the user's original input material, which is the target image; an image processing module, which is configured to identify the type of lamps in the lamp linkage control system, convert the target image into a pixelated image, and send the pixelated image as a lighting effect of a planar lamp type to the planar lamp application, and use the target image to generate lighting effects for target lamps of non-planar lamp types in the lamp linkage control system.

[0161] Based on any embodiment of the device of the present application, the color aggregation module includes: a text acquisition module, which is configured to respond to user input events and determine the user's original input material, where the user's original input material is a content description text of an image; a generation application module, which is configured to call a large language model to generate a corresponding target image based on the user's original input material, use the target image to generate a lighting effect color scheme for a target lamp, and use the content description text to determine a lighting effect motion mode of the target lamp, so that the lighting effect applied by the target lamp is defined by the lighting effect color scheme and the lighting effect motion mode.

[0162] In order to solve the above technical problems, the present application also provides a computer device. Figure 4 As shown, a schematic diagram of the internal structure of a computer device. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. Among them, the computer-readable storage medium of the computer device stores an operating system, a database, and computer-readable instructions. The database may store a control information sequence. When the computer-readable instructions are executed by the processor, the processor can implement a lighting effect configuration method. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device may store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can execute the lighting effect configuration method of the present application. The network interface of the computer device is used to connect and communicate with a terminal. Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0163] In this embodiment, the processor is used to execute Figure 3The memory stores the program code and various data required to execute the above modules or submodules. The network interface is used to transmit data between user terminals or servers. The memory in this embodiment stores the program code and data required to execute all modules / submodules in the lighting effect configuration device of this application, and the server can call the program code and data of the server to execute the functions of all submodules.

[0164] The present application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the lighting effect configuration method of any embodiment of the present application.

[0165] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments of the present application can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a computer-readable storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0166] It will be understood by those skilled in the art that the various operations, methods, steps, measures, and schemes in the processes discussed in this application may be alternated, changed, combined, or deleted. Furthermore, other steps, measures, and schemes in the various operations, methods, and processes discussed in this application may also be alternated, changed, rearranged, decomposed, combined, or deleted. Furthermore, the steps, measures, and schemes in the various operations, methods, and processes in the prior art that are open source and in this application may also be alternated, changed, rearranged, decomposed, combined, or deleted.

[0167] The above description is only a partial implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A lighting effect configuration method, characterized in that: include: Acquire a target image, perform pixel clustering on the target image, obtain multiple pixel clusters and a representative color of each pixel cluster, wherein the representative color is the color of a central pixel of the corresponding pixel cluster; Determine the area ratio of the total number of pixels of each pixel cluster relative to the total number of pixels of the target image; Determine the data distance between each representative color and the average color of the target image to quantify the prominence of the representative color; Select representative colors with a relatively high area ratio as the background color of the lighting effect, select representative colors with a relatively high degree of significance from the remaining representative colors as the decorative colors of the lighting effect, and construct a lighting effect color scheme based on the background color and the decorative color.

2. The lighting effect configuration method according to claim 1, characterized in that: Perform pixel clustering on the target image, including: Convert the target image from RGB color space to LAB color space to obtain the human eye field image; The number of clusters is randomly determined, and a preset clustering algorithm is used to perform multiple iterative pixel clustering on the pixel data in the human eye domain image according to the number of clusters, and each clustering forms a corresponding plurality of pixel clusters and a central pixel of each pixel cluster; The version with the smallest square error in each pixel cluster among all the iterative versions is selected as the valid clustering version, each pixel cluster and its central pixel in the valid clustering version are determined, and the representative color of the corresponding pixel cluster is determined by the color value of the central pixel; Convert the color values ​​from LAB color space to RGB color space.

3. The lighting effect configuration method according to claim 1, characterized in that: Determine the data distance between each representative color and the average color of the target image to quantify the prominence of the representative color, including: Statistically determine the average color of the target image; Based on the color representation of the LAB color space, the data distance between each representative color and the average color is calculated, and all data distances are normalized to the same dimension to represent the prominence of each representative color; Each representative color and its area proportion and prominence are constructed into a characteristic color set to determine the lighting effect color scheme of the target lamp.

4. The lighting effect configuration method according to claim 1, characterized in that: Constructing a lighting effect color scheme according to the background color and the accent color, including: Determine the data distance between two colors in each representative color as the color difference between the two colors; Arrange and combine the representative colors to obtain multiple candidate color matching sequences; The sum of the color differences between each two adjacent colors in each candidate color matching sequence is calculated, and the candidate color matching sequence with the largest sum of color differences is determined as the lighting effect color matching scheme of the non-planar lamps in the lamp joint control system.

5. The lighting effect configuration method according to claim 1, characterized in that: Select representative colors with a relatively high area as the background color of the lighting effect, and select representative colors with a relatively high degree of significance from the remaining representative colors as the decorative colors of the lighting effect, including: Select the top three colors with the largest area as the background colors of the lighting effect; The top five colors with the highest significance are selected from the remaining representative colors as the decorative colors of the lighting effect.

6. The lighting effect configuration method according to any one of claims 1 to 5, characterized in that: After constructing the lighting effect color scheme according to the background color and the accent color, it includes: Inputting the target image into a large language model to generate a content description text describing the image content of the target image; Calling description information in the lighting effect motion mode data set of the target lamp, and determining the lighting effect motion mode corresponding to the description information that is closest in semantics to the content description text; The determined lighting effect motion mode and lighting effect color scheme are sent to the target lamp, so that the target lamp applies the lighting effect motion mode and lighting effect color scheme to play the corresponding lighting effect.

7. The lighting effect configuration method according to claim 6, characterized in that: Get the target image, including: In response to a user input event, determining an original user input material, where the original user input material is a target image; Identify the type of lamp in the lamp joint control system, convert the target image into a pixelated image, send the pixelated image as a lighting effect of a planar lamp type to the planar lamp application, and use the target image to generate lighting effects for a target lamp of a non-planar lamp type in the lamp joint control system.

8. The lighting effect configuration method according to any one of claims 1 to 5, characterized in that: Get the target image, including: In response to a user input event, determining an original user input material, where the original user input material is a content description text of the image; A large language model is called to generate a corresponding target image according to the original input material of the user, the target image is used to generate a lighting effect color scheme of the target lamp, and the content description text is used to determine the lighting effect motion mode of the target lamp, so that the lighting effect applied by the target lamp is defined by the lighting effect color scheme and the lighting effect motion mode.

9. A lighting effect configuration device, characterized in that: include: A color clustering module is configured to obtain a target image, perform pixel clustering on the target image, obtain multiple pixel clusters and a representative color of each pixel cluster, wherein the representative color is the color of a central pixel of the corresponding pixel cluster; A weight determination module is configured to determine an area ratio of a total number of pixels of each pixel cluster relative to a total number of pixels of a target image; a saliency determination module configured to determine a data distance between each representative color and an average color of the target image to quantify the saliency of the representative color; The scheme construction module is configured to select representative colors with a relatively high area ratio as the background color of the lighting effect, select representative colors with a relatively high degree of significance from the remaining representative colors as the accent colors of the lighting effect, and construct a lighting effect color scheme based on the background color and the accent color.

10. A lighting effect configuration device, comprising a controller, wherein the controller is in communication connection with a target lamp, characterized in that: The controller is used to execute the steps of the lighting effect configuration method as described in any one of claims 1 to 8 to control the target lamp to display the lighting effect.

11. A computer-readable storage medium, characterized in that: It stores a computer program implemented according to the method described in any one of claims 1 to 8 in the form of computer-readable instructions, and when the computer program is called and executed by a computer, the steps included in the corresponding method are executed.

Images