Personalized festival light effect adjusting system based on gesture control and mobile phone application

Abstract

The invention relates to the technical field of intelligent light control, and discloses a personalized festival light effect adjusting system based on gesture control and mobile phone application, which comprises a gesture recognition unit, an instruction mapping unit, a mobile phone application interaction unit and the like. The gesture recognition unit captures user gestures to generate signals, the instruction mapping unit converts the signals into control instructions, and the mobile phone application interaction unit receives user preference data. And the data processing unit fuses the instruction and the preference data, generates multi-dimensional light regulation and control parameters, and drives the LED lamp group. The user behavior learning unit constructs a personalized model update instruction library, the environment sensing unit obtains environment data, and the dynamic adjustment unit corrects light output according to the environment data. The cooperative control unit processes multi-user operation, and the safety monitoring unit guarantees the safety of the system. According to the system, convenient gesture and voice control, personalized customization, environment self-adaption and multi-user cooperation are realized, and the light control experience is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent lighting control, and in particular to a personalized holiday lighting effect adjustment system based on gesture control and mobile phone applications. Background Art

[0002] Lighting plays a crucial role in creating an atmosphere during modern festivals, family gatherings, and various other events. Traditional lighting control methods are relatively simple, relying primarily on manual switches, remote controls, or simple button operations to adjust basic parameters such as brightness and color. This control method is not only cumbersome to operate, but also fails to meet users' growing demand for personalization and convenience.

[0003] In early lighting control systems, users were often limited to simple on / off operations, unable to fine-tune parameters like brightness and color temperature. With technological advancements, remote-controlled lighting devices emerged, improving operational convenience to some extent. However, these systems remained limited in functionality, and remote controls were prone to loss or damage. Later, control systems based on keypads became widely used, allowing users to switch between basic lighting modes with the touch of a button. However, these methods lacked the ability to customize lighting effects.

[0004] With the rapid development of artificial intelligence and mobile internet technologies, people are pursuing a higher level of intelligent and personalized living experiences. In the field of lighting control, users expect to be able to easily adjust lighting effects in a variety of ways, tailored to their preferences and specific scenarios. For example, at a birthday party, users want to be able to quickly adjust the lighting to a warm and romantic atmosphere; while watching a movie, they want to be able to switch the lights to a darker tone with a single click. However, existing lighting control systems cannot meet these personalized needs.

[0005] Furthermore, large-scale events such as concerts and stage performances require precise control of a large number of lighting devices. Traditional control methods are inefficient and difficult to implement complex lighting effects. Furthermore, existing lighting control systems often lack the ability to perceive and adapt to environmental factors, failing to automatically adjust lighting effects based on ambient light intensity, color temperature, and human activity, resulting in energy waste and a poor user experience.

[0006] At the same time, with the increase in multi-user scenarios, such as multiple family members wanting to control the lights at the same time, or multiple people participating in lighting interactions in public places, the existing control system cannot effectively coordinate the operating instructions of different users, which is prone to conflicts and confusion.

[0007] In terms of safety, the existing lighting control system is not perfect in monitoring the working status of lamps, and is unable to detect abnormal conditions such as overheating and overcurrent of lamps in time, posing certain safety hazards. Summary of the Invention

[0008] The object of the present invention is to provide a personalized holiday lighting effect adjustment system based on gesture control and mobile phone applications to solve the problems raised in the above background technology.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a personalized holiday lighting effect adjustment system based on gesture control and mobile phone applications, the system comprising: The gesture recognition unit captures user gestures in real time through a camera array placed in the user's activity area, and generates dynamic gesture signals based on a gesture trajectory feature extraction algorithm; An instruction mapping unit matches the dynamic gesture signal with a preset lighting control instruction library and outputs a corresponding control instruction; A mobile phone application interaction unit provides a user interface for customizing lighting parameters and receives user input of preference data, including brightness gradient, color temperature range, and dynamic mode selection; a data processing unit for fusing the control instructions with the preference data to generate multi-dimensional lighting control parameters; A light driving unit drives the LED light groups in the target area according to the multi-dimensional light control parameters and adjusts their working states; A user behavior learning unit, which collects historical user operation data, builds a personalized preference model based on a neural network, and dynamically updates the lighting control instruction library; The environmental perception unit uses distributed environmental sensors to obtain real-time environmental data of the target area, including ambient light intensity, ambient color temperature, and the position of moving objects; a dynamic adjustment unit, which calculates an environmental adaptation coefficient according to the real-time environmental data and the multi-dimensional lighting control parameters, and performs real-time correction on the output parameters of the LED light group based on the environmental adaptation coefficient; The collaborative control unit, when multiple users operate simultaneously, weights the control instructions of each user according to the priority allocation algorithm and generates comprehensive control instructions; The safety monitoring unit detects the operating current and temperature data of the LED lamp group. If the current and temperature exceed the preset threshold, the protection strategy is triggered, which includes reducing the output power or switching to a backup circuit.

[0010] Preferably, the gesture trajectory feature extraction algorithm includes: S101, performing inter-frame difference processing on continuous images captured by the camera array to extract gesture motion areas; S102, using a convolutional neural network to locate key points in the gesture motion area and generate a gesture skeleton topology map; S103: Calculate the curvature change rate and speed distribution parameters of the gesture trajectory according to the skeleton topology map to form a dynamic gesture signal.

[0011] Preferably, the matching process of the lighting control instruction library includes: S201, performing dynamic time warping similarity calculation on the dynamic gesture signal and the standard gesture template in the instruction library; S202: Select the top N templates with the highest similarity, perform weighted voting based on the contextual information of the user's current operation scenario, and determine the final control instruction.

[0012] Preferably, the fusion processing of the data processing unit includes: S301, fuzzifying discrete instructions in the control instructions to generate continuous instruction vectors; S302: Perform orthogonal projection on the continuous instruction vector and the parameter range in the preference data to obtain an optimal solution in the multi-dimensional control space.

[0013] Preferably, the specific steps of the user behavior learning unit to construct the personalized preference model are: S401: Convert the user's historical operation data into a time series feature sequence and input it into a long short-term memory network for training; S402: Extract key operation nodes through the attention mechanism and generate a user behavior pattern vector; S403: Associating the behavior pattern vector with a lighting control instruction library, and dynamically adjusting the instruction matching weight.

[0014] Preferably, the calculation formula of the environmental adaptation coefficient is:

[0015] Where, is the ambient light intensity, is the target light intensity, is the ambient color temperature, is the target color temperature, is the distance between the moving object and the light group, is the maximum allowed distance, is the weighted coefficient and satisfies .

[0016] Preferably, the priority allocation algorithm includes: S601, determining a basic priority according to a user identity; S602: Generate a dynamic priority weight by combining the user operation frequency and the urgency parameter in the real-time environment data; S603: Map each weight to the interval [0, 1] through normalization processing to generate a comprehensive control instruction.

[0017] Preferably, the protection strategy triggering condition of the security monitoring unit is: When the working current of the LED lamp group is satisfied, or the temperature is satisfied, the protection strategy is activated.

[0018] Preferably, the distributed environmental sensors are deployed in the following manner: Multispectral sensors are installed at the four corners and the center of the target area, and the multispectral sensors communicate with the data processing unit through a wireless ad hoc network.

[0019] Preferably, the mobile application interaction unit also supports voice command input, and uploads voice data to the cloud server through an end-to-end encrypted channel for semantic analysis, and the analysis result is input into the data processing unit after a logical OR operation is performed with the gesture control instruction.

[0020] Compared with the prior art, the present invention has the following beneficial effects: The system provides users with diverse and convenient control methods through a gesture recognition unit and a mobile phone application interaction unit. Users only need to make simple gestures, such as waving or clenching their fists, and the camera array can capture and convert them into dynamic gesture signals in real time to control the lights, breaking away from the constraints of traditional remote controls or button operations, and making operations more natural and smooth. The mobile phone application interaction unit allows users to customize lighting parameters according to their preferences, including brightness gradients, color temperature ranges, and dynamic mode selections. For example, when hosting a Christmas party, users can select a warm color temperature range in the mobile phone application, set a gradient brightness effect, and then combine gesture operations to easily create a lighting environment full of festive atmosphere, meeting the personalized needs of users in different scenarios.

[0021] The user behavior learning unit collects historical user data to build a personalized preference model based on a neural network. As user usage increases, the system continuously learns user habits and preferences, dynamically updating the lighting control command library. This means the system can increasingly accurately understand user intent and provide lighting control services that better meet user expectations. The environmental perception unit and dynamic adjustment unit work together, utilizing distributed environmental sensors to capture real-time environmental data such as ambient light intensity, color temperature, and the position of moving objects. Based on this data, they calculate the environmental adaptation coefficient and adjust the output parameters of the LED light cluster in real time. During daytime hours when ambient light intensity is high, the system automatically reduces light brightness to avoid energy waste. When someone approaches the light cluster, the light automatically adjusts brightness or color to provide better lighting effects, achieving intelligent adaptation between light and environment.

[0022] In scenarios where multiple users are operating simultaneously, the collaborative control unit assigns weights to each user's control commands based on a priority allocation algorithm, generating comprehensive control commands. For example, at a family gathering, different family members may want to adjust the lights at the same time. The system determines a basic priority based on user identity and, taking into account factors such as operation frequency and environmental urgency, rationally coordinates commands from different users to avoid conflicts and ensure that every user's needs are properly met, improving the orderliness and efficiency of lighting control in multi-user scenarios.

[0023] The safety monitoring unit monitors the operating current and temperature of the LED light cluster in real time. Once a preset threshold is exceeded, a protection strategy is immediately triggered, such as reducing output power or switching to a backup circuit. This effectively prevents damage to the lamp due to overheating or overcurrent, which could even lead to safety incidents, ensuring user safety and the lifespan of the lamp.

[0024] The mobile app interaction unit supports voice input, allowing users to issue lighting control commands via voice, further enhancing operational convenience. Voice data is uploaded to a cloud server via an end-to-end encrypted channel for semantic analysis. This analysis, combined with gesture control commands, provides users with a wider range of control options. Each unit in the system utilizes a modular design, offering excellent compatibility and scalability, facilitating integration with other smart devices and systems, and adapting to the future development needs of smart homes and smart scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a working principle diagram of the personalized holiday lighting effect adjustment system of the present invention; Figure 2 This is a step diagram of the gesture trajectory feature extraction algorithm; Figure 3 Workflow diagram for matching lighting control instruction library; Figure 4 This is the workflow diagram for the fusion processing of the data processing unit. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0027] See also Figure 1-4 The present invention provides a technical solution: a personalized holiday lighting effect adjustment system based on gesture control and mobile phone application, the system comprising: Gesture recognition unit: A camera array is placed in the user's activity area to capture the user's gestures in real time. It then generates dynamic gesture signals based on a gesture trajectory feature extraction algorithm.

[0028] Command mapping unit: matches the above-generated dynamic gesture signal with the preset lighting control command library, and then outputs the corresponding control command.

[0029] Mobile App Interaction Unit: This unit provides a user-friendly interface for customizing lighting parameters. It also accepts user input for preferences such as brightness gradients, color temperature ranges, and dynamic mode selection.

[0030] Data processing unit: It integrates the control instructions output by the instruction mapping unit and the preference data received by the mobile application interaction unit, and finally generates multi-dimensional lighting control parameters.

[0031] Light driving unit: drives the LED light group in the target area according to the multi-dimensional light control parameters generated by the data processing unit to adjust the working state of the LED light group.

[0032] User behavior learning unit: By collecting user historical operation data, it builds a personalized preference model based on neural networks and dynamically updates the lighting control instruction library.

[0033] Environmental perception unit: Uses distributed environmental sensors to obtain real-time environmental data of the target area, including information such as ambient light intensity, ambient color temperature, and the position of moving objects.

[0034] Dynamic adjustment unit: Calculates the environmental adaptation coefficient based on the real-time environmental data obtained by the environmental perception unit and the multi-dimensional lighting control parameters generated by the data processing unit, and then makes real-time corrections to the output parameters of the LED light group based on the environmental adaptation coefficient.

[0035] Collaborative control unit: When multiple users operate simultaneously, the control instructions of each user are weighted according to the priority allocation algorithm to generate comprehensive control instructions.

[0036] Safety monitoring unit: detects the operating current and temperature data of the LED light group. If the relevant data exceeds the preset threshold, the protection strategy is triggered, which includes reducing the output power or switching to the backup circuit.

[0037] The present invention will be further described below in conjunction with Examples 1 to 5: Example

[0038] This embodiment mainly describes the specific implementation of the gesture trajectory feature extraction algorithm and the lighting control instruction library matching process, which helps the system to more accurately identify user gestures and convert them into effective lighting control instructions.

[0039] In the gesture recognition unit, the gesture trajectory feature extraction algorithm is specifically performed as follows: Extracting Gesture Motion Areas (S101): The camera array continuously captures images of the user's active area and performs frame-by-frame differencing on these images. This inter-frame differencing highlights the changing areas of the image, allowing the extraction of gesture motion areas. For example, in a scene where a user waves their hand, inter-frame differencing can clearly separate the dynamic area generated by the wave from the background, providing an accurate target area for subsequent processing.

[0040] Keypoint Location and Gesture Skeleton Topology Map Generation (S102): A convolutional neural network is used to process the gesture motion area. Convolutional neural networks have powerful feature extraction capabilities and can accurately locate key points of gestures, such as finger joints and palm edges. Based on these key points, a gesture skeleton topology map is generated. This topology map intuitively displays the structure and shape of the gesture, providing a foundation for the subsequent calculation of gesture trajectory parameters.

[0041] Calculating gesture trajectory parameters to form a dynamic gesture signal (S103): Based on the generated gesture skeletal topology, the curvature change rate and velocity distribution parameters of the gesture trajectory are calculated. During a gesture's motion, the curvature change and velocity distribution of the trajectory can reflect the gesture's intent and characteristics. These parameters are integrated to form a dynamic gesture signal. For example, the dynamic gesture signals generated by a fast and slow hand wave will have significant differences in curvature change rate and velocity distribution parameters.

[0042] In the instruction mapping unit, the matching process of the lighting control instruction library is as follows: Dynamic Time Warping Similarity Calculation (S201): The generated dynamic gesture signal is compared to the standard gesture templates in the instruction library using dynamic time warping (DTU) similarity calculation. The DTU algorithm can effectively handle gestures of varying speeds and rhythms. It finds the optimal matching path between the dynamic gesture signal and the standard gesture template, thereby calculating the similarity between the two. For example, even if different users perform the same gesture at varying speeds, the algorithm can still accurately calculate similarity.

[0043] Weighted voting determines the final control instruction (S202): The top N templates with the highest similarity are selected and weighted voting is performed based on the contextual information of the user's current operation scenario. Contextual information includes the current lighting status and ambient light intensity. For example, if the lights are currently off and the user makes a gesture that resembles turning on the lights, the contextual information will be used to determine the gesture as a light-on instruction, leading to the final accurate control instruction. Example

[0044] In the data processing unit, the fusion processing steps are as follows: Discrete Instruction Fuzzification (S301): Control instructions include some discrete instructions, such as the light on / off command. To better integrate these instructions with preference data, these discrete instructions are fuzzified. Fuzzy logic is used to convert these discrete instructions into continuous instruction vectors. For example, the light on command can be converted into a vector representing the probability of turning on the light within a certain range of values. This allows for unified processing with continuous parameters in the preference data (such as brightness gradient and color temperature range).

[0045] Orthogonal Projection to Obtain Optimal Solution (S302): Orthogonal projection is performed on the continuous command vector and the parameter range in the preference data. In the multidimensional control space, orthogonal projection can be used to find an optimal solution that comprehensively considers the control commands and the user's preference data, generating the final multidimensional lighting control parameters. For example, in a control space that considers both brightness and color temperature, orthogonal projection can be used to determine a brightness and color temperature combination that satisfies the user's brightness preference and meets the requirements of the current control command.

[0046] In the user behavior learning unit, the specific steps for building a personalized preference model are as follows: Data conversion and training (S401): Collect historical user operation data and convert it into a time series feature sequence. This time series feature sequence can reflect the changing patterns of user operations over time. This feature sequence is then fed into a long short-term memory (LSTM) network for training. LSTM has the ability to memorize long-term information and can learn long-term dependencies in user operation patterns. For example, the LSTM can learn user habits regarding lighting effects during different holidays through training.

[0047] Extracting Key Operation Nodes (S402): The trained results are processed using an attention mechanism to extract key operation nodes. The attention mechanism allows the model to focus on the important parts of user operations. For example, in a sequence where the user frequently adjusts the brightness and color temperature of a light, the attention mechanism can identify the operation nodes that have a significant impact on the lighting effect.

[0048] Associating and Adjusting Weights (S403): The extracted user behavior pattern vectors are associated with the lighting control command library, and the command matching weights are dynamically adjusted based on the user behavior pattern vectors. If it is found that the user frequently uses a specific gesture to adjust the lighting to a specific brightness and color temperature combination, the command matching weight associated with this gesture and lighting effect will be increased in subsequent command matching, thereby improving the accuracy of personalized control. Example

[0049] This embodiment mainly describes the calculation method of the environmental adaptation coefficient and the specific implementation process of the priority allocation algorithm. These two aspects are crucial for the system to reasonably adjust the lighting according to environmental changes and multi-user operations.

[0050] In the dynamic adjustment unit, the calculation formula of the environmental adaptation coefficient is:

[0051] Where, is the ambient light intensity, is the target light intensity, is the ambient color temperature, is the target color temperature, is the distance between the moving object and the light group, is the maximum allowed distance, is the weighted coefficient and satisfies .

[0052] In actual applications, the system calculates the environmental adaptation coefficient based on the real-time environmental data obtained by the environmental perception unit. For example, when the ambient light intensity When the brightness is high, in order to maintain visual comfort, it is necessary to reduce the brightness of the LED light group appropriately. The environmental adaptation coefficient calculated by the formula , can be used to adjust the output parameters of the LED light group, such as brightness, color temperature, etc. A larger value indicates that the ambient light intensity has a greater impact on lighting adjustment, and the system will be more inclined to adjust the lighting according to changes in ambient light intensity.

[0053] In the collaborative control unit, the priority assignment algorithm is executed as follows: Determining a basic priority (S601): A basic priority is determined based on the user identity. For example, different permission levels can be set for different users. The basic priority of an administrator user can be set higher, while the basic priority of ordinary users can be set relatively lower. In this way, when multiple users are operating simultaneously, the administrator user's instructions will have higher priority.

[0054] Generating Dynamic Priority Weights (S602): Dynamic priority weights are generated based on the user's frequency of operations and the urgency parameter in the real-time environmental data. Frequent operations by a user indicate a more urgent need for lighting adjustments, and their dynamic priority weight will be increased accordingly. Furthermore, if the real-time environmental data indicates an emergency situation, such as a moving object rapidly approaching the lighting cluster, the dynamic priority weight of user operations related to that area will also be increased.

[0055] Normalization Processing to Generate Comprehensive Control Instructions (S603): Through normalization, each weight is mapped to the interval [0, 1], making the weights of different users comparable. Based on the normalized weights, each user's control instructions are weighted to generate a comprehensive control instruction. For example, if user A has a weight of 0.6 and user B has a weight of 0.4, when generating the comprehensive control instruction, their instructions are integrated according to this weight ratio to ensure that the system can respond appropriately to the operations of multiple users. Example

[0056] This embodiment mainly describes the protection strategy triggering conditions of the security monitoring unit and the deployment method of distributed environmental sensors, which are critical for ensuring the safe and stable operation of the system and accurately obtaining environmental data.

[0057] In the safety monitoring unit, the protection strategy is triggered when the operating current or temperature of the LED light cluster exceeds a preset threshold. The system monitors the operating current and temperature of the LED light cluster in real time. For example, the temperature of the LED light cluster may rise during prolonged use or under high load. When the temperature reaches the preset high temperature threshold, the system triggers a protection strategy to prevent damage to the LED light cluster due to overheating. This protection strategy includes reducing the output power, lowering the brightness of the LED light cluster, thereby reducing power consumption and heat generation; or switching to a backup circuit to ensure the basic operation of the lighting system.

[0058] In the environmental perception unit, distributed environmental sensors are deployed as follows: multispectral sensors are installed at the four corners and the center of the target area. Multispectral sensors can simultaneously acquire spectral information in multiple bands, providing a more comprehensive understanding of ambient light intensity, color temperature, and other information. These multispectral sensors communicate with the data processing unit via a wireless ad hoc network. Wireless ad hoc networks are self-organizing and self-healing. Even if some sensors fail or the communication link is interrupted, the remaining sensors can continue to operate normally and transmit data to the data processing unit. For example, at a large festival venue, by deploying multispectral sensors at the four corners and the center of the venue, environmental data from different locations within the venue can be accurately and in real time, providing a reliable basis for the system to adjust lighting. Example

[0059] This embodiment describes in detail the implementation method of the function of the mobile phone application interaction unit supporting voice command input, which provides users with a more convenient operation method and enriches the control means of the system.

[0060] In addition to supporting user input of preference data through the interface, the mobile app's interaction unit also supports voice command input. When a user uses a voice command, the mobile app uploads the voice data to a cloud server via an end-to-end encrypted channel for semantic analysis. This end-to-end encrypted channel ensures the security of voice data during transmission, preventing theft or tampering. The cloud server possesses powerful voice recognition and semantic analysis capabilities, capable of converting user voice commands into computer-recognizable control instructions. For example, if a user says "turn the light brightness to 50%," the cloud server can accurately interpret the meaning of this command and convert it into the corresponding control instruction.

[0061] The parsing results are logically ORed with the gesture control commands and then fed into the data processing unit. This allows the system to handle all user actions, whether through gesture or voice. For example, if a user simultaneously gestures to increase brightness and speaks the voice command "increase brightness," the data processing unit will combine these two commands to more accurately generate multi-dimensional lighting control parameters, achieving the desired lighting effect.

[0062] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0063] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A personalized holiday lighting effect adjustment system based on gesture control and mobile phone application, characterized in that: include: The gesture recognition unit captures user gestures in real time through a camera array placed in the user's activity area, and generates dynamic gesture signals based on a gesture trajectory feature extraction algorithm; An instruction mapping unit matches the dynamic gesture signal with a preset lighting control instruction library and outputs a corresponding control instruction; A mobile phone application interaction unit provides a user interface for customizing lighting parameters and receives user input of preference data, including brightness gradient, color temperature range, and dynamic mode selection; a data processing unit for fusing the control instructions with the preference data to generate multi-dimensional lighting control parameters; A light driving unit drives the LED light groups in the target area according to the multi-dimensional light control parameters and adjusts their working states; A user behavior learning unit, which collects historical user operation data, builds a personalized preference model based on a neural network, and dynamically updates the lighting control instruction library; The environmental perception unit uses distributed environmental sensors to obtain real-time environmental data of the target area, including ambient light intensity, ambient color temperature, and the position of moving objects; a dynamic adjustment unit, which calculates an environmental adaptation coefficient according to the real-time environmental data and the multi-dimensional lighting control parameters, and performs real-time correction on the output parameters of the LED light group based on the environmental adaptation coefficient; The collaborative control unit, when multiple users operate simultaneously, weights the control instructions of each user according to the priority allocation algorithm and generates comprehensive control instructions; The safety monitoring unit detects the operating current and temperature data of the LED lamp group. If the current and temperature exceed the preset threshold, the protection strategy is triggered, which includes reducing the output power or switching to a backup circuit.

2. The personalized holiday lighting effect adjustment system based on gesture control and mobile phone application according to claim 1 is characterized by: The gesture trajectory feature extraction algorithm includes: S101, performing inter-frame difference processing on continuous images captured by the camera array to extract gesture motion areas; S102, using a convolutional neural network to locate key points in the gesture motion area and generate a gesture skeleton topology map; S103: Calculate the curvature change rate and speed distribution parameters of the gesture trajectory according to the skeleton topology map to form a dynamic gesture signal.

3. The personalized holiday lighting effect adjustment system based on gesture control and mobile phone application according to claim 2 is characterized by: The matching process of the lighting control instruction library includes: S201, performing dynamic time warping similarity calculation on the dynamic gesture signal and the standard gesture template in the instruction library; S202: Select the top N templates with the highest similarity, perform weighted voting based on the contextual information of the user's current operation scenario, and determine the final control instruction.

4. The personalized holiday lighting effect adjustment system based on gesture control and mobile phone application according to claim 3 is characterized by: The fusion processing of the data processing unit includes: S301, fuzzifying discrete instructions in the control instructions to generate continuous instruction vectors; S302: Perform orthogonal projection on the continuous instruction vector and the parameter range in the preference data to obtain an optimal solution in the multi-dimensional control space.

5. The personalized holiday lighting effect adjustment system based on gesture control and mobile phone application according to claim 4 is characterized by: The specific steps of the user behavior learning unit to construct a personalized preference model are: S401: Convert the user's historical operation data into a time series feature sequence and input it into a long short-term memory network for training; S402: Extract key operation nodes through the attention mechanism and generate a user behavior pattern vector; S403: Associating the behavior pattern vector with a lighting control instruction library, and dynamically adjusting an instruction matching weight.

6. The personalized holiday lighting effect adjustment system based on gesture control and mobile phone application according to claim 5 is characterized by: The calculation formula of the environmental adaptation coefficient is: ； Where, is the ambient light intensity, is the target light intensity, is the ambient color temperature, is the target color temperature, is the distance between the moving object and the light group, is the maximum allowed distance, is the weighted coefficient and satisfies .

7. The personalized holiday lighting effect adjustment system based on gesture control and mobile phone application according to claim 6 is characterized by: The priority allocation algorithm includes: S601, determining a basic priority according to a user identity; S602: Generate a dynamic priority weight by combining the user operation frequency and the urgency parameter in the real-time environment data; S603: Map each weight to the interval [0, 1] through normalization processing to generate a comprehensive control instruction.

8. The personalized holiday lighting effect adjustment system based on gesture control and mobile phone application according to claim 7 is characterized by: The protection strategy triggering conditions of the safety monitoring unit are: When the working current of the LED lamp group is satisfied, or the temperature is satisfied, the protection strategy is activated.

9. The personalized holiday lighting effect adjustment system based on gesture control and mobile phone application according to claim 8 is characterized by: The distributed environmental sensors are deployed as follows: Multispectral sensors are installed at the four corners and the center of the target area, and the multispectral sensors communicate with the data processing unit through a wireless ad hoc network.

10. The personalized holiday lighting effect adjustment system based on gesture control and mobile phone application according to claim 9 is characterized in that: The mobile phone application interaction unit also supports voice command input, and uploads voice data to the cloud server through an end-to-end encrypted channel for semantic analysis. The analysis result is logically ORed with the gesture control command and then input into the data processing unit.

Images