Starting-up direction control method and system, holder equipment and storage medium
By acquiring historical usage data and current environmental parameters, the power-on direction of the gimbal device is automatically adjusted using a machine learning model, solving the problem of low efficiency in manual adjustment by users and improving ease of use and user experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-25
- Publication Date
- 2026-03-24
AI Technical Summary
The current PTZ device requires users to manually adjust the power-on direction, resulting in low efficiency and a poor user experience.
By acquiring the historical usage dataset and current environmental parameters of the device to be powered on, a user habit model is trained using a machine learning model to automatically adjust the power-on direction.
It achieves adaptive power-on orientation adjustment without manual user intervention, improving ease of use and efficiency, and enhancing the user experience.
Smart Images
Figure CN121728356A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of gimbal control, in particular to a start-up direction control method and system, a gimbal device and a storage medium. BACKGROUND
[0002] With the wide application of intelligent gimbal devices in the fields of photography, live broadcast and security, users have increasingly high requirements for the intelligence of gimbal devices. In the prior art, the start-up direction of a gimbal device usually depends on manual adjustment by a user, which is low in efficiency and poor in user experience. SUMMARY
[0003] Therefore, the present application aims to overcome the deficiencies in the prior art and provide a start-up direction control method and system, a gimbal device and a storage medium. The present application provides the following technical solutions. In a first aspect, the present application provides a start-up direction control method, which comprises: obtaining a historical use data set of a device to be started and a current environment parameter, the current environment parameter comprising at least illumination intensity and sound intensity; training a machine learning model based on the historical use data set to obtain a user habit model; inputting the current environment parameter into the user habit model to obtain a target start-up direction; controlling the device to be started to adjust to the target start-up direction when starting.
[0004] In an embodiment, the historical use data set comprises a plurality of groups of historical use data, each group of historical use data comprising historical operation data corresponding to different historical environment parameters, and the training of the machine learning model based on the historical use data set to obtain the user habit model comprises: performing feature extraction on each group of historical operation data to obtain user behavior features corresponding to the different historical environment parameters; training the machine learning model by taking each historical environment parameter as input and the user behavior features corresponding to each historical environment parameter as a label to obtain a user behavior model.
[0005] In an embodiment, before the training of the machine learning model based on the historical use data set to obtain the user habit model, the method further comprises: performing data preprocessing on each group of historical use data in the historical use data set, the data preprocessing comprising determining whether each group of historical use data meets a preset cleaning condition and deleting the historical use data meeting the preset cleaning condition from the historical use data set.
[0006] In an embodiment, the data preprocessing further comprises: performing normalization processing on each of the historical usage data that does not satisfy the preset cleaning condition.
[0007] In an embodiment, before the controlling the to-be-started device to adjust to the target starting direction at startup, the method further comprises: issuing a direction adjustment prompt information, and determining whether a confirmation adjustment signal is received within a preset time; and if the confirmation adjustment signal is not received, determining a default starting direction of the to-be-started device as the target starting direction.
[0008] In an embodiment, after the controlling the to-be-started device to adjust to the target starting direction at startup, the method further comprises: obtaining current operation data of the to-be-started device under the current environmental parameter; and updating the user habit model according to the current operation data under the current environmental parameter.
[0009] In a second aspect, the present application provides a starting direction control system, the system comprising: an obtaining module, configured to obtain a historical usage data set of a to-be-started device and a current environmental parameter, the current environmental parameter at least comprising: illumination intensity and sound intensity; a model training module, configured to train a machine learning model based on the historical usage data set, to obtain a user habit model; a prediction module, configured to input the current environmental parameter into the user habit model, to obtain a target starting direction; a control module, configured to control the to-be-started device to adjust to the target starting direction at startup.
[0010] In an embodiment, the historical usage data set comprises: a plurality of groups of historical usage data, each of the historical usage data comprising: historical operation data corresponding to different historical environmental parameters respectively, and the model training module is further configured to perform feature extraction on each of the historical operation data respectively, to obtain user behavior features corresponding to the different historical environmental parameters respectively. training a machine learning model with each of the historical environmental parameters as input and the user behavior features corresponding to the historical environmental parameters respectively as labels, to obtain a user behavior model.
[0011] In a third aspect, the present application provides a cloud platform device comprising a memory and a processor, the memory storing a computer program, the computer program being executed on the processor to perform the starting direction control method of the first aspect.
[0012] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the start-up direction control method of the first aspect.
[0013] The beneficial effects of the embodiments of the present application are: The present application realizes adaptive adjustment of the start-up direction of the to-be-started device according to user habits, avoids tedious manual intervention, improves the use convenience and adjustment efficiency, and effectively improves the user experience.
[0014] In order to make the above objectives, characteristics and advantages of the present application more apparent and easy to understand, the following preferred embodiments are specifically described below, and the accompanying drawings are referred to for detailed description. BRIEF DESCRIPTION OF DRAWINGS
[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments, and it should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.
[0016] Figure 1 A flowchart of the start-up direction control method provided by the embodiments of the present application is shown; Figure 2 A structural diagram of the start-up direction control system provided by the embodiments of the present application is shown; Figure 3 A structural diagram of the electronic device provided by the embodiments of the present application is shown.
[0017] Main element symbol explanation: 200 - start-up direction control system; 210 - acquisition module; 220 - model training module; 230 - prediction module; 240 - control module; 300 - gimbal device; 301 - transceiver; 302 - processor; 303 - memory. DETAILED DESCRIPTION
[0018] The embodiments of the present application will be described in detail below, and the examples of the embodiments are shown in the drawings, wherein the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and are only used to explain the present application, and cannot be understood as a limitation on the present application.
[0019] In addition, the terms "first", "second", etc. are used only for descriptive purposes and do not connote or imply relative importance or a quantity of the indicated technical features. Thus, features with "first", "second" designations can include one or more of the features explicitly or implicitly. In the description of the present application, the meaning of "a plurality" is two or more, unless otherwise expressly and specifically limited.
[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the template herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.
[0021] Embodiment 1 With the wide application of intelligent gimbal devices in photography, live broadcast, security and other fields, users' requirements for the intelligentization of gimbal devices are increasing. In different use environments, users' use requirements for gimbal devices often differ. The boot direction of traditional gimbal devices needs to be manually adjusted by users according to requirements, which increases the complexity of operation, reduces the use efficiency, and also adversely affects the user experience. To this end, the present application embodiment provides a boot direction control method, and for details, please refer to Figure 1 , which comprises steps S110-S140.
[0022] Step S110, acquire the historical use data set of the device to be booted and the current environment parameters, the current environment parameters at least including: illumination intensity and sound intensity.
[0023] In the present embodiment, the illumination sensor arranged in the device to be booted is used to acquire the illumination intensity of the environment where the device to be booted is located, and the microphone array is used to acquire the sound intensity of the environment where the device to be booted is located. The current environment parameters can also include any one or more of the following parameters: temperature, humidity, geographic location, altitude, etc., which are only used as examples, and can be set according to actual needs.
[0024] The current environment parameters are different, and the use requirements of the user when using the device to be booted will also be different. By collecting multi-dimensional current environment parameters, the use scene characteristics of the device to be booted can be determined, and then the boot direction suitable for the current use requirement can be determined.
[0025] It can be understood that the historical use data set records corresponding historical operation data under different historical environment parameters, and the use demand or use preference of the user under different environment parameters can be analyzed according to the corresponding historical operation data under different historical environment parameters, so as to lay a data foundation for subsequent personalized boot direction prediction and environment self-adaptation adjustment, and reduce manual operation of the user and improve use convenience.
[0026] In step S120, a machine learning model is trained based on the historical use data set to obtain a user habit model.
[0027] The historical use data set of the to-be-booted device includes a plurality of groups of historical use data in a historical use process of the to-be-booted device, and each group of historical use data includes a group of historical environment parameters and corresponding operation data. The operation data includes user historical shooting direction, operation type (for example, up, down or left-right rotation), operation frequency, operation duration, operation timestamp, device holding direction (horizontal or vertical), recording mode of the to-be-booted device, recording state, and the like. The user behavior characteristics such as the most commonly selected shooting direction, operation frequency per unit time, holding direction preference, and recording mode are analyzed by a machine learning algorithm, and the user behavior characteristics are one-to-one corresponding to the historical environment parameters. Each historical environment parameter is taken as an input vector, and the user behavior characteristics corresponding to each historical environment parameter are taken as a label vector, a machine learning model is trained, and a user habit model is obtained.
[0028] It can be understood that based on the historical use data set, the scene behavior rule of the user is learned, so that the user habit model obtained by training can accurately identify the boot direction demand of the user under different environment parameters, and provides reliable support for subsequent rapid prediction of the target boot direction.
[0029] In an embodiment, the historical use data set includes a plurality of groups of historical use data, and each historical use data includes historical operation data corresponding to different historical environment parameters. The training of the machine learning model based on the historical use data set to obtain the user habit model includes: performing feature extraction on each historical operation data to obtain user behavior characteristics corresponding to different historical environment parameters; and taking each historical environment parameter as an input and the user behavior characteristics corresponding to each historical environment parameter as a label to train the machine learning model and obtain a user behavior model.
[0030] In this embodiment, the machine learning model adopts a machine learning algorithm that combines a decision tree algorithm and a K-means clustering algorithm. First, the decision tree algorithm is used to classify and train the input and labels to identify the user's operation mode in a specific environment. Then, the K-means clustering algorithm is used for behavior division training to mine common habits in similar scenarios. Finally, the training results of the two algorithms are fused to obtain a user habit model that can accurately map the correlation between environmental parameters and user boot direction preferences.
[0031] It can be understood that the fusion algorithm of decision tree and K-means clustering realizes the dual adaptation of individual personalization and group scenario, which not only accurately captures the exclusive boot direction preference of a single user in a specific environment through the decision tree algorithm, avoiding the fixed boot mode of traditional devices, but also mines common operation habits in similar scenarios through the K-means clustering algorithm, improving the generalization ability of the model for different use scenarios, so that the device to be booted can not only fit personal use habits, but also quickly adapt to the basic needs of unfamiliar scenarios.
[0032] In an embodiment, before training the machine learning model based on the historical use data set to obtain a user habit model, the method further includes: performing data preprocessing on each historical use data in the historical use data set, the data preprocessing including: respectively judging whether each historical use data satisfies a preset cleaning condition, and deleting the historical use data that satisfies the preset cleaning condition from the historical use data set.
[0033] In this embodiment, before training the machine learning model, a data cleaning operation needs to be performed on each set of historical use data in the historical use data set. The preset cleaning conditions specifically include three types: one is a field missing condition, that is, judging whether the missing proportion of key fields (such as shooting direction, environmental parameter, operation timestamp) in the historical use data exceeds a preset proportion, and if it exceeds, it is determined that the cleaning condition is met; the second is an outlier condition, which calculates the mean and standard deviation of the data using statistical analysis method, judges whether the data deviates from the mean by more than a preset multiple, and if it deviates, it is determined that the cleaning condition is met; the third is a duplicate data condition, which compares the time stamp, environmental parameter, operation data and other core fields of different data, and if they are completely identical, it is determined that the cleaning condition is met. For the historical use data that satisfies any one of the above preset cleaning conditions, it is directly deleted from the data set, and only the valid data with complete fields, reasonable values and no duplicates are retained.
[0034] It can be understood that by removing low-quality data such as invalid records, outliers and duplicates, the model training process is avoided from being disturbed by such data, ensuring that the model training is based on a high-quality and reliable data set, thereby improving the training accuracy of the user habit model from the data source and reducing the prediction error of the model.
[0035] In an embodiment, the data preprocessing further comprises: performing normalization processing on each of the historical use data that does not satisfy the preset cleaning condition.
[0036] In the embodiment, normalization processing is performed on historical environmental parameter data (including historical light intensity, historical sound intensity, etc.) in the effective historical use data that does not satisfy the preset cleaning condition. Linear normalization method is used to uniformly map environmental parameter data of different magnitudes and different units to a preset interval.
[0037] It can be understood that normalization processing enables various types of environmental data to participate in model training in the same dimension, avoids excessive sensitivity of the model to a certain type of parameter due to differences in data magnitudes, and improves the stability and convergence speed of model training.
[0038] In step S130, the current environmental parameter is input into the user habit model to obtain a target starting direction.
[0039] In the embodiment, the current environmental parameter is input into the user habit model trained by fusing the decision tree algorithm and the K-means clustering algorithm, and the target starting direction predicted by the user habit model is output.
[0040] It can be understood that the target starting direction that matches the current environmental parameter and meets the user's use preference is predicted by the user habit model, and the adaptive direction can be quickly output without manual intervention of the user, thereby solving the cumbersome problem of manually adjusting the direction after starting the traditional gimbal device and reducing the waiting time of the user.
[0041] In step S140, the to-be-started device is controlled to adjust to the target starting direction when starting.
[0042] In the embodiment, the angle adjustment instruction containing the target angle parameter and the adjustment rate parameter is generated based on the target starting direction, and the angle adjustment instruction is sent to the to-be-started device to control the to-be-started device to adjust to the target starting direction.
[0043] It can be understood that the automatic adjustment process solves the problem of manually adjusting the direction of the traditional gimbal device after starting, reduces the user's operation intervention, shortens the preparation time of the to-be-started device after starting, and meets the user's demand for fast response of the device in a fast-paced scene.
[0044] In an embodiment, before the to-be-started device is controlled to adjust to the target starting direction when starting, the method further comprises: issuing a direction adjustment prompt information, and determining whether a confirmation adjustment signal is received within a preset time; if the confirmation adjustment signal is not received, the default starting direction of the to-be-started device is determined as the target starting direction.
[0045] In the embodiment, before the to-be-started device is controlled to adjust to the target boot direction, a direction adjustment prompt information is sent through a user interaction interface of the to-be-started device to prompt the user of the target boot direction to be adjusted and the corresponding current environment feature. If no confirmation adjustment signal sent by the user through the interaction interface is received within a preset time, the boot direction adjustment is performed according to the original predicted target boot direction.
[0046] If no confirmation adjustment signal is received within the preset time, a default boot direction set by the to-be-started device at the factory or saved by the user before is automatically called to be determined as the target boot direction for this time.
[0047] In an embodiment, after the to-be-started device is controlled to adjust to the target boot direction, the method further includes: acquiring current operation data of the to-be-started device under the current environment parameter; and updating the user habit model according to the current operation data under the current environment parameter.
[0048] In the embodiment, the user habit model is updated by collecting the operation data after the to-be-started device is started and adjusted to the target boot direction, so that the user habit model can continuously learn new use data, dynamically adapt to the change of the user habit, and ensure the prediction accuracy of the user learning model.
[0049] The boot direction control method provided in the embodiments of the present application acquires a historical use data set of a to-be-started device and a current environment parameter, the current environment parameter at least including light intensity and sound intensity; trains a machine learning model based on the historical use data set to obtain a user habit model; inputs the current environment parameter into the user habit model to obtain a target boot direction; and controls the to-be-started device to adjust to the target boot direction when starting. The to-be-started device realizes self-adaptive adjustment of the boot direction according to the user habit, avoids tedious manual intervention, improves use convenience and adjustment efficiency, and effectively improves the use experience of the user.
[0050] Embodiment 2 In addition, referring to Figure 2 The embodiments of the present application also provide a boot direction control system 200, which includes: The acquisition module 210 is configured to acquire a historical use data set of a to-be-started device and a current environment parameter, the current environment parameter at least including light intensity and sound intensity. The model training module 220 is configured to train a machine learning model based on the historical use data set to obtain a user habit model. The prediction module 230 is configured to input the current environment parameter into the user habit model to obtain a target boot direction. The control module 240 is configured to control the to-be-started device to adjust to the target starting direction at the start.
[0051] In an embodiment, the historical usage data set includes a plurality of groups of historical usage data, each of the historical usage data including historical operation data corresponding to different historical environment parameters respectively, and the model training module 220 is further configured to perform feature extraction on each of the historical operation data to obtain user behavior features corresponding to the different historical environment parameters respectively. The historical environment parameters are taken as inputs, and the user behavior features corresponding to the historical environment parameters are taken as labels to train a machine learning model to obtain a user behavior model.
[0052] The starting direction control system 200 provided by the embodiments of the present application can execute the starting direction control method provided by the method embodiment 1, and thus will not be described here again to avoid repetition.
[0053] Embodiment 3 In addition, the embodiments of the present application provide a gimbal device 300 including a memory 303 and a processor 302, the memory 303 stores a computer program, and the computer program executes the starting direction control method provided by the embodiment 1 when running on the processor 302.
[0054] Specifically, referring to Figure 3 , the gimbal device 300 includes a transceiver 301, a bus interface, and a processor 302, the processor 302 is configured to obtain a historical usage data set of a to-be-started device and a current environment parameter, the current environment parameter at least includes an illumination intensity and a sound intensity, train a machine learning model based on the historical usage data set to obtain a user habit model, input the current environment parameter into the user habit model to obtain a target starting direction, and control the to-be-started device to adjust to the target starting direction at the start.
[0055] In the embodiments of the present application, the gimbal device 300 further includes a memory 303. Figure 3In particular embodiments, the bus architecture can include any number of interconnecting buses and bridges, and the various circuitry representative of the processor 302 and the memory 303, for example, can be linked through a bus architecture or other means as will occur to those of ordinary skill in the art. The bus architecture can also link various other circuitry, which is well known in the art, including, for example, peripheral devices, voltage regulators, and power management circuitry, under the control of the processor 302, thus not further described.
[0056] The gimbal device 300 provided by the embodiment of the application can execute the start-up direction control method provided by the method embodiment 1, and details are not described herein again to avoid repetition.
[0057] Embodiment 4 In addition, the embodiment of the application provides a computer readable storage medium, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the start-up direction control method provided by the embodiment 1.
[0058] In the embodiment, the computer readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and the like.
[0059] The computer readable storage medium provided by the embodiment can implement the start-up direction control method provided by the embodiment 1, and details are not described herein again to avoid repetition.
[0060] In all the examples shown and described herein, any specific values should be interpreted as merely exemplary and not as a limitation, and thus, other examples of the example embodiments can have different values.
[0061] It should be noted that like reference numerals and letters refer to like items in the following drawings, and thus, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings.
[0062] The above-described embodiments only express several implementation manners of the application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the application. It should be pointed out that, for those skilled in the art, without departing from the concept of the application, a number of modifications and improvements can be made, which are within the protection scope of the application.
Claims
1. A method for controlling the direction of power-on, characterized in that, The method includes: Obtain the historical usage dataset and current environmental parameters of the device to be powered on, wherein the current environmental parameters include at least: light intensity and sound intensity; Based on the historical usage dataset, a machine learning model is trained to obtain a user habit model; The current environmental parameters are input into the user habit model to obtain the target boot direction; Control the device to be powered on to adjust to the target power-on direction when starting up.
2. The power-on direction control method according to claim 1, characterized in that, The historical usage dataset includes: multiple sets of historical usage data, each set of historical usage data including: historical operation data corresponding to different historical environmental parameters. The step of training a machine learning model based on the historical usage dataset to obtain a user habit model includes: For each of the historical operation data, feature extraction is performed to obtain user behavior features corresponding to different historical environment parameters; Using the historical environment parameters as input and the user behavior features corresponding to each historical environment parameter as labels, a machine learning model is trained to obtain a user behavior model.
3. The power-on direction control method according to claim 2, characterized in that, Before training the machine learning model based on the historical usage dataset to obtain the user habit model, the process also includes: Data preprocessing is performed on each of the historical usage data in the historical usage dataset. The data preprocessing includes: determining whether each of the historical usage data meets a preset cleaning condition, and deleting the historical usage data that meets the preset cleaning condition from the historical usage dataset.
4. The power-on direction control method according to claim 3, characterized in that, The data preprocessing also includes: The historical usage data that do not meet the preset cleaning conditions are normalized.
5. The power-on direction control method according to claim 1, characterized in that, Before controlling the device to be powered on to adjust to the target power-on direction during startup, the method further includes: It sends a direction adjustment prompt and determines whether a confirmation signal for adjustment is received within a preset time. If the confirmation adjustment signal is not received, the default power-on direction of the device to be powered on will be determined as the target power-on direction.
6. The power-on direction control method according to claim 5, characterized in that, After controlling the device to be powered on to adjust to the target power-on direction during startup, the method further includes: Obtain the current operation data of the device to be powered on under the current environmental parameters; The user habit model is updated based on the current operation data under the current environmental parameters.
7. A start-up direction control system, characterized in that, The system includes: The acquisition module is used to acquire the historical usage dataset and current environmental parameters of the device to be powered on, wherein the current environmental parameters include at least: light intensity and sound intensity; The model training module is used to train the machine learning model based on the historical usage dataset to obtain a user habit model; The prediction module is used to input the current environmental parameters into the user habit model to obtain the target power-on direction; The control module is used to control the device to be powered on to adjust to the target power-on direction when it is started.
8. The start-up direction control system according to claim 7, characterized in that, The historical usage dataset includes: multiple sets of historical usage data, each set of historical usage data includes: historical operation data corresponding to different historical environment parameters. The model training module is also used to extract features from each set of historical operation data to obtain user behavior features corresponding to different historical environment parameters. Using the historical environment parameters as input and the user behavior features corresponding to each historical environment parameter as labels, a machine learning model is trained to obtain a user behavior model.
9. A gimbal device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed on the processor, performs the power-on direction control method according to any one of claims 1-6.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the power-on direction control method according to any one of claims 1-6.