Scene intelligent adjustment method and device, equipment and storage medium

Abstract

Embodiments of the present application relate to the technical field of information processing, and disclose a scene intelligent adjustment method and device, equipment and a storage medium, comprising: periodically identifying a user posture appearing in a place; judging whether an abnormal posture meeting a preset abnormal posture condition exists in the identified user posture; and when it is determined that the identified user posture includes an abnormal posture, intelligently adjusting a scene in the place where the abnormal posture appears. The scene intelligent adjustment method, device, equipment and storage medium disclosed by the present application solve the problem that the existing method requires a user to manually select a scene and set a trigger condition, which is inconvenient to use and cannot adjust the scene according to the actual use, so that the scene in the place is intelligently adjusted based on the user posture, without the need for manual selection of the scene or setting of a scene adjustment trigger instruction, and the scene is adjusted according to the actual situation, thereby effectively improving the convenience and flexibility of the user in adjusting the scene.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of information processing technology, and in particular to a method, apparatus, device and storage medium for intelligent scene adjustment. Background Technology

[0002] Currently, the smart home industry is developing rapidly, using residences as a carrier and leveraging next-generation information technologies such as the Internet of Things, cloud computing, big data, and artificial intelligence to achieve interconnection and interoperability between system platforms and home appliances.

[0003] In some technologies, smart home scenarios are either preset default scenarios or user-created scenarios. However, using scenarios requires users to manually select scenarios and set trigger conditions, which is inconvenient and makes it impossible to adjust scenarios to suit actual usage. Summary of the Invention

[0004] The purpose of this invention is to provide at least one method, apparatus, device, and storage medium for intelligent scene adjustment. This invention can at least solve the technical problem that users need to manually select scenes and set trigger conditions when using a scene, which is inconvenient and makes it impossible to adjust the scene according to the actual usage situation. At least it can achieve intelligent adjustment of the scene in the venue based on the user's posture in real time, without the need for manual selection of scenes or setting scene adjustment trigger commands, and can adjust the scene according to the actual usage situation, effectively improving the convenience and flexibility of scene adjustment for users.

[0005] To address the aforementioned technical problems, at least one embodiment of this application provides a scene intelligent adjustment method, comprising:

[0006] Periodically identify user postures within a given location;

[0007] Determine whether any of the identified user gestures are abnormal gestures that meet preset abnormal gesture conditions;

[0008] When an abnormal posture is detected in the identified user posture, the scene in the location where the abnormal posture occurs is intelligently adjusted.

[0009] At least one embodiment of this application also provides a scene intelligent adjustment device, including:

[0010] The acquisition module is used to periodically identify user poses appearing in the venue;

[0011] The judgment module is used to determine whether there is an abnormal posture among the identified user postures that meets the preset abnormal posture conditions.

[0012] The adjustment module is used to intelligently adjust the scene in the location where the abnormal posture occurs when it is determined that there is an abnormal posture in the identified user posture.

[0013] At least one embodiment of this application also provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the above-described scene intelligent adjustment method.

[0014] At least one embodiment of this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described scene intelligent adjustment method.

[0015] The embodiments of this application provide a method for periodically identifying user postures within a given location and determining whether any of the identified user postures meet preset abnormal posture conditions. When an abnormal posture is detected, the scene in the location where the abnormal posture occurs is intelligently adjusted. By monitoring user postures in real time and intelligently adjusting the scene based on these postures, the system eliminates the need for manual scene selection or setting of scene adjustment trigger commands. This approach effectively improves the convenience and flexibility of scene adjustments for users by adapting to actual usage scenarios.

[0016] In some optional embodiments, the periodic identification of user gestures occurring within the location includes:

[0017] A preset human posture recognition model is used to identify the periodically collected images in the location to obtain the user postures appearing in the location.

[0018] Human pose recognition models can extract high-quality individual poses, thereby improving the accuracy of pose estimation and achieving better pose estimation results.

[0019] In some optional embodiments, the step of using a preset human posture recognition model to identify periodically acquired images within the location to obtain user postures appearing within the location includes:

[0020] Features of the images within the location in each period are extracted to generate feature maps;

[0021] The feature map is input into the human posture recognition model for recognition, and the coordinates of key points used to describe the human posture are output.

[0022] The user's pose is determined based on the coordinates of the key points.

[0023] By detecting the bounding box of each person, extracting the features of the image, and inputting the feature map into the human pose recognition model for recognition, the coordinates of each key point are obtained, and the user pose is determined based on the coordinates of each key point, making the obtained user pose more accurate and the recognition accuracy higher.

[0024] In some optional embodiments, determining whether any of the identified user gestures meet preset abnormal gesture conditions includes:

[0025] A preset abnormal posture classification model is used to classify the user posture output by the human posture recognition model in order to determine whether there is an abnormal posture that meets the preset abnormal posture conditions; wherein, the output classification of the abnormal posture detection model includes a normal posture category and at least one abnormal posture category under the abnormal posture conditions.

[0026] By classifying user postures using an abnormal posture classification model, it is possible to determine whether a user's posture is abnormal. Based on the abnormal posture, the scene of the location can be intelligently adjusted, and the scene can be adjusted according to the actual usage situation, which effectively improves the convenience and flexibility of scene adjustment for users.

[0027] In some alternative embodiments, the location includes multiple sub-areas;

[0028] The determination of whether any of the identified user gestures meet the preset abnormal gesture conditions includes:

[0029] For each of the user postures identified in each of the sub-regions, it is determined whether there is an abnormal posture that meets the preset abnormal posture conditions; wherein, each of the sub-regions corresponds to a different location function, and the abnormal posture conditions corresponding to the sub-regions with different location functions are also different.

[0030] By determining whether there are abnormal user postures in sub-regions within the venue, the scene in each sub-region can be intelligently adjusted accordingly.

[0031] In some optional embodiments, when it is determined that there is an abnormal posture in the identified user posture, the step of intelligently adjusting the scene in the location where the abnormal posture occurs includes:

[0032] When it is determined that an abnormal posture exists in the user posture identified in the sub-region, the sub-region containing the abnormal posture is determined to be an adjustment sub-region.

[0033] Based on the scene of the sub-region adjacent to the adjustment sub-region, determine the scene adjustment task of the adjustment sub-region;

[0034] The scene of the adjustment sub-region is intelligently adjusted based on the scene adjustment task; wherein, the scene difference value at the junction between the scene of the intelligently adjusted adjustment sub-region and the scene of the adjacent sub-region is less than a preset difference threshold.

[0035] When making intelligent adjustments to sub-regions within a venue, the transition between the adjusted sub-region and adjacent sub-regions is made smooth, reducing the abruptness of the transition between the adjusted sub-region and adjacent sub-regions.

[0036] In some optional embodiments, the intelligent adjustment of the scene in the location where the abnormal posture occurs includes:

[0037] Intelligent adjustments are made to at least one of the following in the scene where the abnormal posture occurs: temperature, humidity, light color, sound, and light brightness.

[0038] By intelligently adjusting at least one of the following parameters of smart devices—temperature, humidity, light color, sound, and light brightness—the scene within the venue can be intelligently adjusted. Attached Figure Description

[0039] One or more embodiments are illustrated by way of example with reference to the accompanying drawings, and these illustrative descriptions do not constitute a limitation on the embodiments.

[0040] Figure 1 This is a flowchart illustrating a scene intelligent adjustment method provided in one embodiment of this application;

[0041] Figure 2 This is a schematic flowchart of a scene intelligent adjustment device provided in one embodiment of this application;

[0042] Figure 3 This is a schematic diagram of the structure of an electronic device provided in another embodiment of this application;

[0043] Figure 4 This is a flowchart illustrating a scene intelligent adjustment method provided in one embodiment of this application;

[0044] Figure 5 This is a schematic diagram of the sub-region distribution provided in one embodiment of this application. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the various embodiments of this application will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been provided in the various embodiments of this application to help readers better understand this application. However, the technical solutions claimed in this application can be implemented even without these technical details and various changes and modifications based on the following embodiments. The division of the various embodiments below is for the convenience of description and should not constitute any limitation on the specific implementation of this application. The various embodiments can be combined with and referenced by each other without contradiction.

[0046] To facilitate understanding of the embodiments of this application, the relevant content of the scene intelligent adjustment method will be introduced first.

[0047] Currently, the smart home industry is developing rapidly, using residences as a carrier and leveraging next-generation information technologies such as the Internet of Things, cloud computing, big data, and artificial intelligence to achieve interconnection and interoperability between system platforms and home appliances.

[0048] In some technologies, smart home scenarios are either preset default scenarios or user-created scenarios. However, using scenarios requires users to manually select scenarios and set trigger conditions, which is inconvenient and makes it impossible to adjust scenarios to suit actual usage.

[0049] To address the technical problem mentioned above, where users need to manually select scenarios and set trigger conditions, which is inconvenient and makes it impossible to adjust scenarios according to actual usage, this invention proposes a scenario intelligent adjustment method. The implementation details of the scenario intelligent adjustment method in this embodiment are described below. The following content is only for ease of understanding and is not necessary for implementing this solution.

[0050] Example 1:

[0051] The scene intelligent adjustment method of this embodiment can be applied to electronic devices with communication, computing, and data storage capabilities. Its specific process can be as follows: Figure 1 As shown, it includes:

[0052] Step 101: Periodically identify user poses appearing in the location.

[0053] Specifically, by setting up monitoring equipment such as cameras in the venue, images of users in the venue are acquired in real time or at intervals, and the user images acquired in multiple cycles are identified to obtain the user postures appearing in the venue.

[0054] Step 102: Determine whether there are any abnormal postures among the identified user postures that meet the preset abnormal posture conditions.

[0055] Specifically, abnormal posture refers to a posture that is inconsistent with the user's posture in the current scene of the venue. The abnormal posture will vary depending on the abnormal posture conditions. In this embodiment, determining whether there is an abnormal posture among the identified user postures that meets the preset abnormal posture conditions is actually to determine whether the current scene of the venue needs to be adjusted, and thus adjust the scene according to the actual usage.

[0056] Step 103: When it is determined that there is an abnormal posture in the identified user posture, the scene in the place where the abnormal posture occurs is intelligently adjusted.

[0057] Specifically, when an abnormal posture is detected in the identified user gestures, it indicates that the current scene of the current location needs to be adjusted. In this embodiment, the scene in the location where the abnormal posture occurs is intelligently adjusted. In some examples, each location is equipped with multiple smart devices, such as smart curtains, smart lights, smart speakers, and smart air conditioners. Each location has a smart scene controlled by these smart devices. By adjusting the smart devices in the location where the abnormal posture occurs, the scene within that location is intelligently adjusted.

[0058] In this embodiment, user postures appearing in a location are periodically identified, and it is determined whether any of the identified user postures meet preset abnormal posture conditions. When an abnormal posture is detected, the scene in the location where the abnormal posture occurs is intelligently adjusted. By monitoring user postures in real time and intelligently adjusting the scene based on these postures, there is no need for manual scene selection or setting scene adjustment trigger commands. This effectively improves the convenience and flexibility of scene adjustment for users by adjusting the scene according to actual usage.

[0059] In some embodiments, periodically identifying user poses occurring within a location includes:

[0060] A pre-defined human posture recognition model is used to identify images collected periodically within a location, thereby obtaining the user postures appearing within the location.

[0061] Specifically, the human pose recognition model is a neural network model, implemented using the AlphaPose estimation algorithm based on the Region-Based Multi-Person Pose Recognition (RMPE) algorithm. This enables the extraction of high-quality single-person poses, thereby improving the accuracy of pose estimation, even with inaccurate bounding boxes. In this embodiment, the human pose recognition model is used to identify images within a location to obtain the user poses appearing within that location.

[0062] In some embodiments, the human pose recognition model is obtained through the following steps:

[0063] Construct the original recognition model;

[0064] Based on the acquired image information samples of users from multiple periods and the corresponding user posture samples, the original recognition model is trained to obtain a human posture recognition model.

[0065] Specifically, by collecting image information samples of users from multiple historical periods and corresponding user posture samples as training sample data, reinforcement learning algorithms or deep learning algorithms are used to train the original recognition model, thereby obtaining a human posture recognition model.

[0066] In some embodiments, a preset human posture recognition model is used to identify periodically collected images within a location, and the user postures appearing within the location include:

[0067] Features of the images within each period are extracted to generate feature maps;

[0068] The feature map is input into the human pose recognition model for recognition, and the coordinates of key points used to describe the human pose are output.

[0069] The user's posture is determined based on the coordinates of key points.

[0070] Specifically, after acquiring images of the location over multiple periods, image features are extracted, using an AlphaPose estimation algorithm based on the Region-Based Multi-Person Pose Recognition (RMPE) algorithm. Multiple frames are captured from the images to detect the bounding boxes of each person. Specifically, by sliding across the image, local features are extracted, obtaining X, Y, and Z coordinates corresponding to colors (R, G, B). Two adjacent color patches are compared; if the colors match, they are defined as the same patch. This generates a feature map to capture different features in the image (such as edges, corners, etc.). The feature map is then input into the human pose recognition model for identification. Since the user's pose has a slope—for example, a person's arm is composed of two joints and not a straight line—the following formula represents the entire pose: Pose value = {(Input 1 * Weight 1) + (Input 2 * Weight 2) + ...} + Bias value, where the pose value represents the entire pose, the input refers to the slope formula of the pose, and the bias value > 0. Then, output the coordinates of each key point. For example, human pose 1 = X1, Y1, Z1, X11, Y11, Z11, X12, Y12, Z12, ...; human pose 2 = X2, Y2, Z2, X3, Y3, Z3, X5, Y5, Z5, X6, Y6, Z6.

[0071] In some cases, multiple user gestures can form a gesture combination, which can reflect user behavior in a certain scenario in a certain place.

[0072] In some embodiments, determining whether there is an abnormal posture among the identified user postures that meets preset abnormal posture conditions includes:

[0073] A pre-defined abnormal posture classification model is used to classify the user postures output by the human posture recognition model in order to determine whether there are abnormal postures that meet the pre-defined abnormal posture conditions. The output classification of the abnormal posture detection model includes a normal posture category and at least one abnormal posture category under abnormal posture conditions.

[0074] Specifically, the abnormal posture classification model is a neural network model. After the human posture recognition model outputs the user's posture, the abnormal posture classification model classifies the user's posture to determine whether it belongs to the normal posture category or the abnormal posture category. When the user's posture belongs to the normal posture category, it indicates that no event requiring scene adjustment has occurred in the venue, and no scene adjustment is needed. When the user's posture belongs to the abnormal posture category, it indicates that an event requiring scene adjustment has occurred in the venue, and scene adjustment is necessary. In some examples, there can be one or more abnormal posture conditions, each corresponding to an abnormal posture category. For example, when the abnormal posture condition is that the venue has different sub-regions and a certain sub-region needs adjustment, the abnormal posture category is the partition scene category. When the abnormal posture condition is that a user's posture is different from that of other users, or that a user's posture is different from a previous posture, the abnormal posture category is the special posture category. For example, if the current scene is a coffee shop and music is playing, and a user's posture is identified as a special posture category (falling down), then the music playback will stop.

[0075] In some embodiments, the abnormal pose classification model is obtained through the following steps:

[0076] Construct the original classification model;

[0077] The original classification model is trained based on user pose samples and pose category samples to obtain an abnormal pose classification model.

[0078] Specifically, user pose samples include normal pose samples and abnormal pose samples, and pose category samples include normal pose categories and abnormal pose categories. By collecting user pose samples and pose category samples from multiple historical periods as training sample data, reinforcement learning algorithms or deep learning algorithms are used to train the original classification model, thereby obtaining an abnormal pose classification model.

[0079] In some embodiments, the venue includes multiple sub-areas;

[0080] Determining whether any of the identified user poses meet preset abnormal pose conditions includes:

[0081] For each sub-region, the user postures identified are judged to determine whether there are any abnormal postures that meet the preset abnormal posture conditions. Each sub-region corresponds to a different location function, and the abnormal posture conditions for sub-regions with different location functions are also different.

[0082] Specifically, such as Figure 5 As shown, the venue contains multiple sub-areas, which can be either closed or open. These sub-areas combine to form the venue, such as area A and area B. In some examples, a sub-area is simply a specific area within the venue and cannot be combined to form the venue itself; for example, area A cannot form venue B. Abnormal posture recognition is performed on each sub-area to determine if any user posture is abnormal. It should be noted that sub-areas can have the same venue function or different functions. Correspondingly, when the venue function of each sub-area differs, the abnormal posture conditions for sub-areas with different functions also differ. In some examples, the venue has two sub-areas: a rest area and a learning area. In the rest area, the abnormal posture condition could be a user sitting or standing, while in the learning area, it could be a user lying down or standing. In some examples, if all sub-areas within the venue have the same venue function (e.g., all sub-areas are for learning), then the abnormal posture condition is a user lying down or standing. In some examples, the venue includes area A and area B. If the human posture in area A is identified as posture combination 1 (standing, waving his hands constantly, and walking around frequently in area A), and the human posture in area B is identified as posture combination 2 (sitting), then the abnormal posture category can be determined as the partition scene category.

[0083] In some embodiments, when an abnormal posture is detected in the identified user posture, intelligent adjustment of the scene in the location where the abnormal posture occurs includes:

[0084] When it is determined that there is an abnormal posture in the user posture identified in the sub-region, the sub-region with the abnormal posture is determined as the adjustment sub-region.

[0085] Based on the scene of the sub-regions adjacent to the adjustment sub-region, determine the scene adjustment task for the adjustment sub-region;

[0086] The scene of the sub-region is intelligently adjusted based on the scene adjustment task; wherein, the scene difference value at the junction between the scene of the intelligently adjusted sub-region and the scene of the adjacent sub-region is less than a preset difference threshold.

[0087] Specifically, when it is determined that at least one of the multiple sub-regions of the venue has an abnormal posture in the user's posture, the sub-region with the abnormal posture (i.e., the sub-region) is the sub-region that needs to be intelligently adjusted in the scene, and the scene of the adjusted sub-region needs to be intelligently adjusted.

[0088] It's important to understand that when intelligently adjusting the scene within a venue, not only the overall function of the venue but also the specific functions of different sub-areas are considered. When different sub-areas have different functions, simply intelligently adjusting the scene of a sub-area can easily lead to a lack of smooth transition between the adjusted sub-area and adjacent sub-areas, appearing abrupt. For example, in a coffee shop with a dining area and a sales area, the dining area is the adjusted sub-area. Simply adjusting the lighting in the dining area can easily make the transition between the two open areas appear abrupt. In this embodiment, the scene adjustment task for the adjusted sub-area is determined based on the scenes of the sub-areas adjacent to it, and then the scene of the adjusted sub-area is intelligently adjusted based on the scene adjustment task. After intelligently adjusting the scene of the adjusted sub-area, the scene difference value at the transition point between the intelligently adjusted scene of the adjusted sub-area and the scenes of the adjacent sub-areas is less than a preset difference threshold. In some examples, a coffee shop has a dining area and a sales area. The dining area is an adjustment sub-area. The dining area and the sales area are adjacent. The scene adjustment task of the dining area is determined based on the scene of the sales area (intelligent light color temperature is 3000K). The scene adjustment task of the dining area is to adjust the intelligent light color temperature to 4500K. The scene of the dining area is intelligently adjusted based on the scene adjustment task, and the intelligent light color temperature difference at the junction between the dining area and the sales area is less than the preset difference threshold.

[0089] In some examples, the venue includes a nap area and a study area, both of which are adjustable sub-areas. The study area and nap area are adjacent. The scene adjustment task for the nap area is determined based on the scene in the study area. This task is to close the smart curtains in the nap area. The scene in the nap area is then intelligently adjusted based on this task. Similarly, the scene adjustment task for the study area is determined based on the nap area's scene. This task is to turn on the desk lamps in the study area. The scene in the study area is then intelligently adjusted based on this task, and the brightness difference at the transition point between the study area and the nap area is less than a preset difference threshold.

[0090] In some embodiments, intelligent adjustment of the scene in a location where abnormal postures occur includes:

[0091] Intelligent adjustments are made to at least one of the following parameters in a location where an abnormal posture occurs: temperature, humidity, lighting color, sound, and lighting brightness.

[0092] Specifically, each venue is equipped with a variety of smart devices, such as smart curtains, smart lights, smart speakers, and smart air conditioners. Each venue has a smart scene controlled by smart devices. By adjusting the smart devices in the venue that exhibit abnormal behavior, at least one of the smart devices' temperature, humidity, light color, sound, and light brightness is intelligently adjusted, thereby intelligently adjusting the scene within the venue.

[0093] Example 2:

[0094] like Figure 4 As shown, this embodiment provides an exemplary content of Embodiment 1, namely, an exemplary process of a scene intelligent adjustment method, specifically including: a smart home management platform, a mobile terminal (central control), and smart devices (including millimeter-wave radar).

[0095] S1. Smart Building Equipment Binding:

[0096] (1) Networking of home central control devices:

[0097] ① Use a mobile device (such as an app) to scan the QR code on the screen of the home control device (if no account is bound, the QR code data content is fixed) to obtain the home control device ID information.

[0098] ② Simultaneously turn on the Bluetooth function of your mobile device, use Bluetooth to connect to the home control device, and enter the desired WiFi name or password.

[0099] ③ The home host obtains the WiFi name or password information, connects to the WiFi, and registers the home control device ID with the server.

[0100] (2) Adding sub-devices to the central control unit:

[0101] ① (After completing step 1 above, continue) Use a mobile terminal (such as an app) to connect to the discovered WiFi+BLE (GMesh) device via the Gmesh Bluetooth protocol, send WiFi information to the sub-device, and wait for the first device to reply that it has successfully connected to the router. Wait for subsequent devices to reply that they have successfully received the WiFi information. The maximum connection time is 90 seconds; exceeding this timeout indicates a connection failure. If a device cannot find the SSID or fails to connect to the router for other reasons, it is considered a connection failure.

[0102] ② For successfully connected devices, the home control device uploads its home control device ID and the device IDs of the added sub-devices to the server. Set the room location of the devices.

[0103] As shown in Table 1 below, the default scene setup includes setting the scene, including the usage location, commonly used smart device categories in the location, number of people (single / multiple), and scene mode.

[0104] Table 1 Relationship between Location and Scene

[0105]

[0106]

[0107] S2. Human Posture Model Training

[0108] (1) The AlphaPose estimation algorithm based on the multi-person pose recognition algorithm (RMPE) is adopted to obtain multiple frames of images and detect the bounding box of each person.

[0109] ① Slide the image to extract local features and obtain the X, Y, and Z coordinates corresponding to the color. Compare two adjacent color blocks; if the colors are the same, they are defined as the same color block. This generates a feature map that captures different features in the image (such as edges, corners, etc.).

[0110] ② Training image, value = {(input 1 * weight 1) + (input 2 * weight 2) + ...} + bias value; (bias value > 0), because an action has a slope, just like a person's arm, which is composed of two joints and not a straight line, so the above relationship is needed to represent the whole posture.

[0111] ③ Output the coordinates of each key point (such as x and y coordinates).

[0112] (2) It can be concluded that the human body's posture and movement 1 = X1, Y1, Z1, X11, Y11, Z11, X12, Y12, Z12...; the human body's posture and movement 2 = X2, Y2, Z2, X3, Y3, Z3, X5, Y5, Z5, X6, Y6, Z6.

[0113] (3) Partition Scene Model: If the human posture in area A of the venue is identified as combination 1 (standing, waving his hands constantly, and walking frequently within area A); and the human posture in area B of the venue is identified as combination 2 (sitting), then it is defined as the partition type.

[0114] (4) Special case model: human posture is identified as combination 3 (falling on the ground) and combination 4 (hanging upside down).

[0115] In this embodiment, the model input parameters are a set of feature maps of the same type, such as thousands of pictures of the same action, such as sitting, but with different sitting positions and image angles. The model output parameters are the user's posture.

[0116] S3. Initiate execution of a specific scenario by sending execution instructions to the device from the cloud. Based on the combination of the number of people and their posture in the image recognition, if the location of the person is identified as a zone, the scenario adjustment is triggered.

[0117] (1) If the current scene mode is a coffee shop, and the device is running smoothly, recognizing human posture combinations and positions similar to the training model, then the scene adjustment = scene * weight + sub-scene * offset. In this case, the sub-scene is executed: Dining area: smart light color temperature 4500K, playing soft music; Sales area: smart light color temperature 3000K. The sub-scene refers to recognizing postures of special case models. This means that when a scene is started (e.g., coffee shop operation, such as B), special case model human postures occur during execution, and the scene is adjusted in this specific area. (e.g., fine-tuning the scene in area A).

[0118] (2) If the current scene mode is a coffee shop, music is playing; if a special posture (falling down) is detected, the music will stop playing.

[0119] Example 3:

[0120] Another embodiment of this application relates to a scene intelligent adjustment device. The implementation details of this scene intelligent adjustment device are described below. The following implementation details are provided for ease of understanding and are not essential for implementing this solution. A schematic diagram of the scene intelligent adjustment device in this embodiment can be seen as follows: Figure 2 As shown, it includes:

[0121] The acquisition module 201 is used to periodically identify user postures appearing in the venue;

[0122] The judgment module 202 is used to determine whether there is an abnormal posture in the identified user posture that meets the preset abnormal posture conditions.

[0123] The adjustment module 203 is used to intelligently adjust the scene in the location where the abnormal posture occurs when it is determined that there is an abnormal posture in the identified user posture.

[0124] In some embodiments, the acquisition module 201 is further configured to use a preset human posture recognition model to identify the periodically collected images in the venue, thereby obtaining the user posture appearing in the venue.

[0125] In some embodiments, the acquisition module 201 includes:

[0126] The extraction unit is used to extract features from the images within the location in each period and generate feature maps.

[0127] The coordinate output unit is used to input the feature map into the human pose recognition model for recognition and output the coordinates of key points that describe the human pose.

[0128] The pose determination unit is used to determine the user's pose based on the coordinates of key points.

[0129] In some embodiments, the judgment module 202 is further configured to classify the user posture output by the human posture recognition model using a preset abnormal posture classification model, so as to determine whether there is an abnormal posture that meets the preset abnormal posture conditions; wherein, the output classification of the abnormal posture detection model includes a normal posture category and at least one abnormal posture category under abnormal posture conditions.

[0130] In some embodiments, the judgment module 202 is further configured to determine whether there is an abnormal posture that meets the preset abnormal posture conditions among the user postures identified in each sub-region; wherein, the venue includes multiple sub-regions; each sub-region corresponds to a different venue function, and the abnormal posture conditions corresponding to sub-regions with different venue functions are also different.

[0131] In some embodiments, the adjustment module 203 includes:

[0132] The sub-region determination unit is used to determine the sub-region with the abnormal posture as the adjustment sub-region when it is determined that there is an abnormal posture in the user posture identified in the sub-region.

[0133] The task determination unit is used to determine the scene adjustment task of the adjustment sub-region based on the scene of the sub-region adjacent to the adjustment sub-region;

[0134] The adjustment unit is used to intelligently adjust the scene of the adjustment sub-region based on the scene adjustment task; wherein, the scene difference value at the junction between the scene of the intelligently adjusted adjustment sub-region and the scene of the adjacent sub-region is less than a preset difference threshold.

[0135] In some embodiments, the adjustment module 203 is also used to intelligently adjust at least one of the temperature, humidity, light color, sound, and light brightness of the scene in the location where the abnormal posture occurs.

[0136] It is worth mentioning that all modules involved in this embodiment are logical modules. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. Furthermore, to highlight the innovative aspects of this application, this embodiment does not introduce units that are not closely related to solving the technical problems proposed in this application; however, this does not mean that other units are absent in this embodiment.

[0137] Example 4:

[0138] Another embodiment of this application relates to an electronic device, such as... Figure 3 As shown, it includes: at least one processor 901; and a memory 902 communicatively connected to the at least one processor 901; wherein the memory 902 stores instructions executable by the at least one processor 901, the instructions being executed by the at least one processor 901 to enable the at least one processor 901 to execute the scene intelligent adjustment method in the above embodiments.

[0139] The memory and processor are connected via a bus, which can include any number of interconnecting buses and bridges, connecting various circuits of one or more processors and memories. The bus can also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and will not be described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over the wireless medium via an antenna, which further receives data and transmits it to the processor.

[0140] The processor manages the bus and general processing, and also provides various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory is used to store data used by the processor during operation.

[0141] Example 5:

[0142] Another embodiment of this application relates to a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the method embodiments described above.

[0143] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0144] Those skilled in the art will understand that the above embodiments are specific embodiments for implementing this application, and in practical applications, various changes can be made to them in form and detail without departing from the spirit and scope of this application.

Claims

1. A method for intelligent scene adjustment, characterized in that, include: Periodically identify user postures within a given location; The venue includes multiple sub-areas; Determine whether any of the identified user gestures are abnormal gestures that meet preset abnormal gesture conditions; When it is determined that there is an abnormal posture among the identified user postures, the scene in the place where the abnormal posture occurs is intelligently adjusted; When it is determined that an abnormal posture exists in the identified user postures, the intelligent adjustment of the scene in the location where the abnormal posture occurs includes: When it is determined that an abnormal posture exists in the user posture identified in the sub-region, the sub-region containing the abnormal posture is determined to be an adjustment sub-region. The scene adjustment task of the adjustment sub-region is determined based on the scene of the sub-region adjacent to the adjustment sub-region; the scene functions of the adjustment sub-region and the sub-region adjacent to the adjustment sub-region are different. The scene of the adjustment sub-region is intelligently adjusted based on the scene adjustment task; wherein, the scene difference value at the junction between the scene of the intelligently adjusted adjustment sub-region and the scene of the adjacent sub-region is less than a preset difference threshold.

2. The scene intelligent adjustment method according to claim 1, characterized in that, The periodic identification of user gestures within the location includes: A preset human posture recognition model is used to identify the periodically collected images in the location to obtain the user postures appearing in the location.

3. The scene intelligent adjustment method according to claim 2, characterized in that, The step of using a preset human posture recognition model to identify periodically collected images within the location, and obtaining the user postures appearing within the location, includes: Features of the images within the location in each period are extracted to generate feature maps; The feature map is input into the human posture recognition model for recognition, and the coordinates of key points used to describe the human posture are output. The user's pose is determined based on the coordinates of the key points.

4. The scene intelligent adjustment method according to claim 2, characterized in that, The determination of whether any of the identified user gestures meet the preset abnormal gesture conditions includes: A preset abnormal posture classification model is used to classify the user posture output by the human posture recognition model in order to determine whether there is an abnormal posture that meets the preset abnormal posture conditions; wherein, the output classification of the abnormal posture classification model includes a normal posture category and at least one abnormal posture category under the abnormal posture conditions.

5. The scene intelligent adjustment method according to claim 1, characterized in that, The determination of whether any of the identified user gestures meet the preset abnormal gesture conditions includes: For each of the user postures identified in each of the sub-regions, it is determined whether there is an abnormal posture that meets the preset abnormal posture conditions; wherein, each of the sub-regions corresponds to a different location function, and the abnormal posture conditions corresponding to the sub-regions with different location functions are also different.

6. The scene intelligent adjustment method according to claim 1, characterized in that, The intelligent adjustment of the scene in the location where the abnormal posture occurs includes: Intelligent adjustments are made to at least one of the following in the scene where the abnormal posture occurs: temperature, humidity, light color, sound, and light brightness.

7. A scene intelligent adjustment device, characterized in that, include: The acquisition module is used to periodically identify user poses appearing in the venue; The venue includes multiple sub-areas; The judgment module is used to determine whether there is an abnormal posture among the identified user postures that meets the preset abnormal posture conditions. The adjustment module is used to intelligently adjust the scene in the location where the abnormal posture occurs when it is determined that there is an abnormal posture in the identified user posture. The adjustment module includes: The sub-region determination unit is used to determine the sub-region containing the abnormal posture as an adjustment sub-region when it is determined that the abnormal posture exists in the user posture identified in the sub-region. The task determination unit is used to determine the scene adjustment task of the adjustment sub-region based on the scene of the sub-region adjacent to the adjustment sub-region; the scene function of the adjustment sub-region and the sub-region adjacent to the adjustment sub-region is different; An adjustment unit is used to intelligently adjust the scene of the adjustment sub-region based on the scene adjustment task; wherein the scene difference value at the junction between the scene of the intelligently adjusted adjustment sub-region and the scene of the adjacent sub-region is less than a preset difference threshold.

8. An electronic device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the scene intelligent adjustment method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the scene intelligent adjustment method as described in any one of claims 1 to 6.

Images