Machine learning control method and apparatus, electronic device, and storage medium

Abstract

This invention provides a learning machine control method, device, electronic device, and storage medium, relating to the field of control technology. The learning machine control method includes: controlling a distance sensor to start based on a start command from the learning machine; acquiring the distance value between a target object and the learning machine collected by the distance sensor; performing position detection on the target object based on the distance value to obtain a position detection result; if the position detection result indicates that the target object is within a preset range of the learning machine, then controlling the distance sensor to turn off and controlling a temperature sensor to start; acquiring the temperature value of the target object detected by the temperature sensor; performing type identification on the target object based on the temperature value to obtain a type identification result; if the type identification result indicates that the target object is a human body, then controlling the temperature sensor and distance sensor to turn off and controlling a camera to start for posture detection, thereby reducing energy consumption, extending the battery life of the learning machine, reducing the number of charging cycles, and improving the user experience.

Description

Technical Field

[0001] This invention relates to the field of control method technology, and in particular to a learning machine control method, device, electronic device and storage medium. Background Technology

[0002] Learning devices can meet students' needs for reading aloud, searching for answers, attending online classes, and taking photos. Therefore, the requirements for the front-facing camera on learning devices are higher and more numerous compared to electronic devices such as mobile phones and tablets. For example, when using a learning device, the front-facing camera detects the user's posture to remind them to maintain good posture and avoid slouching. However, the continuous operation of the front-facing camera leads to increased power consumption, resulting in frequent charging of the learning device and affecting the user experience. Summary of the Invention

[0003] This invention provides a learning machine control method, device, electronic device, and storage medium to solve the defects of existing learning machines that consume power quickly and require frequent charging.

[0004] This invention provides a learning machine control method, comprising: controlling a distance sensor to start based on a start command of the learning machine; acquiring a distance value between a target object and the learning machine collected by the distance sensor; performing position detection on the target object based on the distance value to obtain a position detection result; if the position detection result indicates that the target object is within a preset range of the learning machine, then controlling the distance sensor to turn off and controlling a temperature sensor to start; acquiring a temperature value of the target object detected by the temperature sensor; performing type identification on the target object based on the temperature value to obtain a type identification result; if the type identification result indicates that the target object is a human body, then controlling the temperature sensor and the distance sensor to turn off and controlling the camera to start for posture detection.

[0005] According to a learning machine control method provided by the present invention, controlling the camera to start includes: controlling the camera to capture an image of a target object with a first pixel; acquiring the image of the target object, performing human detection on the target object based on the image of the target object, and obtaining a human detection result; if the human detection result determines that the target object is a human body, then controlling the camera to acquire an image of the user's sitting posture with a second pixel; wherein, the first pixel is smaller than the second pixel.

[0006] According to a learning machine control method provided by the present invention, after controlling the camera to acquire a user's sitting posture image with a second pixel, the method further includes: acquiring the user's sitting posture image; performing sitting posture detection on the user based on the sitting posture image to obtain a sitting posture detection result; if the sitting posture detection result indicates that the user's sitting posture is normal, controlling the camera to start shooting once at a second pixel interval of a first preset time, wherein the first preset time is longer than the initial shooting interval of the camera.

[0007] According to a learning machine control method provided by the present invention, if the posture detection result indicates that the user's posture is abnormal, the alarm of the learning machine is controlled to remind the user to adjust the posture; if the posture detection result indicates that the user's posture is abnormal again within a preset time period, the learning machine is controlled to go into sleep mode.

[0008] According to a learning machine control method provided by the present invention, the step of performing position detection on the target object based on the distance value to obtain a position detection result further includes: if the position detection result indicates that the target object is outside the preset range of the learning machine, then controlling the distance sensor to start detection once every second preset time interval, wherein the second preset time interval is longer than the initial acquisition interval of the distance sensor.

[0009] According to a learning machine control method provided by the present invention, the step of identifying the type of the target object based on the temperature value of the target object to obtain the type identification result further includes: if the type identification result indicates that the target object is not a human body, then controlling the temperature sensor to turn off and controlling the distance sensor to turn on again.

[0010] According to a learning machine control method provided by the present invention, if the type identification result indicates that the target object is not a human body, the method further includes: controlling the temperature sensor to turn off and controlling the distance sensor to turn on again; acquiring the distance value between the current target object and the learning machine detected by the distance sensor; performing target identification on the current target object based on the distance value between the current target object and the learning machine to obtain a target identification result; and if the target identification result indicates that the target object has changed, performing position detection on the target object based on the latest acquired distance value.

[0011] The present invention also provides a learning machine control device, comprising:

[0012] The control unit is configured to: control the distance sensor to start based on the learning machine's activation command; acquire the distance value between the target object and the learning machine collected by the distance sensor; perform position detection on the target object based on the distance value to obtain a position detection result; if the position detection result indicates that the target object is within a preset range of the learning machine, control the distance sensor to turn off and control the temperature sensor to start; acquire the temperature value of the target object detected by the temperature sensor; perform type identification on the target object based on the temperature value to obtain a type identification result; if the type identification result indicates that the target object is a human body, control the temperature sensor and the distance sensor to turn off and control the camera to start for posture detection.

[0013] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of any of the learning machine control methods described above.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the learning machine control methods described above.

[0015] The learning machine control method, device, electronic device, and storage medium provided by this invention control a distance sensor to start based on the learning machine's start command. The distance sensor detects the position of the target object based on the distance value between the target object and the learning machine. If the target object is within a preset range of the learning machine, the distance sensor is turned off and a temperature sensor is started. The target object's type is identified based on the temperature value detected by the temperature sensor. If the target object is a human body, the temperature sensor and distance sensor are turned off, and a camera is started to detect sitting posture. The camera is activated only after position detection and type identification, thereby reducing energy consumption, extending the learning machine's battery life, reducing charging frequency, and improving the user experience. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is one of the flowcharts illustrating the learning machine control method provided by the present invention;

[0018] Figure 2This is a flowchart illustrating step 1320 in the learning machine control method provided by the present invention;

[0019] Figure 3 This is a flowchart illustrating the process after step 2400 in the learning machine control method provided by the present invention.

[0020] Figure 4 This is the second flowchart of the learning machine control method provided by the present invention;

[0021] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0023] The following is combined with Figures 1-4 The learning machine control method of the present invention is described.

[0024] This invention provides a learning machine control method, such as... Figure 1 As shown, it includes: Step 1100, controlling the distance sensor to start based on the learning machine's start command. Step 1200, acquiring the distance value between the target object and the learning machine collected by the distance sensor; Step 1210, performing position detection on the target object based on the distance value to obtain a position detection result; Step 1220, if the position detection result indicates that the target object is within a preset range of the learning machine, then controlling the distance sensor to turn off and controlling the temperature sensor to start. Step 1300, acquiring the temperature value of the target object detected by the temperature sensor; Step 1310, performing type identification on the target object based on the temperature value to obtain a type identification result; Step 1320, if the type identification result indicates that the target object is a human body, then controlling the temperature sensor and distance sensor to turn off, and controlling the camera to start for posture detection.

[0025] The activation command is a power-on command. Specifically, it receives the trigger command from the power button, controls the learning machine to turn on, and simultaneously controls the proximity sensor to turn on. The activation command can also be a power-on command or a wake-up command. Specifically, during power-on, the proximity sensor is activated based on the power-on command. When the learning machine's screen is in sleep mode, the screen is woken up via the power button or the human-computer interaction module, and the proximity sensor is activated simultaneously with the screen being woken up.

[0026] Before detecting the sitting posture of the target object, the distance value between the target object and the learning machine is first acquired by the distance sensor. Based on the distance value, the position of the target object is detected to obtain the position detection result. Specifically, the distance between the user and the learning machine is within a certain range when the user is using the learning machine, and a preset range is determined based on the size of this range. If the distance value exceeds the preset range, the user is considered to be in a non-learning state, and no sitting posture detection is required; if the distance value does not exceed the preset range, the user is considered to be in a learning state, and sitting posture detection is required.

[0027] In one optional embodiment, the lower limit of the preset range is a preset value, while the upper limit is not limited. The preset value represents the minimum distance that must be satisfied between the user and the learning machine when the user is in a learning state. Specifically, if the distance value is greater than or equal to the preset value, the user is considered to be in a learning state, posture detection is performed, and the location recognition result is determined to indicate that the target object is within the preset range. In another optional embodiment, the upper and lower limits of the preset range are respectively the maximum and minimum distances that must be satisfied between the user and the learning machine when the user is in a learning state. Specifically, if the distance value is between the upper and lower limits, the user is considered to be in a learning state, posture detection is performed, and the location detection result is determined to indicate that the target object is within the preset range. Alternatively, the upper limit of the preset range can be set to the maximum distance between the user and the learning machine when the user is in a learning state, without specifically limiting the lower limit. Correspondingly, if the distance value is less than or equal to the upper limit, the user is considered to be in a learning state, posture detection is performed, and the location recognition result is determined to indicate that the target object is within the preset range.

[0028] Because various situations may occur when using the learning machine, the distance sensor may detect the user, or it may detect objects such as water cups or books. Therefore, if the position detection result indicates that the target object is within the preset range of the learning machine, the distance sensor is turned off and the temperature sensor is activated. This is to identify the type of target object before performing posture detection, thus excluding cases where the target object is an item or a pet.

[0029] Specifically, the process first acquires the temperature value of the target object detected by a temperature sensor; then, based on the temperature value, the target object is classified, yielding a classification result. The classification process can be as follows: the target object's temperature value is compared to the range of human body temperature. If the target object's temperature value falls within the range of human body temperature, the target object is identified as a human body, requiring posture detection, and the classification result is determined to be a human body. Alternatively, the classification process can involve acquiring the temperature values ​​of the target object at different locations. If the temperature values ​​at different locations are inconsistent but all fall within the range of human body temperature, the target object is identified as a human body, requiring posture detection, and the classification result is determined to be a human body.

[0030] After determining the target object to be a human body in step 1310, the sitting posture of the human body needs to be detected. At this time, the temperature sensor and distance sensor are turned off and the camera is turned on.

[0031] The learning machine is equipped with a temperature sensor, a distance sensor, and a camera. Optionally, the temperature sensor is an infrared thermometer, and the distance sensor is a Time-of-Flight (TOF) sensor.

[0032] The learning machine control method provided in this embodiment of the invention controls the activation of a distance sensor based on the activation command of the learning machine. The distance sensor detects the position of the target object based on the distance value between the target object and the learning machine. If the target object is within a preset range of the learning machine, the distance sensor is deactivated and a temperature sensor is activated. The target object's type is identified based on the temperature value detected by the temperature sensor. If the target object is a human body, both the temperature sensor and the distance sensor are deactivated, and a camera is activated to detect sitting posture. The camera is activated only after position detection and type identification, thereby reducing energy consumption, extending the battery life of the learning machine, reducing the number of charging cycles, and improving the user experience.

[0033] In a specific embodiment of the present invention, such as Figure 2 As shown, controlling the camera to start includes: step 2100, controlling the camera to capture an image of the target object with a first pixel; step 2200, acquiring the image of the target object; step 2300, performing human detection on the target object based on the image of the target object to obtain a human detection result; step 2400, if the human detection result determines that the target object is a human figure, then controlling the camera to capture an image of the user's sitting posture with a second pixel. Wherein, the first pixel is smaller than the second pixel.

[0034] When using the learning machine, the target object may be a pet such as a cat or dog. Since pets' body temperature is close to that of humans, identifying the type of target object based on its temperature is insufficient to distinguish between humans and pets. Therefore, the camera is first controlled to move at a lower first pixel level. Then, human detection is performed on the target object image. Once the human detection confirms that the target object is human, the camera is controlled to move at a higher second pixel level to capture a clear image of the sitting posture.

[0035] The process of human detection on a target object can be as follows: The acquired target object image is input into a pre-built human detection model. The human detection model performs human detection on the target object it represents based on the input target object image and outputs the human detection result. The human detection model is obtained through pre-training. The training process of the human detection model includes: acquiring a large number of human images of sample people and labeling the sample human images with human data tags; then, based on the sample human images of sample people and the human data tags of the sample human images, training the initial human detection model to obtain the trained human detection model.

[0036] The process of human detection of a target object can also be as follows: image processing is performed on the target object image to obtain image contour data, the image contour data is matched with human standard contour data, and if the matching degree between the image contour data and the human standard contour data reaches a preset value, then the target object image is determined to be a human image.

[0037] Optionally, the first pixel is 300,000 pixels and the second pixel is 2,000,000 pixels.

[0038] The learning machine control method provided in this embodiment of the invention first controls the camera to capture an image of the target object with a lower first pixel; then, it performs human detection on the target object based on the target object image. If the target object is determined to be human, the camera is controlled to capture an image of the user's sitting posture with a higher second pixel. This avoids inaccurate identification of the target object based on its temperature value, which would otherwise lead to high power consumption when the camera operates at high pixel count during posture detection. This further reduces the power consumption of the camera and extends the standby time of the learning machine.

[0039] like Figure 3 As shown, after controlling the camera to acquire the user's sitting posture image at the second pixel, the method further includes: step 3100, acquiring the sitting posture image; step 3200, performing sitting posture detection on the user based on the sitting posture image to obtain a sitting posture detection result; step 3300, if the sitting posture detection result indicates that the user's sitting posture is normal, then controlling the camera to start a shooting session at the second pixel interval for a first preset time. The first preset time is longer than the initial shooting interval of the camera.

[0040] If the posture detection result indicates that the user's posture is abnormal, the alarm in the learning machine will be activated to remind the user to adjust their posture, thus achieving the purpose of posture correction. If the posture detection result indicates that the user's posture is normal, the camera will be activated to take a picture once at a first preset time interval, thereby reducing the camera's shooting frequency and reducing energy consumption while effectively monitoring posture.

[0041] The process of detecting user posture based on seated images involves pre-building a database of normal sitting postures, and then matching the user's sitting posture image against this database. If any standard sitting posture data matches the user's sitting posture image, the user's posture is considered normal. Conversely, if the user's sitting posture image does not match any standard sitting posture data, the user's posture is considered abnormal.

[0042] The process of detecting user posture based on seated images can also be as follows: First, an abnormal sitting posture database is pre-built. Then, the user's sitting posture image is matched against this database. If any abnormal sitting posture data matches the user's sitting posture image, the user's posture is determined to be abnormal. Conversely, if the user's sitting posture image does not match any abnormal sitting posture data, the user's posture is considered normal.

[0043] In addition to the above, the process of detecting user posture based on posture images can also be as follows: inputting the user's posture image into a posture detection model, the posture detection model performs posture detection on the user represented by the input posture image and outputs the posture detection result. The posture detection model is obtained through pre-training. The training process of the posture detection model includes: acquiring a large number of sample posture images of people and labeling the posture state of the sample posture images; then, based on the sample posture images of people and the posture state labels, training the initial posture detection model to obtain the trained posture detection model.

[0044] Optionally, the first preset duration is 5 minutes, with the camera detecting the second pixel every 5 minutes to effectively monitor the user's posture. Of course, the first preset duration can be 8 minutes, 10 minutes, or other durations, as long as it is longer than the camera's initial shooting interval to reduce the shooting frequency.

[0045] The learning machine control method provided in this embodiment of the invention controls the camera to start shooting once at a second pixel interval for a first preset time when the posture detection result indicates that the user's posture is normal. By reducing the shooting frequency of the camera, the power consumption of the camera is reduced, the usage time of the learning machine is extended, frequent charging is avoided, and the user experience is improved.

[0046] In one specific embodiment of the present invention, if the posture detection result indicates that the user's posture is abnormal, the alarm of the learning machine is controlled to remind the user to adjust the posture; if the posture detection result indicates that the user's posture is abnormal again within a preset time, the learning machine is controlled to go into sleep mode.

[0047] After step 3200 determines the user's posture detection result, if the result indicates an abnormal posture, an alarm is activated to remind the user to adjust their posture. Optionally, the alarm can be a voice alarm or an audible and visual alarm, playing a built-in voice prompt to remind the user to adjust their posture. Alternatively, the alarm can be a display screen, displaying an image of the user's posture or reminder text to remind the user to adjust their posture.

[0048] The preset duration is longer than the first preset duration. The preset duration is 20 minutes or 30 minutes. Based on the number of times abnormal posture is detected in the user within the preset duration, the learning machine will be put into sleep mode to remind the user to pay attention to their posture and enhance the posture correction effect.

[0049] The learning machine control method provided in this embodiment of the invention performs posture detection on the user and obtains the posture detection result. If the posture detection result shows that the user's posture is abnormal, an alarm is triggered to remind the user to adjust their posture, thereby correcting the abnormal posture and avoiding the adverse effects of prolonged abnormal posture on the body. If the detection result indicates that the user's posture is abnormal again within a preset time, it means that the user has difficulty maintaining a normal posture and needs to pay close attention. The learning machine is then put into sleep mode to force the user to rest and urge the user to adjust their posture as prompted.

[0050] In addition, if the posture detection results indicate that the user's posture is abnormal again within a preset time, the learning machine will upload the data to the parent's end to remind the parent to pay attention to the child's posture problem.

[0051] The target object is located based on the distance value to obtain the location detection result. Then, if the location detection result indicates that the target object is outside the preset range of the learning machine, the distance sensor is controlled to start detection once every second preset time interval.

[0052] Specifically, the lower limit of the preset range is a preset value, while the upper limit is not limited. If the distance between the target object detected by the distance sensor and the learning machine is not greater than the preset value, it indicates that the target object is in a non-learning state, and posture detection is unnecessary; the position detection result is determined to be that the target object is outside the preset range. Alternatively, the upper limit of the preset range can be set to the maximum distance between the user and the learning machine when the user is in a learning state, without specifically limiting the lower limit. If the distance between the target object detected by the distance sensor and the learning machine is not less than the upper limit of the preset range, it indicates that the target object is not in a learning state, and the position detection result is determined to be that the target object is outside the preset range. Or, the upper limit of the preset range can be a set value, while the lower limit is not specifically limited. If the distance value is greater than the upper limit, it is determined that the user is in a non-learning state, and posture detection is unnecessary; the position detection result is determined to be that the target object is outside the preset range of the learning machine.

[0053] The second preset duration is longer than the initial acquisition interval of the distance sensor. Optionally, the second preset duration can be 0.02s or 0.1s, etc., as long as it achieves the purpose of reducing the frequency of distance sensor information acquisition.

[0054] The learning machine control method provided in this embodiment of the invention controls the distance sensor to start detection once at a second preset time interval when the position detection result indicates that the target object is outside the preset range of the learning machine, so that it can be detected in time when the user approaches the learning machine. Furthermore, by controlling the detection frequency of the distance sensor, energy consumption is reduced, the battery power supply time is extended, and the user experience is further improved.

[0055] Based on the target object's temperature value, the target object is identified as a type, and the type identification result is obtained. Then, if the type identification result indicates that the target object is not a human body, the temperature sensor is turned off and the distance sensor is turned on again.

[0056] Specifically, if the temperature value detected by the temperature sensor for the target object does not fall within the human body temperature range, it indicates that the target object is an object such as a table or chair, and the category identification result is determined that the target object is not a human body. Alternatively, if the temperature values ​​of the target object are the same at different locations, it indicates that the target object is an object, and the category identification result is determined that the target object is not a human body. In this case, the temperature sensor is turned off to reduce the power consumption of components in the learning machine.

[0057] To reduce energy consumption, the distance sensor can be restarted by performing a detection every second preset time interval, thereby reducing its energy consumption. Alternatively, the distance sensor can be restarted by performing a detection every third preset time interval, which is longer than the second preset time interval.

[0058] The learning machine control method provided in this embodiment of the invention controls the temperature sensor to turn off and the distance sensor to turn on again when the type recognition result indicates that the target object is not a human body. This keeps both the camera and the temperature sensor in the learning machine in a turned-off state, reducing energy consumption and extending standby time. At the same time, the distance sensor is in a working state, which can detect the user in time so as to provide posture detection for the user.

[0059] In this process, the distance sensor, which is in the activated state, re-detects the distance between the target object and the learning machine. If the target object remains unchanged, the temperature sensor will be activated again, resulting in frequent switching of both the temperature and distance sensors. Therefore, as follows: Figure 4 As shown, after the control distance sensor is restarted, the method further includes: step 4100, acquiring the distance value between the current target object and the learning machine detected by the distance sensor; step 4200, performing target recognition on the current target object based on the distance value between the current target object and the learning machine, and obtaining the target recognition result; step 4300, if the target recognition result indicates that the target object has changed, then performing position detection on the target object based on the latest acquired distance value.

[0060] The process of target recognition based on the distance between the current target and the learning machine is as follows: the latest distance value is compared with the previous distance value. If the comparison results are inconsistent, it is determined that the target has changed, and the target recognition result is determined to be a change in the target.

[0061] If the target object is determined to have changed, step 1210 is executed based on the latest acquired distance value to determine whether the changed target object is within the preset range.

[0062] The process includes the following steps after step 4200: If the target recognition result indicates that the target object has not changed, then the distance value between the current target object and the learning machine is obtained again, and steps 4100 and 4200 are repeated. Specifically, the latest obtained distance value is compared with the previously obtained distance value. If the comparison result is consistent, it is determined that the target object has not changed, and the target recognition result is determined to be that the target object has not changed.

[0063] The learning machine control method provided in this embodiment of the invention identifies the current target object based on the distance value between the current target object and the learning machine after the distance sensor is restarted. It restarts the position detection only after the target object has changed, thereby avoiding the frequent switching of the distance sensor and temperature sensor when the target object has not changed, and extending the service life of the distance sensor and temperature sensor.

[0064] In another specific embodiment, the target object is identified based on its temperature value to obtain a type identification result. The method then further includes: if the type identification result indicates that the target object is not a human body, controlling the temperature sensor to perform detection at a first preset frequency. The first preset frequency is lower than the initial frequency of the temperature sensor.

[0065] In this embodiment, the category identification result indicates that the current target object is not a human body, so there is no need to perform posture detection. To reduce the number of times the distance sensor and temperature sensor are switched on and off, the temperature sensor is controlled to reduce its detection frequency, thereby reducing energy consumption. Simultaneously, the target object's category is identified based on the temperature value detected again, thus enabling timely detection of the user.

[0066] This invention also provides a learning machine control device, which includes a control unit. The control unit is used to control a distance sensor to start based on a start command of the learning machine; acquire a distance value between a target object and the learning machine collected by the distance sensor; perform position detection on the target object based on the distance value to obtain a position detection result; if the position detection result indicates that the target object is within a preset range of the learning machine, then control the distance sensor to turn off and control a temperature sensor to start; acquire a temperature value of the target object detected by the temperature sensor; perform type identification on the target object based on the temperature value to obtain a type identification result; if the type identification result indicates that the target object is a human body, then control the temperature sensor and the distance sensor to turn off and control the camera to start for posture detection.

[0067] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, such as... Figure 5 As shown, the electronic device may include a processor 510, a communications interface 520, a memory 530, and a communication bus 540, wherein the processor 510, communications interface 520, and memory 530 communicate with each other via the communication bus 540. The processor 510 can call logical commands in the memory 530 to execute the following methods: controlling the distance sensor to start based on the learning machine's start command; acquiring the distance value between the target object and the learning machine collected by the distance sensor; performing position detection on the target object based on the distance value to obtain a position detection result; if the position detection result indicates that the target object is within a preset range of the learning machine, then controlling the distance sensor to turn off and controlling the temperature sensor to start; acquiring the temperature value of the target object detected by the temperature sensor; performing type identification on the target object based on the target object temperature value to obtain a type identification result; if the type identification result indicates that the target object is a human body, then controlling the temperature sensor and the distance sensor to turn off, and controlling the camera to start for posture detection.

[0068] Furthermore, the logical commands in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several commands to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. 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.

[0069] This invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program performs the methods provided in the above embodiments, including, for example,: controlling a distance sensor to start based on a learning machine's start command; acquiring a distance value between a target object collected by the distance sensor and the learning machine; performing position detection on the target object based on the distance value to obtain a position detection result; if the position detection result indicates that the target object is within a preset range of the learning machine, controlling the distance sensor to turn off and controlling a temperature sensor to start; acquiring a temperature value of the target object detected by the temperature sensor; performing type identification on the target object based on the target object temperature value to obtain a type identification result; if the type identification result indicates that the target object is a human body, controlling the temperature sensor and the distance sensor to turn off and controlling the camera to start for posture detection.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A learning machine control method, characterized in that, include: The distance sensor is activated based on the start command from the learning machine; The distance value between the target object and the learning machine is obtained from the distance sensor. The target object is located based on the distance value to obtain a location detection result; if the location detection result indicates that the target object is within a preset range of the learning machine, the distance sensor is turned off and the temperature sensor is turned on. Obtain the temperature value of the target object detected by the temperature sensor; Based on the temperature value of the target object, the target object is identified as a type, and a type identification result is obtained; if the type identification result indicates that the target object is a human body, the temperature sensor and the distance sensor are turned off, and the camera is turned on to perform sitting posture detection.

2. The learning machine control method according to claim 1, characterized in that, The step of controlling the camera to start includes: controlling the camera to capture an image of the target object with a first pixel; acquiring the image of the target object, performing human detection on the target object based on the image of the target object, and obtaining a human detection result; if the human detection result determines that the target object is a human body, then controlling the camera to capture an image of the user's sitting posture with a second pixel; wherein, the first pixel is smaller than the second pixel.

3. The learning machine control method according to claim 2, characterized in that, After controlling the camera to acquire the user's sitting posture image with the second pixel, the method further includes: acquiring the user's sitting posture image; performing sitting posture detection on the user based on the sitting posture image to obtain a sitting posture detection result; if the sitting posture detection result indicates that the user's sitting posture is normal, then controlling the camera to start shooting once at a second pixel interval of a first preset time, wherein the first preset time is longer than the initial shooting interval of the camera.

4. The learning machine control method according to claim 3, characterized in that, If the posture detection result indicates that the user's posture is abnormal, the alarm of the learning machine will be activated to remind the user to adjust their posture; if the posture detection result indicates that the user's posture is abnormal again within a preset time, the learning machine will be put into sleep mode.

5. The learning machine control method according to claim 1, characterized in that, The step of performing position detection on the target object based on the distance value to obtain a position detection result further includes: if the position detection result indicates that the target object is outside the preset range of the learning machine, then controlling the distance sensor to start detection once every second preset time interval, wherein the second preset time interval is longer than the initial acquisition interval of the distance sensor.

6. The learning machine control method according to claim 1, characterized in that, The process of identifying the type of the target object based on its temperature value and obtaining a type identification result further includes: if the type identification result indicates that the target object is not a human body, then controlling the temperature sensor to turn off and controlling the distance sensor to turn on again.

7. The learning machine control method according to claim 6, characterized in that, If the type identification result indicates that the target object is not a human body, then the temperature sensor is turned off and the distance sensor is turned on again. The process further includes: acquiring the distance value between the current target object and the learning machine detected by the distance sensor; performing target identification on the current target object based on the distance value between the current target object and the learning machine to obtain a target identification result; if the target identification result indicates that the target object has changed, then performing position detection on the target object based on the latest acquired distance value.

8. A learning machine control device, characterized in that, include: The control unit is used to control the activation of the distance sensor based on the activation command from the learning machine; Obtain the distance value between the target object and the learning machine collected by the distance sensor; The target object is located based on the distance value to obtain a location detection result; if the location detection result indicates that the target object is within a preset range of the learning machine, the distance sensor is turned off and the temperature sensor is turned on. Obtain the temperature value of the target object detected by the temperature sensor; Based on the temperature value of the target object, the target object is identified as a type, and a type identification result is obtained; if the type identification result indicates that the target object is a human body, the temperature sensor and the distance sensor are turned off, and the camera is turned on to perform sitting posture detection.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the learning machine control method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the learning machine control method as described in any one of claims 1 to 7.

Images