A hibernation method and system applied to smart glasses

Abstract

The application provides a hibernation method and system applied to smart glasses, and comprises the following steps: S1, predicting a control rule of a user in a current use cycle based on historical control records of the smart glasses; S2, determining first prediction weights of each control mode at different time points in the current use cycle based on the control rule; S3, determining second prediction weights of each use posture at different time points in the current use cycle and hibernation judgment cycles of each control mode at different time points in the current use cycle based on the first prediction weights; and S4, setting a hibernation state of the smart glasses based on the hibernation judgment cycles, eye monitoring videos on the inner side of the smart glasses and the second prediction weights. The application overcomes the single judgment mode of the traditional hibernation state control method, realizes accurate setting of the hibernation state of the smart glasses, and reduces the damage to the smart glasses while ensuring the use experience of the user.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to a sleep method and system for smart glasses. Background Technology

[0002] Currently, with the intelligent development of wearable devices, smart glasses have become a popular emerging smart wearable device. Existing smart glasses can achieve a variety of functions, such as motion recognition, Internet communication, virtual reality interaction, voice recognition, and audio recording.

[0003] However, many users have poor usage habits and cannot manually control the sleep state of smart devices in a timely manner. In this case, the sleep state control methods of traditional smart glasses are mostly based on single motion monitoring, fixed settings, or environmental monitoring to determine whether the smart glasses have entered sleep state. Such sleep state control methods are not precise enough, which will lead to a decrease in the battery life of smart glasses or cause unnecessary damage to smart glasses.

[0004] Therefore, this invention proposes a sleep method and system for smart glasses. Summary of the Invention

[0005] This invention provides a sleep mode method and system for smart glasses. Based on the historical control records of the smart glasses, it predicts the user's control patterns, further predicts the prediction weight of each control method used by the user at different times, and further determines the sleep judgment cycle and the prediction weight of the usage posture at different times within the current usage cycle. This overcomes the drawback of the single judgment method in traditional sleep state control methods, and enables precise setting of the sleep state of smart glasses. In this way, it reduces wear and tear on smart glasses while ensuring the user experience.

[0006] This invention provides a sleep method for smart glasses, comprising:

[0007] S1: Predict the user's control patterns during the current usage cycle based on the historical control records of smart glasses;

[0008] S2: Determine the first prediction weight of each control method at different times within the current usage cycle based on the control law;

[0009] S3: Based on the first prediction weight, determine the second prediction weight of each usage posture at different times in the current usage cycle and the sleep judgment cycle of each control method at different times in the current usage cycle;

[0010] S4: Set the sleep state of the smart glasses based on the sleep judgment cycle, the eye monitoring video inside the smart glasses, and the second prediction weight.

[0011] Preferably, the sleep method applied to smart glasses, S1: predicting the user's control pattern during the current usage cycle based on the smart glasses' historical control records, includes:

[0012] S101: Based on the sub-historical records of each control method in the historical control records of the smart glasses in multiple usage cycles, determine the receiving time of the control command of the corresponding control method received by the smart glasses in each usage cycle, mark the receiving time of the corresponding control method in the cycle thread, and obtain the receiving thread of the corresponding control method in the corresponding usage cycle.

[0013] S102: Determine whether there is a time period clustering in the receiving times of all receiving threads of the corresponding control method, and obtain the determination result;

[0014] S103: Based on the judgment results, predict the user's control pattern in the current usage cycle.

[0015] Preferably, in the sleep method for smart glasses, S103: predicting the user's control pattern during the current usage cycle based on the judgment result, including:

[0016] When the judgment result is that there is a time period cluster in the receiving time of all receiving threads of the corresponding control method, the clustered time period of the corresponding control method is determined, and all the clustered time periods of the corresponding control method are sorted according to the time sequence to obtain the clustered time period sequence.

[0017] Based on the start and end times of each aggregation period in the aggregation period sequence, the start and end time change vectors between each aggregation period and the next adjacent aggregation period are determined. Based on the start and end time change vectors of each aggregation period, the predicted control time of the corresponding control method in the current usage cycle is predicted. The predicted control time of all control methods in the current usage cycle is taken as the control rule of the user in the current usage cycle.

[0018] Preferably, the sleep method for smart glasses predicts the predicted control time of the corresponding control mode in the current usage cycle based on the start time change vector and end time change vector of each aggregation period, including:

[0019] All start-time change vectors and all end-time change vectors are sorted according to time sequence and their coordinates are unified to obtain a unified result. Based on the unified result, the rate of change of the first vector at the end point of all start-time change vectors and the rate of change of the second vector at the end point of all end-time change vectors are determined.

[0020] The predicted start time of the corresponding control mode in the current operating cycle is determined based on the first vector change rate and the endpoint coordinate of the last start time change vector. The predicted end time of the corresponding control mode in the current operating cycle is determined based on the second vector change rate and the endpoint coordinate of the last end time change vector. The predicted control time of the corresponding control mode in the current operating cycle is determined based on the predicted start time and the predicted end time.

[0021] Preferably, in the sleep method for smart glasses, S103: predicting the user's control pattern during the current usage cycle based on the judgment result, including:

[0022] When the judgment result is that there is no time period clustering in the receiving time of all receiving threads of the corresponding control method, the control instruction with the smallest time difference with the corresponding receiving time is determined in the receiving thread of the next usage cycle based on the receiving time of each control instruction. The control instructions with the corresponding control instruction are then connected to obtain multiple receiving time fluctuation curves.

[0023] The first derivative function of the functional expression of the receiving time fluctuation curve is determined. Based on the value of the first derivative function in the current usage cycle and the last receiving time in the corresponding receiving time fluctuation curve, the predicted control time of the corresponding control mode in the current usage cycle is predicted. The predicted control time of all control modes in the current usage cycle is taken as the user's control law in the current usage cycle.

[0024] Preferably, in the sleep method for smart glasses, S2: determining the first prediction weight of each control mode at different times within the current usage cycle based on control laws, including:

[0025] Based on the predicted control time of each control method in the current usage cycle in the control law, the reciprocal of the total number of overlapping control methods is used as the first prediction weight of the overlapping control methods at the corresponding time. The first prediction weight of each control method at the time corresponding to the non-overlapping predicted control time is set to 1. The first prediction weight of each control method at different times in the remaining time of the current usage cycle other than the predicted control time of the corresponding control method is set to 0.

[0026] Preferably, the sleep method applied to smart glasses, S3: determining the second prediction weight of each usage posture at different times within the current usage cycle and the sleep judgment cycle of each control method at different times within the current usage cycle based on the first prediction weight, including:

[0027] The first predicted weight of each control method at different times within the current usage cycle is queried from the weight correspondence table to determine the second predicted weight of each usage posture at different times within the current usage cycle.

[0028] The product of the standard sleep cycle and the first prediction weight of the corresponding control method at the corresponding moment within the current usage cycle is used as the sleep judgment cycle of the corresponding control method at the corresponding moment within the current usage cycle.

[0029] Preferably, in the aforementioned sleep method for smart glasses, S4: based on the sleep judgment period, the eye monitoring video inside the smart glasses, and the second prediction weight, the sleep state of the smart glasses is set, including:

[0030] Based on the sleep judgment period of each control method at different times within the current usage cycle, the target judgment period of each control method at the current time is determined. It is then determined whether the smart glasses receive the control command of the corresponding control method within the latest target judgment period. If so, it is determined that the current time does not belong to the sleep period of the smart glasses. Otherwise, based on the second prediction weight of each usage posture at the current time and the eye monitoring video inside the smart glasses, it is determined whether the current time belongs to the sleep period of the smart glasses.

[0031] When it is determined that the current moment belongs to the sleep period of the smart glasses, the smart glasses are controlled to enter the sleep state.

[0032] Preferably, the sleep method for smart glasses, based on a second prediction weight for each usage posture at the current moment and eye monitoring video inside the smart glasses, determines whether the current moment belongs to the sleep period of the smart glasses, including:

[0033] The current posture data of the smart glasses is obtained by the posture sensor installed on the smart glasses, and the current posture data is used to query the posture list to determine the current posture of the smart glasses.

[0034] Based on the current posture and posture correlation coefficient table of the smart glasses, as well as the second prediction weight of each usage posture at the current moment, the probability coefficient of the smart glasses being worn at the current moment is determined.

[0035] When the probability coefficient is not lower than the coefficient threshold, blink detection is performed on the user based on the eye monitoring video inside the smart glasses to obtain the blink detection result. Based on the blink detection result, it is determined whether the user is in a closed eye state. If so, it is determined that the current moment belongs to the sleep period of the smart glasses; otherwise, it is determined that the current moment does not belong to the sleep period of the smart glasses.

[0036] When the probability coefficient is lower than the coefficient threshold, it is determined that the current moment belongs to the sleep period of the smart glasses.

[0037] This invention provides a sleep system for smart glasses, comprising:

[0038] The pattern prediction end is used to predict the user's control patterns during the current usage cycle based on the historical control records of smart glasses;

[0039] The first determining end is used to determine the first prediction weight of each control mode at different times within the current usage cycle based on the control law;

[0040] The second determining end is used to determine the second prediction weight of each usage posture at different times in the current usage cycle and the sleep judgment cycle of each control method at different times in the current usage cycle based on the first prediction weight.

[0041] The status setting terminal is used to set the sleep state of the smart glasses based on the sleep judgment cycle, the eye monitoring video inside the smart glasses, and the second prediction weight.

[0042] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0043] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0044] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0045] Figure 1 This is a flowchart of a sleep method applied to smart glasses according to an embodiment of the present invention;

[0046] Figure 2 This is a flowchart of another sleep method applied to smart glasses in an embodiment of the present invention;

[0047] Figure 3 This is a sleep system for smart glasses as described in an embodiment of the present invention. Detailed Implementation

[0048] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0049] Example 1:

[0050] This invention provides a sleep mode method for smart glasses, referencing... Figure 1 ,include:

[0051] S1: Predict the user's control patterns during the current usage cycle based on the historical control records of smart glasses;

[0052] S2: Determine the first prediction weight of each control method at different times within the current usage cycle based on the control law;

[0053] S3: Based on the first prediction weight, determine the second prediction weight of each usage posture at different times in the current usage cycle and the sleep judgment cycle of each control method at different times in the current usage cycle;

[0054] S4: Set the sleep state of the smart glasses based on the sleep judgment cycle, the eye monitoring video inside the smart glasses, and the second prediction weight.

[0055] In this embodiment, smart glasses are glasses devices equipped with a wealth of intelligent functions (motion recognition, internet communication, virtual reality interaction, voice recognition, audio recording, etc.).

[0056] In this embodiment, the historical control record is a record of control commands issued by the user received during multiple usage cycles of the smart glasses.

[0057] In this embodiment, the usage cycle refers to the periodicity of the user's habit of using glasses, such as 24 hours or a week.

[0058] In this embodiment, the current usage cycle is the usage cycle at the current moment.

[0059] In this embodiment, the control pattern is the pattern in which the user adopts each control method during the current usage cycle, specifically including the time or time period in which the user adopts each control method during the current usage cycle.

[0060] In this embodiment, the control method is the way the user controls the smart glasses, such as button control, voice control, motion control, etc.

[0061] In this embodiment, the first prediction weight is a value that represents the probability that the user will use the corresponding control method to control the smart glasses at the corresponding time in the current usage cycle.

[0062] In this embodiment, the posture used refers to the body posture of the user when wearing and using smart glasses, such as sitting, standing, or lying down.

[0063] In this embodiment, the second prediction weight is a value that represents the probability of a user being in each body posture at a corresponding moment in the current usage cycle when wearing and using smart glasses.

[0064] In this embodiment, the sleep judgment period is the judgment period threshold used to determine whether the smart glasses should enter a sleep state, and to determine whether the control command of the corresponding control method has been received in the latest judgment period. For example, if the user does not receive the control command of the corresponding control method in the latest sleep judgment period of each control method, further judgment is made based on the eye monitoring video inside the smart glasses and the second prediction weight; otherwise, it is determined that the current time is not the sleep period of the smart glasses.

[0065] In this embodiment, based on the first prediction weight, the second prediction weight of each usage posture at different times in the current usage cycle and the sleep judgment cycle of each control method at different times in the current usage cycle are determined. This allows for the prediction of the first prediction weight of each control method at the corresponding time by analyzing the user's control habits, and enables the setting of sleep judgment cycles based on the type of control method, thereby making the control of the sleep state of smart glasses more flexible and precise.

[0066] In this embodiment, the eye monitoring video is a monitoring video that includes the user's eye area, obtained from the inside of the smart glasses.

[0067] In this embodiment, the sleep state of the smart glasses is set based on the sleep judgment period, the eye monitoring video inside the smart glasses, and the second prediction weight, namely:

[0068] Based on the sleep judgment cycle, the eye monitoring video inside the smart glasses, and the second prediction weight, it is determined whether the current moment belongs to the sleep period of the smart glasses (i.e., the period when the smart glasses should enter the sleep state). If the current moment belongs to the sleep period of the smart glasses, the smart glasses are set to enter the sleep state; otherwise, the smart glasses are set not to enter the sleep state.

[0069] In this embodiment, the hibernation state is a state in which the currently running program of the smart glasses' operating system is saved in memory and the power is turned off.

[0070] The beneficial effects of the above technologies are as follows: based on the historical control records of smart glasses, the user's control patterns can be predicted, and the prediction weight of each control method adopted by the user at different times can be further predicted. Furthermore, the prediction weight of the sleep judgment cycle and usage posture at different times within the current usage cycle can be determined. This overcomes the drawback of the single judgment method of traditional sleep state control methods, realizes the precise setting of the sleep state of smart glasses, and reduces the wear and tear on smart glasses while ensuring the user experience.

[0071] Example 2:

[0072] Based on Example 1, the sleep mode method applied to smart glasses, S1: predicts the user's control pattern during the current usage cycle based on the historical control records of the smart glasses, referring to... Figure 2 ,include:

[0073] S101: Based on the sub-historical records of each control method in the historical control records of the smart glasses in multiple usage cycles, determine the receiving time of the control command of the corresponding control method received by the smart glasses in each usage cycle, mark the receiving time of the corresponding control method in the cycle thread, and obtain the receiving thread of the corresponding control method in the corresponding usage cycle.

[0074] S102: Determine whether there is a time period clustering in the receiving times of all receiving threads of the corresponding control method, and obtain the determination result;

[0075] S103: Based on the judgment results, predict the user's control pattern in the current usage cycle.

[0076] In this embodiment, the sub-historical record is the record of the control commands received by the smart glasses in the corresponding control mode during multiple usage cycles, which is included in the historical control record of the smart glasses.

[0077] In this embodiment, a periodic thread is a thread that represents the duration of a period of use.

[0078] In this embodiment, the receiving thread is the thread that marks the receiving time of the corresponding control method after the periodic thread, representing the time when the smart glasses receive the control command of the corresponding control method in the corresponding usage period.

[0079] In this embodiment, it is determined whether there is a time period clustering in the receiving times of all receiving threads of the corresponding control method, and the determination result is obtained, including:

[0080] Calculate the clustering degree of the receiving time in each receiving thread:

[0081]

[0082] In the formula, ε represents the clustering degree of the receiving moments in the currently calculated receiving thread, i represents the currently calculated receiving moment in the currently calculated receiving thread, n represents the total number of receiving moments included in the currently calculated receiving thread, and t i+1 For the (i+1)th receiving moment in the currently calculated receiving thread, t i This represents the i-th receiving moment in the currently calculated receiving thread, and T represents the duration of the usage period.

[0083] Determine whether the average clustering degree of the receiving time in all receiving threads exceeds the clustering degree threshold (that is, the average clustering degree of the receiving time in all receiving threads of the corresponding control method must be greater than the threshold). If so, determine that there is time-segment clustering in the receiving time of all receiving threads of the corresponding control method; otherwise, determine that there is no time-segment clustering in the receiving time of all receiving threads of the corresponding control method.

[0084] Based on the above formula, the clustering degree of the receiving time in each receiving thread is accurately calculated, and based on the average value of the clustering degree of the receiving time in all receiving threads, an accurate judgment result is obtained to determine whether there is time period clustering in the receiving time of all receiving threads of the corresponding control method.

[0085] The beneficial effects of the above technology are as follows: it enables the processing of the time when the control command of the corresponding control method is received in each usage cycle in the historical control records of smart glasses into a receiving thread, and further predicts the user's control pattern in the current usage cycle based on the result of judging whether there is a time period cluster in the receiving time in the receiving thread, thus realizing the processing and analysis of historical control records and the accurate prediction of the control pattern in the current usage cycle based on historical control records.

[0086] Example 3:

[0087] Based on Example 2, the sleep method applied to smart glasses, S103: predicting the user's control pattern during the current usage period based on the judgment result, including:

[0088] When the judgment result is that there is a time period cluster in the receiving time of all receiving threads of the corresponding control method, the clustered time period of the corresponding control method is determined, and all the clustered time periods of the corresponding control method are sorted according to the time sequence to obtain the clustered time period sequence.

[0089] Based on the start and end times of each aggregation period in the aggregation period sequence, the start and end time change vectors between each aggregation period and the next adjacent aggregation period are determined. Based on the start and end time change vectors of each aggregation period, the predicted control time of the corresponding control method in the current usage cycle is predicted. The predicted control time of all control methods in the current usage cycle is taken as the control rule of the user in the current usage cycle.

[0090] In this embodiment, the clustered time period is the time period in the receiving thread where the receiving time is clustered.

[0091] In this embodiment, the aggregation time period sequence is the sequence obtained by sorting all aggregation time periods of the corresponding control method according to the time sequence.

[0092] In this embodiment, based on the start and end times of each aggregation period in the aggregation period sequence, the start time change vector and end time change vector between each aggregation period and the next adjacent aggregation period in the aggregation period sequence are determined, namely:

[0093] The time interval between the aggregation period and the next adjacent aggregation period is used as the x-coordinate of the start time change vector and the end time change vector, respectively. The time interval between the start time of each aggregation period and the start time of the next adjacent aggregation period is used as the y-coordinate of the start time change vector, and the time interval between the end time of each aggregation period and the end time of the next adjacent aggregation period is used as the y-coordinate of the end time change vector. Thus, the start time change vector and the end time change vector between each aggregation period and the next adjacent aggregation period in the aggregation period sequence are determined.

[0094] In this embodiment, the predicted control time is the time during which the user controls the smart glasses using the corresponding control method within the current usage cycle, based on the start time change vector and end time change vector of each aggregation period.

[0095] The beneficial effects of the above technology are as follows: when the judgment result is that there is a time period cluster in the receiving time of all receiving threads of the corresponding control method, based on the analysis of the start time and end time of the clustered time period in each usage cycle according to the time sequence, the time when the user adopts each control method in the current usage cycle can be accurately predicted, and thus the control pattern of the user in the current usage cycle can be accurately predicted.

[0096] Example 4:

[0097] Based on Example 3, the sleep method applied to smart glasses predicts the predicted control time of the corresponding control mode in the current usage cycle based on the start time change vector and end time change vector of each aggregation period, including:

[0098] All start-time change vectors and all end-time change vectors are sorted according to time sequence and their coordinates are unified to obtain a unified result. Based on the unified result, the rate of change of the first vector at the end point of all start-time change vectors and the rate of change of the second vector at the end point of all end-time change vectors are determined.

[0099] The predicted start time of the corresponding control mode in the current operating cycle is determined based on the first vector change rate and the endpoint coordinate of the last start time change vector. The predicted end time of the corresponding control mode in the current operating cycle is determined based on the second vector change rate and the endpoint coordinate of the last end time change vector. The predicted control time of the corresponding control mode in the current operating cycle is determined based on the predicted start time and the predicted end time.

[0100] In this embodiment, the unified result is the result obtained by sorting all start time change vectors and all end time change vectors according to time sequence and then unifying their coordinates.

[0101] In this embodiment, coordinate unification means unifying the starting point of all start time change vectors and all end time change vectors after sorting them according to time sequence to the origin of a unified preset coordinate system.

[0102] In this embodiment, based on the unified result, the rate of change of the first vector at the end of all starting time change vectors and the rate of change of the second vector at the end of all ending time change vectors are determined, namely:

[0103] Based on the unified results, the coordinates of the endpoints of all start-time change vectors and all end-time change vectors in the preset coordinate system are determined. The vector pointing from the origin of the preset coordinate system to the endpoint coordinates of the corresponding start-time change vector or end-time change vector is taken as the position vector of the endpoint of the corresponding start-time change vector or end-time change vector.

[0104] The rate of change of the position vectors of all start-time change vectors after being sorted by time sequence is determined as the first rate of change, and the rate of change of the position vectors of all end-time change vectors after being sorted by time sequence is determined as the second rate of change.

[0105] In this embodiment, the predicted start time of the corresponding control mode in the current usage cycle is determined based on the first vector change rate and the endpoint coordinates of the last start time change vector, including:

[0106] Based on the first vector change rate and the endpoint coordinate of the last start time change vector, the start time of the last clustering period in the clustering period sequence is determined, pointing to the predicted start time of the predicted control period of the corresponding control mode in the current usage cycle. Based on the predicted start time change vector and the start time of the last clustering period in the clustering period sequence, the predicted start time of the predicted control period of the corresponding control mode in the current usage cycle is determined.

[0107] In this embodiment, the predicted termination time of the corresponding control mode in the current usage cycle is determined based on the second position vector change rate and the endpoint coordinates of the change vector at the last termination time, including:

[0108] Based on the second position vector change rate and the endpoint coordinates of the last termination time change vector, the termination time prediction change vector of the last aggregation period in the aggregation period sequence is determined, pointing to the predicted termination time of the predicted control period of the corresponding control mode in the current usage cycle. Based on the termination time prediction change vector and the termination time of the last aggregation period in the aggregation period sequence, the predicted termination time of the predicted control period of the corresponding control mode in the current usage cycle is determined.

[0109] In this embodiment, the prediction control time is the duration from the prediction start time to the prediction end time.

[0110] The beneficial effects of the above technology are as follows: by analyzing the pattern of the position vectors of the starting time change vector and the ending time change vector of each aggregation period, the predicted start time and predicted end time of the corresponding control method in the current usage cycle can be predicted, thereby accurately predicting the time when the user adopts each control method in the current usage cycle.

[0111] Example 5:

[0112] Based on Example 2, the sleep method applied to smart glasses, S103: predicting the user's control pattern during the current usage period based on the judgment result, including:

[0113] When the judgment result is that there is no time period clustering in the receiving time of all receiving threads of the corresponding control method, the control instruction with the smallest time difference with the corresponding receiving time is determined in the receiving thread of the next usage cycle based on the receiving time of each control instruction. The control instructions with the corresponding control instruction are then connected to obtain multiple receiving time fluctuation curves.

[0114] The first derivative function of the functional expression of the receiving time fluctuation curve is determined. Based on the value of the first derivative function in the current usage cycle and the last receiving time in the corresponding receiving time fluctuation curve, the predicted control time of the corresponding control mode in the current usage cycle is predicted. The predicted control time of all control modes in the current usage cycle is taken as the user's control law in the current usage cycle.

[0115] In this embodiment, the receiving time fluctuation curve is a curve that includes one receiving time in each receiving thread and represents the fluctuation of the receiving time of the control command corresponding to the control mode with the usage cycle.

[0116] In this embodiment, based on the value of the first derivative function in the current usage cycle and the last reception moment in the corresponding reception time fluctuation curve, the predicted control time of the corresponding control method in the current usage cycle is predicted, which is:

[0117] The first derivative value of the first derivative function in the current usage period is determined as the corresponding value (for example, if the first derivative function is y = 0.5t, where t represents the current usage period, then when the current usage period is the 4th usage period, the value of the first derivative function in the current usage period is 2). The total number of usage periods between the usage period to which the last reception moment in the corresponding reception time fluctuation curve belongs and the current usage period is determined. The sum of the corresponding values ​​of the last reception moment in the corresponding reception time fluctuation curve and the total number of usage periods is taken as the predicted control moment of the corresponding control mode in the current usage period, and the predicted control moment is taken as the corresponding predicted control time.

[0118] The beneficial effects of the above technology are as follows: when the judgment result is that there is no time period clustering in the reception time of all receiving threads of the corresponding control method, the reception time of each control instruction in the receiving thread is tracked and analyzed according to the time sequence, and the predicted control time of the corresponding control method in the current usage cycle is accurately predicted. In addition, the time when the user adopts each control method in the current usage cycle is accurately predicted, and the control pattern of the user in the current usage cycle is accurately predicted.

[0119] Example 6:

[0120] Based on Embodiment 3 or 5, the sleep method applied to smart glasses, S2: determining the first prediction weight of each control mode at different times within the current usage cycle based on control laws, including:

[0121] Based on the predicted control time of each control method in the current usage cycle in the control law, the reciprocal of the total number of overlapping control methods is used as the first prediction weight of the overlapping control methods at the corresponding time. The first prediction weight of each control method at the time corresponding to the non-overlapping predicted control time is set to 1. The first prediction weight of each control method at different times in the remaining time of the current usage cycle other than the predicted control time of the corresponding control method is set to 0.

[0122] The beneficial effect of the above technology is that by judging whether the predicted control time of each control method in the current usage cycle overlaps, the probability of the user adopting each control method at different times in the current usage cycle can be determined.

[0123] Example 7:

[0124] Based on Example 1, the sleep method applied to smart glasses, S3: determining the second prediction weight of each usage posture at different times within the current usage cycle and the sleep judgment cycle of each control method at different times within the current usage cycle based on the first prediction weight, including:

[0125] The first predicted weight of each control method at different times within the current usage cycle is queried from the weight correspondence table to determine the second predicted weight of each usage posture at different times within the current usage cycle.

[0126] The product of the standard sleep cycle and the first prediction weight of the corresponding control method at the corresponding moment within the current usage cycle is used as the sleep judgment cycle of the corresponding control method at the corresponding moment within the current usage cycle.

[0127] In this embodiment, the weight correspondence table is a list containing the correspondence between the first predicted weight of each control mode at the same time and the second predicted weight of each usage posture at the corresponding time.

[0128] The beneficial effects of the above technologies are as follows: First prediction weights based on the probability that a user will use each control method to control the smart glasses at a corresponding time are realized; second prediction weights for each usage posture at different times within the current usage cycle and sleep judgment cycles for each control method at different times within the current usage cycle are determined; and sleep judgment cycles are determined based on each control method, thereby further ensuring the timeliness and accuracy of sleep state control.

[0129] Example 8:

[0130] Based on Example 1, the sleep mode method applied to smart glasses, S4: Based on the sleep judgment period, the eye monitoring video inside the smart glasses, and the second prediction weight, the sleep state of the smart glasses is set, including:

[0131] Based on the sleep judgment period of each control method at different times within the current usage cycle, the target judgment period of each control method at the current time is determined. It is then determined whether the smart glasses receive the control command of the corresponding control method within the latest target judgment period. If so, it is determined that the current time does not belong to the sleep period of the smart glasses. Otherwise, based on the second prediction weight of each usage posture at the current time and the eye monitoring video inside the smart glasses, it is determined whether the current time belongs to the sleep period of the smart glasses.

[0132] When it is determined that the current moment belongs to the sleep period of the smart glasses, the smart glasses are controlled to enter the sleep state.

[0133] In this embodiment, the target judgment period is the sleep judgment period of the corresponding control method at the current moment of the current usage period.

[0134] In this embodiment, the latest target judgment period is the duration from the current time back to one corresponding target judgment period. For example, if the current time is 9 o'clock and the corresponding target judgment period is 2 hours, then the latest target judgment period is the time period from 7 o'clock to 9 o'clock.

[0135] In this embodiment, the sleep period is the period during which the smart glasses should enter a sleep state.

[0136] The beneficial effects of the above technology are as follows: by determining whether the smart glasses have received the control command of the corresponding control method within the latest target judgment period, the first stage of the judgment is realized to determine whether the current moment belongs to the sleep period of the smart glasses, thus realizing the judgment of whether the smart glasses should be controlled to enter the sleep state.

[0137] Example 9:

[0138] Based on Example 8, the sleep method for smart glasses, which determines whether the current moment belongs to the sleep period of the smart glasses based on the second prediction weight of each usage posture at the current moment and the eye monitoring video inside the smart glasses, includes:

[0139] The current posture data of the smart glasses is obtained by the posture sensor installed on the smart glasses, and the current posture data is used to query the posture list to determine the current posture of the smart glasses.

[0140] Based on the current posture and posture correlation coefficient table of the smart glasses, as well as the second prediction weight of each usage posture at the current moment, the probability coefficient of the smart glasses being worn at the current moment is determined.

[0141] When the probability coefficient is not lower than the coefficient threshold, blink detection is performed on the user based on the eye monitoring video inside the smart glasses to obtain the blink detection result. Based on the blink detection result, it is determined whether the user is in a closed eye state. If so, it is determined that the current moment belongs to the sleep period of the smart glasses; otherwise, it is determined that the current moment does not belong to the sleep period of the smart glasses.

[0142] When the probability coefficient is lower than the coefficient threshold, it is determined that the current moment belongs to the sleep period of the smart glasses.

[0143] In this embodiment, the attitude sensor is a sensor used to detect the real-time attitude data of the smart glasses.

[0144] In this embodiment, the current posture data is the data representing the current posture of the smart glasses obtained based on the posture sensor, such as the tilt angle of the smart glasses lenses.

[0145] In this embodiment, the posture list is a list containing the correspondence between the current posture data and the current posture of the smart glasses.

[0146] In this embodiment, the current posture of the smart glasses is the posture corresponding to different angles formed by the lens and the parallel bottom surface.

[0147] In this embodiment, the posture correlation coefficient table is a preset list containing the correlation coefficients between each posture of the smart glasses and each usage posture of the user.

[0148] In this embodiment, based on the current posture of the smart glasses and the posture correlation coefficient table, as well as the second prediction weight of each usage posture at the current moment, the probability coefficients of the smart glasses being worn at the current moment are determined, including:

[0149] Based on the posture correlation coefficient table, the current posture of the smart glasses and the sub-correlation coefficients for each usage posture are determined. Based on the sub-correlation coefficients and the second prediction weight of each usage posture at the current moment, the probability coefficients of the smart glasses being worn at the current moment are calculated.

[0150]

[0151] In the formula, β is the probability coefficient that the smart glasses are being worn at the current moment, j is the j-th usage posture with a non-zero second prediction weight at the current moment, m is the total number of usage postures with a non-zero second prediction weight at the current moment, and g j Let α be the sub-correlation coefficient between the current pose of the smart glasses and the j-th usage pose. j The second prediction weight for the j-th usage posture at the current time;

[0152] The above formula accurately calculates the probability coefficient of the smart glasses being worn at the current moment based on the current posture of the smart glasses, the sub-correlation coefficient of each usage posture, and the second prediction weight of each usage posture at the current moment.

[0153] In this embodiment, the coefficient threshold is the minimum possible coefficient required to detect blinking of the user based on the eye monitoring video inside the smart glasses.

[0154] In this embodiment, blink detection is to determine the user's blink frequency per unit time based on the eye monitoring video inside the smart glasses.

[0155] In this embodiment, the blink detection result is the blink frequency of the user per unit time.

[0156] In this embodiment, the user is determined to be in a closed-eye state based on the blink detection results. Specifically, if the blink detection frequency in the blink detection results is lower than a preset blink frequency threshold, the user is determined to be in a closed-eye state; otherwise, the user is determined not to be in a closed-eye state.

[0157] The beneficial effects of the above technology are as follows: Based on the current posture of the smart glasses and the second prediction weight of each usage posture at the current moment, as well as the posture correlation coefficient table, the probability coefficient of the smart glasses being worn at the current moment is determined. Combined with the blink detection results obtained by the user blink detection based on the eye monitoring video on the inside of the smart glasses, it is determined whether the current moment belongs to the sleep period of the smart glasses, thus realizing the judgment of whether the smart glasses should be controlled to enter the sleep state.

[0158] Example 10:

[0159] This invention provides a sleep system for smart glasses, referenced Figure 3 ,include:

[0160] The pattern prediction end is used to predict the user's control patterns during the current usage cycle based on the historical control records of smart glasses;

[0161] The first determining end is used to determine the first prediction weight of each control mode at different times within the current usage cycle based on the control law;

[0162] The second determining end is used to determine the second prediction weight of each usage posture at different times in the current usage cycle and the sleep judgment cycle of each control method at different times in the current usage cycle based on the first prediction weight.

[0163] The status setting terminal is used to set the sleep state of the smart glasses based on the sleep judgment cycle, the eye monitoring video inside the smart glasses, and the second prediction weight.

[0164] The beneficial effects of the above technologies are as follows: based on the historical control records of smart glasses, the user's control patterns can be predicted, and the prediction weight of each control method adopted by the user at different times can be further predicted. Furthermore, the prediction weight of the sleep judgment cycle and usage posture at different times within the current usage cycle can be determined. This overcomes the drawback of the single judgment method of traditional sleep state control methods, realizes the precise setting of the sleep state of smart glasses, and reduces the wear and tear on smart glasses while ensuring the user experience.

[0165] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A sleep mode method for smart glasses, characterized in that, include: S1: Predict the user's control patterns during the current usage cycle based on the historical control records of smart glasses; S2: Determine the first prediction weight of each control method at different times within the current usage cycle based on the control law; S3: Based on the first prediction weight, determine the second prediction weight of each usage posture at different times in the current usage cycle and the sleep judgment cycle of each control method at different times in the current usage cycle; S4: Set the sleep state of the smart glasses based on the sleep judgment cycle, the eye monitoring video inside the smart glasses, and the second prediction weight; S1: Based on the historical control records of the smart glasses, predict the user's control patterns during the current usage cycle, including: S101: Based on the sub-historical records of each control method in the historical control records of the smart glasses in multiple usage cycles, determine the receiving time of the control command of the corresponding control method received by the smart glasses in each usage cycle, mark the receiving time of the corresponding control method in the cycle thread, and obtain the receiving thread of the corresponding control method in the corresponding usage cycle. S102: Determine whether there is a time period clustering in the receiving times of all receiving threads of the corresponding control method, and obtain the determination result; S103: Predict the user's control patterns during the current usage cycle based on the judgment results; Wherein, S2: Based on the control law, the first prediction weight of each control mode at different times within the current usage cycle is determined, including: Based on the predicted control time of each control method in the current usage cycle in the control law, the reciprocal of the total number of overlapping control methods is used as the first prediction weight of the overlapping control methods at the corresponding time. The first prediction weight of each control method at the time corresponding to the non-overlapping predicted control time is set to 1. The first prediction weight of each control method at different times in the remaining time of the current usage cycle, excluding the predicted control time of the corresponding control method, is set to 0. Wherein, S3: Based on the first prediction weight, the second prediction weight of each usage posture at different times within the current usage cycle and the sleep judgment cycle of each control method at different times within the current usage cycle are determined, including: The first predicted weight of each control method at different times within the current usage cycle is queried from the weight correspondence table to determine the second predicted weight of each usage posture at different times within the current usage cycle. The product of the standard sleep cycle and the first prediction weight of the corresponding control method at the corresponding moment within the current usage cycle is used as the sleep judgment cycle of the corresponding control method at the corresponding moment within the current usage cycle. S4: Based on the sleep judgment period, the eye monitoring video inside the smart glasses, and the second prediction weight, the sleep state of the smart glasses is set, including: Based on the sleep judgment period of each control method at different times within the current usage cycle, the target judgment period of each control method at the current time is determined. It is then determined whether the smart glasses receive the control command of the corresponding control method within the latest target judgment period. If so, it is determined that the current time does not belong to the sleep period of the smart glasses. Otherwise, based on the second prediction weight of each usage posture at the current time and the eye monitoring video inside the smart glasses, it is determined whether the current time belongs to the sleep period of the smart glasses. When it is determined that the current moment belongs to the sleep period of the smart glasses, the smart glasses are controlled to enter the sleep state.

2. The sleep method for smart glasses according to claim 1, characterized in that, S103: Based on the judgment result, predict the user's control patterns during the current usage cycle, including: When the judgment result is that there is a time period cluster in the receiving time of all receiving threads of the corresponding control method, the clustered time period of the corresponding control method is determined, and all the clustered time periods of the corresponding control method are sorted according to the time sequence to obtain the clustered time period sequence. Based on the start and end times of each aggregation period in the aggregation period sequence, the start and end time change vectors between each aggregation period and the next adjacent aggregation period are determined. Based on the start and end time change vectors of each aggregation period, the predicted control time of the corresponding control method in the current usage cycle is predicted. The predicted control time of all control methods in the current usage cycle is taken as the control rule of the user in the current usage cycle.

3. The sleep method for smart glasses according to claim 2, characterized in that, Based on the start and end time change vectors of each aggregation period, the predicted control time for the corresponding control mode in the current usage cycle is predicted, including: All start-time change vectors and all end-time change vectors are sorted according to time sequence and their coordinates are unified to obtain a unified result. Based on the unified result, the rate of change of the first vector at the end point of all start-time change vectors and the rate of change of the second vector at the end point of all end-time change vectors are determined. The predicted start time of the corresponding control mode in the current operating cycle is determined based on the first vector change rate and the endpoint coordinate of the last start time change vector. The predicted end time of the corresponding control mode in the current operating cycle is determined based on the second vector change rate and the endpoint coordinate of the last end time change vector. The predicted control time of the corresponding control mode in the current operating cycle is determined based on the predicted start time and the predicted end time.

4. The sleep method for smart glasses according to claim 1, characterized in that, S103: Based on the judgment result, predict the user's control patterns during the current usage cycle, including: When the judgment result is that there is no time period clustering in the receiving time of all receiving threads of the corresponding control method, the control instruction with the smallest time difference with the corresponding receiving time is determined in the receiving thread of the next usage cycle based on the receiving time of each control instruction. The control instructions with the corresponding control instruction are then connected to obtain multiple receiving time fluctuation curves. The first derivative function of the functional expression of the receiving time fluctuation curve is determined. Based on the value of the first derivative function in the current usage cycle and the last receiving time in the corresponding receiving time fluctuation curve, the predicted control time of the corresponding control mode in the current usage cycle is predicted. The predicted control time of all control modes in the current usage cycle is taken as the user's control law in the current usage cycle.

5. The sleep method for smart glasses according to claim 1, characterized in that, Based on the second prediction weight of each usage posture at the current moment and the eye monitoring video inside the smart glasses, it is determined whether the current moment belongs to the sleep period of the smart glasses, including: The current posture data of the smart glasses is obtained by the posture sensor installed on the smart glasses, and the current posture data is used to query the posture list to determine the current posture of the smart glasses. Based on the current posture and posture correlation coefficient table of the smart glasses, as well as the second prediction weight of each usage posture at the current moment, the probability coefficient of the smart glasses being worn at the current moment is determined. When the probability coefficient is not lower than the coefficient threshold, blink detection is performed on the user based on the eye monitoring video inside the smart glasses to obtain the blink detection result. Based on the blink detection result, it is determined whether the user is in a closed eye state. If so, it is determined that the current moment belongs to the sleep period of the smart glasses; otherwise, it is determined that the current moment does not belong to the sleep period of the smart glasses. When the probability coefficient is lower than the coefficient threshold, it is determined that the current moment belongs to the sleep period of the smart glasses.

6. A sleep system for smart glasses, characterized in that, A method for performing a sleep mode applied to smart glasses according to any one of claims 1 to 5, comprising: The pattern prediction end is used to predict the user's control patterns during the current usage cycle based on the historical control records of smart glasses; The first determining end is used to determine the first prediction weight of each control mode at different times within the current usage cycle based on the control law; The second determining end is used to determine the second prediction weight of each usage posture at different times in the current usage cycle and the sleep judgment cycle of each control method at different times in the current usage cycle based on the first prediction weight. The status setting terminal is used to set the sleep state of the smart glasses based on the sleep judgment cycle, the eye monitoring video inside the smart glasses, and the second prediction weight.

Images