CN119694070B - Remote monitoring methods and children's watches - Google Patents
Remote monitoring methods and children's watchesInfo
- Publication number
- CN119694070B CN119694070B CN202411841247.6A CN202411841247A CN119694070B CN 119694070 B CN119694070 B CN 119694070B CN 202411841247 A CN202411841247 A CN 202411841247A CN 119694070 B CN119694070 B CN 119694070B
- Authority
- CN
- China
- Prior art keywords
- mobile terminal
- data
- current
- distance threshold
- sound
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Telephone Function (AREA)
Abstract
The invention provides a remote monitoring method and a child watch, which relate to the technical field of monitoring and comprise the steps of extracting motion characteristics and sound characteristics of a mobile terminal from various sensor data, generating a current track, identifying a behavior mode of a corresponding user of the mobile terminal through a preset behavior mode library based on the motion characteristics and the current track, obtaining the current behavior mode, inputting the motion characteristics and the sound characteristics into a pre-trained prediction model, judging whether a mobile object is approaching the mobile terminal or not by using the prediction model, dynamically adjusting a preset distance threshold based on the motion characteristics, the sound characteristics and the current track at the current moment when the mobile object is judged to be approaching the mobile terminal, judging whether the relative distance between the mobile object and the mobile terminal is smaller than the dynamically adjusted distance threshold, generating alarm information when the relative distance is smaller than the dynamically adjusted distance threshold, and pushing the alarm information to an associated terminal pre-paired with the mobile terminal.
Description
Technical Field
The invention relates to the technical field of monitoring, in particular to a remote monitoring method and a child watch.
Background
With the increasing concern of society for child safety, technologies and products directed to child safety monitoring are continuously emerging. The portable terminals such as the child watch are used as daily wearing equipment for children, have basic functions such as communication and positioning, and gradually integrate with a remote monitoring technology so as to realize more comprehensive and careful safety monitoring for the children.
The traditional child watch monitoring method mainly relies on GPS positioning technology, and parents can know the track of a child at any time by acquiring the position information of the child in real time. However, in complex environments (such as indoor environments, high-rise dense areas and the like), the positioning accuracy is often seriously affected, and accurate and reliable child position information cannot be provided for parents.
Accordingly, there is a need to provide a remote monitoring method and a child watch that address the above-described issues.
Disclosure of Invention
In order to solve the technical problems, the invention provides a remote monitoring method and a child watch, and provides a more intelligent and personalized remote monitoring method so as to meet the higher requirements of parents on child safety monitoring.
The invention provides a remote monitoring method, which comprises the following steps:
extracting motion features and sound features of a mobile terminal from a plurality of sensor data, wherein the plurality of sensor data are collected by a sensor assembly mounted on the mobile terminal, and generating a current track based on the motion features;
based on the motion characteristics and the current track, identifying the behavior mode of the corresponding user of the mobile terminal through a preset behavior mode library to obtain a current behavior mode;
Inputting the motion characteristics and the sound characteristics into a pre-trained prediction model, and judging whether a mobile object is approaching to the mobile terminal or not by utilizing the prediction model;
When the mobile object is judged to be approaching the mobile terminal, dynamically adjusting a preset distance threshold value based on the motion characteristic, the sound characteristic and the current track at the current moment, wherein the distance threshold value represents the safety distance which should be kept between the mobile terminal and the mobile object in the current behavior mode;
And judging whether the relative distance between the mobile object and the mobile terminal is smaller than a dynamically adjusted distance threshold value, and generating alarm information and pushing the alarm information to an associated terminal pre-paired with the mobile terminal when the relative distance is smaller than the dynamically adjusted distance threshold value.
Preferably, the plurality of sensor data includes accelerometer data, gyroscope data, GPS data, and sound data, wherein the accelerometer data, the gyroscope data, and the GPS data are used to extract the motion feature, and the sound data is used to extract the sound feature, wherein the motion feature includes an accelerometer data feature, a gyroscope data feature, and a GPS data feature corresponding to the accelerometer data, the gyroscope data, and the GPS data, respectively.
Preferably, the generating step of the current track includes:
And generating the current track of the mobile terminal by using a map matching algorithm based on the extracted accelerometer data characteristics and gyroscope data characteristics and combining the position change information and the timestamp data in the GPS data characteristics.
Preferably, the identifying, based on the motion feature and the current track, the behavior pattern of the corresponding user of the mobile terminal through a preset behavior pattern library, to obtain a current behavior pattern includes:
Extracting a plurality of behavior pattern related feature vectors from the accelerometer data features, gyroscope data features and GPS data features, wherein the feature vectors comprise acceleration, angular velocity, speed, direction and speed change rate;
matching the feature vector with a plurality of behavior patterns in a preset behavior pattern library, wherein the behavior pattern library comprises a plurality of known behavior patterns and corresponding feature vectors thereof;
And calculating the similarity between the feature vector and the feature vector of each known behavior mode, and selecting the known behavior mode with the highest similarity as the current behavior mode.
Preferably, the inputting the motion feature and the sound feature into a pre-trained prediction model, and determining whether a moving object is approaching the mobile terminal by using the prediction model includes:
Performing fusion processing on the extracted motion characteristics and the extracted sound characteristics to form fusion characteristic vectors;
inputting the fusion feature vector into a pre-trained prediction model, wherein the prediction model is a neural network model which is identified through training and analyzes the fusion feature vector to judge whether a moving object is approaching a mobile terminal;
outputting a judging result, and triggering a subsequent distance threshold dynamic adjustment step if the judging result is that the moving object is close.
Preferably, the training process of the prediction model includes:
collecting a labeling data set containing motion characteristics, sound characteristics and corresponding moving object approaching conditions;
Training a neural network model by utilizing the data set, and adjusting model parameters in the training process to minimize the prediction error rate;
and evaluating the performance of the neural network model by a cross-validation method, and finally storing the trained model as a pre-trained prediction model.
Preferably, when the mobile object is determined to be approaching the mobile terminal, the dynamically adjusting the preset distance threshold based on the motion feature, the sound feature and the current track at the current moment includes:
adjusting a preset basic distance threshold to obtain a preliminary distance threshold, wherein the adjustment adopts the following formula:
Wherein, T p is a preliminary distance threshold, T 0 is a basic distance threshold, F v is a speed adjustment factor, F d is a direction adjustment factor, and F n is a noise adjustment factor;
Calculating a behavior pattern adjustment factor, adjusting the preliminary distance threshold based on the behavior pattern adjustment factor to obtain a final dynamic threshold, and taking the final dynamic threshold as an adjusted distance threshold, wherein the adjustment of the preliminary distance threshold adopts the following formula:
Wherein T f is the final dynamic threshold, F b is the behavior pattern adjustment factor, v is the speed of the mobile terminal at the current time, and v max is the maximum speed in the current track of the mobile terminal.
Preferably, in the adjustment of the basic distance threshold, a calculation formula of the speed adjustment factor is:
Wherein F v is a speed adjustment factor, v is the speed of the mobile terminal at the current moment, deltav is the speed change rate in the current track, and alpha is the influence coefficient of the preset speed change rate;
The calculation formula of the direction adjustment factor is as follows:
Wherein F d is a direction adjustment factor, d is the direction of the mobile terminal at the current moment, Δd is the direction change rate in the current track, and β is the influence coefficient of the preset direction change rate;
the calculation formula of the noise adjustment factor is as follows:
Wherein F n is a noise adjustment factor, n is an environmental noise level of the mobile terminal at the current moment, deltan is a noise change rate in the current track, and gamma is an influence coefficient of a preset noise change rate.
Preferably, the calculation formula of the behavior pattern adjustment factor is:
Fb＝Kb+δ×Δb
Wherein F b is a behavior pattern adjustment factor, Δb is a behavior pattern change rate, and represents a ratio of the number of times the user behavior pattern changes in the current track to the total time, δ is an influence coefficient of a preset behavior pattern change rate, and K b is a behavior pattern adjustment coefficient.
The invention also provides a child watch comprising a sensor assembly and a processor,
The sensor assembly is for a variety of sensor data;
The processor is used for extracting the motion characteristics and sound characteristics of the mobile terminal from various sensor data and generating a current track based on the motion characteristics;
The processor is used for identifying the behavior mode of the corresponding user of the mobile terminal through a preset behavior mode library based on the motion characteristics and the current track to obtain a current behavior mode;
the processor is used for inputting the motion characteristics and the sound characteristics into a pre-trained prediction model, and judging whether a moving object is approaching to the mobile terminal or not by utilizing the prediction model;
The processor is used for dynamically adjusting a preset distance threshold value based on the motion characteristic, the sound characteristic and the current track at the current moment when the mobile object is judged to be approaching the mobile terminal, wherein the distance threshold value represents the safety distance which should be kept between the mobile terminal and the mobile object in the current behavior mode;
The processor is used for judging whether the relative distance between the mobile object and the mobile terminal is smaller than a distance threshold after dynamic adjustment, and generating alarm information and pushing the alarm information to an associated terminal pre-paired with the mobile terminal when the relative distance is smaller than the distance threshold.
Compared with the related art, the remote monitoring method and the child watch provided by the invention have the following beneficial effects:
according to the invention, the motion characteristics and the sound characteristics of the mobile terminal are extracted from various sensor data, and the current track and the behavior pattern recognition result are generated based on the characteristics, so that the accurate perception of the user behavior pattern is realized. Meanwhile, the method also utilizes a pre-trained prediction model to intelligently judge the approaching condition of the moving object, and dynamically adjusts the distance threshold according to the judging result, thereby realizing personalized monitoring of the safety distance under different behavioral modes.
Drawings
FIG. 1 is a flow chart of a remote monitoring method provided by the invention;
fig. 2 is a block diagram of a child watch according to the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings. Furthermore, embodiments of the invention and features of the embodiments may be combined with each other without conflict.
It should be further noted that, for convenience of description, only some, but not all of the matters related to the present invention are shown in the accompanying drawings. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart depicts operations (or steps) as a sequential process, many of the operations can be performed in parallel, concurrently, or at the same time. Furthermore, the order of the operations may be rearranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figures. The processes may correspond to methods, functions, procedures, subroutines, and the like.
Example 1
The invention provides a remote monitoring method, which is shown by referring to fig. 1, and comprises the following steps:
s1, extracting motion characteristics and sound characteristics of a mobile terminal from various sensor data, and generating a current track based on the motion characteristics, wherein the various sensor data are collected by a sensor assembly installed on the mobile terminal.
The various sensor data includes accelerometer data, gyroscope data, GPS data and sound data, wherein the accelerometer data, the gyroscope data and the GPS data are used for extracting the motion characteristics, and the sound data is used for extracting the sound characteristics, wherein the motion characteristics include accelerometer data characteristics, gyroscope data characteristics and GPS data characteristics corresponding to the accelerometer data, the gyroscope data and the GPS data, respectively.
In this embodiment, the accelerometer is derived from an accelerometer mounted on the mobile terminal, and linear accelerations in three axes (X, Y, Z) can be measured. Accelerometer data is mainly used for extracting linear motion characteristics of the mobile terminal. The acceleration magnitude can be obtained by calculating the square root of the sum of squares of acceleration values in three axes. The acceleration direction is represented by dividing the acceleration value in each axis by the magnitude of the acceleration to obtain a unit vector. The acceleration change rate is obtained by dividing the acceleration difference between two consecutive time points by the time interval.
The gyroscope data is derived from a gyroscope mounted on the mobile terminal, and angular velocities in three axes (X, Y, Z) can be measured. The gyroscope data are mainly used for extracting the rotation motion characteristics of the mobile terminal. The angular velocity magnitude can be obtained by calculating the square root of the sum of squares of the angular velocity values in the three axes. The angular velocity direction is represented by dividing the angular velocity value on each axis by the magnitude of the angular velocity to obtain a unit vector. The rate of change of angular velocity is obtained by dividing the difference in angular velocity between two successive time points by the time interval.
The GPS data originates from a GPS receiver mounted on the mobile terminal and can provide accurate geographical location information. The GPS data is mainly used for extracting the location and speed characteristics of the mobile terminal. The location information includes longitude, latitude, and altitude of the mobile terminal. The speed information includes an instantaneous speed of the mobile terminal. The direction information includes a heading angle of the mobile terminal. The rate of change of position is then obtained by dividing the difference between longitude, latitude and altitude at two consecutive time points by the time interval.
The sound data originates from a microphone mounted on the mobile terminal and can collect audio signals of the surrounding environment. The sound data is mainly used for extracting acoustic environmental features around the mobile terminal, and the environmental noise level can be obtained by calculating the average sound pressure level of the sound signal.
From the above detailed description, it can be seen that the accelerometer data provides linear motion characteristics of the mobile terminal, including the magnitude, direction and rate of change of acceleration. The gyroscope data provides the rotational motion characteristics of the mobile terminal, including the magnitude, direction and rate of change of angular velocity. The GPS data provides location and speed characteristics of the mobile terminal, including longitude, latitude, altitude, speed and direction. The sound data provides acoustic environmental features around the mobile terminal, i.e. the level of environmental noise, which together constitute the motion features and sound features of the mobile terminal, providing a rich data support for subsequent behavior pattern recognition and environmental monitoring.
Meanwhile, in step S1, the step of generating the current track includes:
And generating the current track of the mobile terminal by using a map matching algorithm based on the extracted accelerometer data characteristics and gyroscope data characteristics and combining the position change information and the timestamp data in the GPS data characteristics.
In this embodiment, the collected accelerometer data, gyroscope data and GPS data are filtered, noise and outliers are removed, and accuracy and consistency of the data are ensured.
The accelerometer data features and the gyroscope data features are combined to form a comprehensive motion feature vector. This vector includes information such as acceleration, angular velocity, speed and direction.
And matching the comprehensive motion characteristic vector with map data by utilizing the position and the timestamp information in the GPS data. The specific method comprises the following steps:
And estimating the path of the mobile terminal between two GPS points based on the acceleration and angular speed data.
Map matching, namely matching the estimated path with a road network on a map, and correcting errors caused by GPS signal drift.
Track optimization, namely combining speed and direction information to further optimize the track so as to enable the track to be smoother and more accurate.
S2, based on the motion characteristics and the current track, identifying the behavior mode of the corresponding user of the mobile terminal through a preset behavior mode library to obtain the current behavior mode.
Specifically, step S2 includes the steps of:
And S21, extracting a plurality of characteristic vectors related to the behavior patterns from the accelerometer data characteristics, the gyroscope data characteristics and the GPS data characteristics, wherein the characteristic vectors comprise acceleration, angular velocity, speed, direction and speed change rate.
In the present embodiment, a plurality of behavior pattern-related feature vectors are extracted from the accelerometer data features, the gyroscope data features, and the GPS data features.
Where accelerometer data provides linear motion characteristics, gyroscope data provides rotational motion characteristics, and GPS data provides position and velocity characteristics, these feature vectors include acceleration, angular velocity, direction, rate of change of velocity, and the like.
For example, the acceleration may be obtained by calculating the square root of the sum of squares of the acceleration values in the three axes, the angular velocity may be obtained by calculating the square root of the sum of the squares of the angular velocity values in the three axes, the velocity may be obtained directly from the GPS data, the direction may be calculated from the gyroscope data, and the rate of change of velocity may be obtained by calculating the difference in velocity between two consecutive points in time divided by the time interval.
And S22, matching the feature vector with a plurality of behavior patterns in a preset behavior pattern library, wherein the behavior pattern library comprises a plurality of known behavior patterns and corresponding feature vectors thereof.
In this embodiment, the extracted feature vector is matched with a plurality of behavior patterns in a preset behavior pattern library. The behavior pattern library comprises a plurality of known behavior patterns and corresponding feature vectors thereof. These known patterns of behavior include, but are not limited to, common patterns of behavior such as walking, running, riding, resting, and the like. The matching process involves comparing the extracted feature vector with the feature vector of each known behavior pattern in the library to find the most similar known behavior pattern. Specifically, the matching may be performed by calculating euclidean distances or cosine similarities between feature vectors.
S23, calculating the similarity between the feature vector and the feature vector of each known behavior mode, and selecting the known behavior mode with the highest similarity as the current behavior mode.
In this embodiment, the similarity between the feature vector and the feature vector of each known behavior pattern is calculated, and the known behavior pattern with the highest similarity is selected as the current behavior pattern.
Illustratively, the Euclidean distance between the extracted feature vector and the feature vector of each known behavior pattern in the library is calculated, and the known behavior pattern with the smallest distance is selected as the current behavior pattern. This ensures that the identified behavior pattern is the pattern that best matches the actual behavior of the current user.
And S3, inputting the motion characteristics and the sound characteristics into a pre-trained prediction model, and judging whether a moving object is approaching to the mobile terminal or not by utilizing the prediction model.
Specifically, step S3 includes the following steps:
S31, fusing the extracted motion features and the extracted sound features to form fused feature vectors.
In this embodiment, first, the motion features and the sound features extracted from the various sensor data need to be fused to form a comprehensive feature vector. Specifically:
motion characteristics including acceleration, angular velocity, speed, direction, and rate of change of speed, etc., are extracted from accelerometer data, gyroscope data, and GPS data.
Sound features, including ambient noise levels, are extracted from sound data collected by a microphone.
The specific steps of the fusion treatment include:
and data preprocessing, namely normalizing the extracted motion characteristics and sound characteristics, ensuring that the characteristics are on the same order of magnitude, and facilitating the subsequent model input.
Feature stitching, namely stitching the normalized motion features and the sound features into a high-dimensional feature vector. Illustratively, if the dimension of the motion feature vector is five (acceleration, angular velocity, direction, rate of change of velocity), the dimension of the sound feature vector is one (ambient noise level), then the dimension of the final fusion feature vector is six.
S32, inputting the fusion feature vector into a pre-trained prediction model, wherein the prediction model is a neural network model which is identified through training and analyzed to judge whether a mobile object is approaching the mobile terminal.
In this embodiment, it is first ensured that the neural network model used has been sufficiently trained, and fusion feature vectors can be effectively identified and analyzed. The model is trained on a large number of labeled datasets that contain the motion and sound characteristics of the mobile terminal under various circumstances, and tag information about whether the mobile object is in close proximity.
The fused feature vector formed in step S31 is passed as input to a pre-trained neural network model. The fusion feature vector contains comprehensive information extracted from various sensors such as an accelerometer, a gyroscope, a GPS, a microphone and the like, and can comprehensively reflect the current state and the surrounding environment of the mobile terminal.
After the neural network model receives the fusion feature vector, the neural network model gradually processes the input data through a series of forward propagation operations. This process includes:
Feature transformation, namely performing nonlinear transformation on input features through multiple layers of neurons, and capturing complex relations in data.
Feature extraction-hidden layers inside the model are responsible for extracting higher level abstract features from the input features that help to distinguish between different moving object proximity situations.
And finally, generating a numerical value between 0 and 1 by the model through a sigmoid activation function of the output layer, and representing the probability of approaching a moving object. A threshold (e.g., 0.5) is typically set above which a moving object is considered to be approaching.
And outputting a result, namely analyzing the result generated by the model into a judgment of whether the moving object approaches. If the judgment result is that the moving object is close, the subsequent distance threshold dynamic adjustment step is triggered.
Specifically, the training process of the prediction model includes:
and collecting a labeling data set containing motion characteristics, sound characteristics and corresponding approaching conditions of the moving object.
In the beginning of model training, a set of annotation data containing rich information needs to be collected first. This data set should cover the motion characteristics, sound characteristics, and corresponding moving object proximity labels in a variety of environments.
The motion characteristics are collected through an accelerometer, a gyroscope and a GPS, and comprise acceleration, angular velocity, speed, direction, speed change rate and the like.
Sound features-sound features such as ambient noise level are collected by a microphone.
Labeling, in which each data point requires a distinct label to indicate whether a moving object is in close proximity.
After the data is collected, preprocessing is required, including data cleaning, missing value processing, outlier detection, etc., to ensure the accuracy and consistency of the data. In addition, for motion and sound features, normalization processing is also needed to make them in the same order of magnitude, so that subsequent model input is facilitated.
Training the neural network model by using the data set, and adjusting model parameters in the training process to minimize the prediction error rate.
In this embodiment, a neural network model is constructed to identify and analyze the fused feature vectors. This model is a multi-layer perceptron model comprising:
And the input layer is used for receiving the fusion feature vector as input.
And the hidden layer comprises a plurality of layers of neurons and is used for carrying out nonlinear transformation and feature extraction on input features.
And an output layer, wherein a value between 0 and 1 is generated by using a sigmoid activation function, and the value represents the probability of approaching a moving object.
After the model is built, the model is trained using the collected annotation dataset. During training, model parameters are continually adjusted to minimize the prediction error rate, which is typically achieved by a back-propagation algorithm.
And evaluating the performance of the neural network model by a cross-validation method, and finally storing the trained model as a pre-trained prediction model.
In this embodiment, in order to evaluate the performance of the model, a cross-validation method is employed. Specifically, the data set is divided into k equal parts (e.g., k=5), each time using the k-1 part as the training set and the remaining part as the test set. Thus, k evaluation results can be obtained, and an average value is taken as a final performance evaluation index.
During the cross-validation process, metrics including, but not limited to, accuracy, recall, F1 score, etc., are used to evaluate the performance of the model. These metrics can help to understand the behavior of the model under different conditions and provide basis for adjusting model parameters.
After sufficient training and evaluation, the trained model is saved as a pre-trained predictive model. This model can be loaded and used in the actual application to determine if there is a moving object approaching the mobile terminal.
And S33, outputting a judging result, and triggering a subsequent distance threshold dynamic adjustment step if the judging result shows that the moving object is close.
In this embodiment, the prediction model outputs a classification result, which indicates whether a mobile object is approaching the mobile terminal, and if the determination result indicates that the mobile object is approaching, step S4 is performed, and the preset distance threshold is dynamically adjusted based on the current motion characteristics, the sound characteristics and the current track, so as to ensure the safety of the user.
And S4, when the mobile object is judged to be approaching the mobile terminal, dynamically adjusting a preset distance threshold based on the motion characteristic, the sound characteristic and the current track at the current moment, wherein the distance threshold represents the safety distance which should be kept between the mobile terminal and the mobile object in the current behavior mode.
Specifically, step S4 includes the steps of:
S41, adjusting a preset basic distance threshold to obtain a preliminary distance threshold, wherein the adjustment adopts the following formula:
Wherein T p is a preliminary distance threshold, T 0 is a base distance threshold, F v is a speed adjustment factor, F d is a direction adjustment factor, and F n is a noise adjustment factor.
In this embodiment, the basic distance threshold is a preset initial value, which indicates a safe distance that should be maintained between the mobile terminal and the mobile object in general. This value is typically determined based on empirical and experimental data, and in order to accommodate different environments and behavioral patterns, a speed adjustment factor, a direction adjustment factor, and a noise adjustment factor are introduced, which are calculated based on current motion characteristics and sound characteristics.
The speed adjustment factor reflects the current speed and the impact of the speed change on the safe distance. The faster the child watch wearer, the greater the safety distance required, since both the reaction time and stopping distance increase when moving at high speed, requiring greater buffer space to avoid collisions. Further, the rate of change of the speed reflects the trend of the change of the speed. If the rate of speed change is large, meaning that the speed change is frequent or severe, this increases uncertainty, requiring a greater safety distance to cope with a possible emergency.
The direction adjustment factor reflects the current direction and the effect of the direction change on the safe distance. The direction of the child watch wearer can affect his or her trajectory. Illustratively, the safe distance required for straight travel is different from that required for cornering. The straight-going path is stable, the safety distance can be properly reduced, and the path change is large when turning, and a larger safety distance is needed. Furthermore, the direction change rate reflects the change frequency and amplitude of the direction. If the direction change rate is large, meaning that children change directions frequently, the uncertainty and complexity of the path are increased, and a larger safety distance is required to ensure safety.
The noise adjustment factor reflects the impact of the current ambient noise on the safe distance. High ambient noise levels can interfere with the child's attention and perception, increasing the risk of accidents. Thus, a greater safety distance is required to compensate for this interference in a high noise environment. In addition, the noise change rate reflects the trend of the change in the environmental noise. If the rate of change of noise is large, meaning that the ambient noise is unstable, uncertainty and potential risk are increased, a greater safety distance is required to cope with a possible emergency.
By combining the speed adjustment factor, the direction adjustment factor, and the noise adjustment factor, the preliminary distance threshold can more fully reflect the current environment and the needs of child behavior for a safe distance. Specifically, the speed adjustment factor ensures that the safe distance is large enough to prevent collisions due to insufficient reaction time at high speeds or frequent acceleration and deceleration. The direction adjustment factor ensures that in case of frequent changes of direction or turns, the safety distance is large enough to prevent accidents due to path changes. The noise adjustment factor ensures that in environments with high noise or frequent noise changes, the safety distance is large enough to prevent hazards due to distraction.
S42, calculating a behavior pattern adjustment factor, and adjusting the preliminary distance threshold based on the behavior pattern adjustment factor to obtain a final dynamic threshold and using the final dynamic threshold as an adjusted distance threshold, wherein the adjustment of the preliminary distance threshold adopts the following formula:
Wherein T f is the final dynamic threshold, F b is the behavior pattern adjustment factor, v is the speed of the mobile terminal at the current time, and v max is the maximum speed in the current track of the mobile terminal.
In the remote monitoring method provided by the invention, the behavior mode influence factor can have an important influence on the final dynamic threshold value, because the behavior mode of the user directly influences the safety requirement and the risk degree of the user in a specific environment. Different behavior patterns mean that the user may be in different active states and environmental conditions, and these changes in state and conditions require the security system to be able to flexibly adjust the security distance threshold to provide more effective protection, thus requiring further adjustment of the preliminary distance threshold by the behavior pattern adjustment factor.
In the adjustment of the basic distance threshold, the calculation formula of the speed adjustment factor is as follows:
Wherein F v is a speed adjustment factor, v is the speed of the mobile terminal at the current moment, deltav is the speed change rate in the current track, and alpha is the influence coefficient of the preset speed change rate.
In this embodiment, the value 10 is used to normalize the terminal speed so that it affects the threshold value within a reasonable range.
The calculation formula of the direction adjustment factor is as follows:
wherein F d is a direction adjustment factor, d is the direction of the mobile terminal at the current moment, Δd is the direction change rate in the current track, and β is the influence coefficient of the preset direction change rate.
In this embodiment, the value 90 is also used to normalize the terminal direction so that it affects the threshold value within a reasonable range.
The calculation formula of the noise adjustment factor is as follows:
Wherein F n is a noise adjustment factor, n is an environmental noise level of the mobile terminal at the current moment, deltan is a noise change rate in the current track, and gamma is an influence coefficient of a preset noise change rate.
In this embodiment, the value 100 is also used to normalize the ambient noise level so that it affects the threshold value within a reasonable range.
Specifically, the calculation formula of the behavior pattern adjustment factor is as follows:
Fb＝Kb+δ×Δb
Wherein F b is a behavior pattern adjustment factor, Δb is a behavior pattern change rate, and represents a ratio of the number of times the user behavior pattern changes in the current track to the total time, δ is an influence coefficient of a preset behavior pattern change rate, and K b is a behavior pattern adjustment coefficient.
In this embodiment, the behavior pattern adjustment coefficients are adjustment coefficients under different behavior patterns, and corresponding adjustment coefficients can be found through a preset mapping table, where the mapping table defines adjustment coefficients corresponding to different behavior patterns, so that the adjustment coefficients are used when calculating the behavior pattern adjustment coefficients, and the construction process of the mapping table includes:
Defining behavior patterns-all possible behavior patterns are defined first, such as stationary, walking, jogging, sprinting, running, bicycling, driving, etc.
Setting an adjustment coefficient, namely setting a reasonable adjustment coefficient for each behavior mode. These coefficients may be determined based on practical testing and experience to ensure that the safe distance threshold can be reasonably adjusted in different behavioral modes.
And S5, judging whether the relative distance between the mobile object and the mobile terminal is smaller than a dynamically adjusted distance threshold value, and generating alarm information and pushing the alarm information to an associated terminal pre-paired with the mobile terminal when the relative distance is smaller than the dynamically adjusted distance threshold value.
In this embodiment, the first obtaining the relative distance specifically includes:
And acquiring the position information of the mobile terminal, namely acquiring the current position information of the mobile terminal through GPS data, wherein the current position information comprises longitude, latitude and altitude.
Estimating the position information of the moving object by combining the sound data acquired by the sound sensor, the accelerometer, the gyroscope and the GPS data. The specific method comprises the following steps:
And (3) sound data processing, namely calculating the average sound pressure level of the sound signal and evaluating the environmental noise level.
And (3) motion data processing, namely estimating the motion trail and direction of the mobile terminal by utilizing accelerometer and gyroscope data.
Position estimation the position of a moving object is estimated using a sound source localization algorithm, in particular a method based on sound intensity and time difference of arrival, in combination with sound data and motion data.
And calculating a relative distance according to the estimated position information of the mobile terminal and the mobile object, wherein the relative distance is calculated through Euclidean distance formula.
The relative distance is then compared to a dynamic threshold, specifically comprising:
and (3) acquiring a dynamic threshold, namely acquiring a final dynamic threshold after dynamic adjustment from the step S4.
And comparing the calculated relative distance with the final dynamic threshold value, and judging whether the calculated relative distance is smaller than the dynamic threshold value or not.
Then, generating alarm information specifically including:
If the relative distance is less than the final dynamic threshold, the mobile object is considered approaching the mobile terminal, a potential security risk exists, and alert information is generated, including but not limited to alert time, location information of the mobile object, location information of the mobile terminal, relative distance, dynamic threshold, and current behavior pattern of the user.
Finally, pushing alarm information, which specifically comprises:
And determining the associated terminal pre-paired with the mobile terminal. The associated terminal can be a guardian mobile phone, a smart watch and other devices of the user.
Pushing the alarm information to the associated terminal through wireless communication technology (including but not limited to Bluetooth, wi-Fi, cellular network and the like).
And displaying alarm information, namely displaying the alarm information on the associated terminal to remind a user of paying attention to safety. The display may be, but is not limited to, a pop-up notification, a vibration alert, an audible alarm, etc.
Example two
The present invention also provides a child watch, as shown with reference to fig. 2, comprising a sensor assembly 100 and a processor 200.
The sensor assembly 100 is used for a variety of sensor data.
The processor 200 is configured to extract motion features and sound features of the mobile terminal from a variety of sensor data and generate a current trajectory based on the motion features.
The processor 200 is configured to identify, based on the motion feature and the current track, a behavior pattern of a corresponding user of the mobile terminal through a preset behavior pattern library, so as to obtain a current behavior pattern.
The processor 200 is configured to input the motion characteristics and the sound characteristics into a pre-trained prediction model, and determine whether a moving object is approaching the mobile terminal using the prediction model.
The processor 200 is configured to dynamically adjust a preset distance threshold based on the motion feature, the sound feature and the current track at the current moment when the mobile object is determined to be approaching the mobile terminal, where the distance threshold characterizes a safe distance between the mobile terminal and the mobile object that should be maintained in the current behavior mode.
The processor 200 is configured to determine whether a relative distance between the mobile object and the mobile terminal is smaller than a dynamically adjusted distance threshold, and generate alarm information and push the alarm information to an associated terminal pre-paired with the mobile terminal when the relative distance is smaller than the dynamically adjusted distance threshold.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flowchart and/or block of the flowchart illustrations and/or block diagrams, and combinations of flowcharts and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Those of ordinary skill in the art will appreciate that all or part of the steps of the various methods of the above embodiments may be implemented by hardware associated with a program that may be stored in a computer-readable storage medium, including Read-Only Memory (ROM), random-access Memory (Random Access Memory, RAM), programmable Read-Only Memory (Programmable Read-Only Memory, PROM), erasable programmable Read-Only Memory (Erasable Programmabl e Read Only Memory, EPROM), one-time programmable Read-Only Memory (OTPROM), electrically erasable programmable Read-Only Memory (EEPROM), compact disc Read-Only Memory (CompactDis c Read-Only Memory, CD-ROM), or other optical disk Memory, magnetic disk Memory, tape Memory, or any other medium that can be used to carry or store data that is readable by a computer.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.
Claims (10)
1. A remote monitoring method, characterized in that the monitoring method comprises the following steps:
extracting motion features and sound features of a mobile terminal from a plurality of sensor data, wherein the plurality of sensor data are collected by a sensor assembly mounted on the mobile terminal, and generating a current track based on the motion features;
based on the motion characteristics and the current track, identifying the behavior mode of the corresponding user of the mobile terminal through a preset behavior mode library to obtain a current behavior mode;
Inputting the motion characteristics and the sound characteristics into a pre-trained prediction model, and judging whether a mobile object is approaching to the mobile terminal or not by utilizing the prediction model;
When the mobile object is judged to be approaching the mobile terminal, dynamically adjusting a preset distance threshold value based on the motion characteristic, the sound characteristic and the current track at the current moment, wherein the distance threshold value represents the safety distance which should be kept between the mobile terminal and the mobile object in the current behavior mode;
And judging whether the relative distance between the mobile object and the mobile terminal is smaller than a dynamically adjusted distance threshold value, and generating alarm information and pushing the alarm information to an associated terminal pre-paired with the mobile terminal when the relative distance is smaller than the dynamically adjusted distance threshold value.
2. The method of claim 1, wherein the plurality of sensor data includes accelerometer data, gyroscope data, GPS data, and sound data, wherein the accelerometer data, the gyroscope data, and the GPS data are used to extract the motion characteristics, and the sound data is used to extract the sound characteristics, wherein the motion characteristics include accelerometer data characteristics, gyroscope data characteristics, and GPS data characteristics corresponding to the accelerometer data, the gyroscope data, and the GPS data, respectively.
3. The remote monitoring method according to claim 2, wherein the generating step of the current track is:
And generating the current track of the mobile terminal by using a map matching algorithm based on the extracted accelerometer data characteristics and gyroscope data characteristics and combining the position change information and the timestamp data in the GPS data characteristics.
4. The remote monitoring method according to claim 3, wherein the identifying, based on the motion feature and the current track, the behavior pattern of the corresponding user of the mobile terminal through a preset behavior pattern library, to obtain the current behavior pattern includes:
Extracting a plurality of behavior pattern related feature vectors from the accelerometer data features, gyroscope data features and GPS data features, wherein the feature vectors comprise acceleration, angular velocity, speed, direction and speed change rate;
matching the feature vector with a plurality of behavior patterns in a preset behavior pattern library, wherein the behavior pattern library comprises a plurality of known behavior patterns and corresponding feature vectors thereof;
And calculating the similarity between the feature vector and the feature vector of each known behavior mode, and selecting the known behavior mode with the highest similarity as the current behavior mode.
5. The method of claim 4, wherein said inputting the motion feature and the sound feature into a pre-trained predictive model and using the predictive model to determine whether a mobile object is approaching the mobile terminal comprises:
Performing fusion processing on the extracted motion characteristics and the extracted sound characteristics to form fusion characteristic vectors;
inputting the fusion feature vector into a pre-trained prediction model, wherein the prediction model is a neural network model which is identified through training and analyzes the fusion feature vector to judge whether a moving object is approaching a mobile terminal;
outputting a judging result, and triggering a subsequent distance threshold dynamic adjustment step if the judging result is that the moving object is close.
6. The method of claim 5, wherein the training process of the predictive model comprises:
collecting a labeling data set containing motion characteristics, sound characteristics and corresponding moving object approaching conditions;
Training a neural network model by utilizing the data set, and adjusting model parameters in the training process to minimize the prediction error rate;
and evaluating the performance of the neural network model by a cross-validation method, and finally storing the trained model as a pre-trained prediction model.
7. The remote monitoring method according to claim 6, wherein dynamically adjusting a preset distance threshold based on the motion characteristic, the sound characteristic, and the current trajectory at the current time when the mobile object is determined to be approaching the mobile terminal comprises:
adjusting a preset basic distance threshold to obtain a preliminary distance threshold, wherein the adjustment adopts the following formula:
Wherein, T p is a preliminary distance threshold, T 0 is a basic distance threshold, F v is a speed adjustment factor, F d is a direction adjustment factor, and F n is a noise adjustment factor;
Calculating a behavior pattern adjustment factor, adjusting the preliminary distance threshold based on the behavior pattern adjustment factor to obtain a final dynamic threshold, and taking the final dynamic threshold as an adjusted distance threshold, wherein the adjustment of the preliminary distance threshold adopts the following formula:
Wherein T f is the final dynamic threshold, F b is the behavior pattern adjustment factor, v is the speed of the mobile terminal at the current time, and v max is the maximum speed in the current track of the mobile terminal.
8. The method according to claim 7, wherein in the adjustment of the base distance threshold, the speed adjustment factor is calculated by the formula:
Wherein F v is a speed adjustment factor, v is the speed of the mobile terminal at the current moment, deltav is the speed change rate in the current track, and alpha is the influence coefficient of the preset speed change rate;
The calculation formula of the direction adjustment factor is as follows:
Wherein F d is a direction adjustment factor, d is the direction of the mobile terminal at the current moment, Δd is the direction change rate in the current track, and β is the influence coefficient of the preset direction change rate;
the calculation formula of the noise adjustment factor is as follows:
Wherein F n is a noise adjustment factor, n is an environmental noise level of the mobile terminal at the current moment, deltan is a noise change rate in the current track, and gamma is an influence coefficient of a preset noise change rate.
9. The remote monitoring method according to claim 8, wherein the calculation formula of the behavior pattern adjustment factor is:
Fb＝Kb+δ×Δb
Wherein F b is a behavior pattern adjustment factor, Δb is a behavior pattern change rate, and represents a ratio of the number of times the user behavior pattern changes in the current track to the total time, δ is an influence coefficient of a preset behavior pattern change rate, and K b is a behavior pattern adjustment coefficient.
10. A child watch is characterized by comprising a sensor assembly and a processor,
The sensor assembly is for a variety of sensor data;
The processor is used for extracting the motion characteristics and sound characteristics of the mobile terminal from various sensor data and generating a current track based on the motion characteristics;
The processor is used for identifying the behavior mode of the corresponding user of the mobile terminal through a preset behavior mode library based on the motion characteristics and the current track to obtain a current behavior mode;
the processor is used for inputting the motion characteristics and the sound characteristics into a pre-trained prediction model, and judging whether a moving object is approaching to the mobile terminal or not by utilizing the prediction model;
The processor is used for dynamically adjusting a preset distance threshold value based on the motion characteristic, the sound characteristic and the current track at the current moment when the mobile object is judged to be approaching the mobile terminal, wherein the distance threshold value represents the safety distance which should be kept between the mobile terminal and the mobile object in the current behavior mode;
The processor is used for judging whether the relative distance between the mobile object and the mobile terminal is smaller than a distance threshold after dynamic adjustment, and generating alarm information and pushing the alarm information to an associated terminal pre-paired with the mobile terminal when the relative distance is smaller than the distance threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411841247.6A CN119694070B (en) | 2024-12-13 | 2024-12-13 | Remote monitoring methods and children's watches |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411841247.6A CN119694070B (en) | 2024-12-13 | 2024-12-13 | Remote monitoring methods and children's watches |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119694070A CN119694070A (en) | 2025-03-25 |
| CN119694070B true CN119694070B (en) | 2025-07-18 |
Family
ID=95036322
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411841247.6A Active CN119694070B (en) | 2024-12-13 | 2024-12-13 | Remote monitoring methods and children's watches |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119694070B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104994218A (en) * | 2015-05-14 | 2015-10-21 | 努比亚技术有限公司 | Mobile terminal and method thereof for achieving use prompt |
| CN105049589A (en) * | 2015-05-20 | 2015-11-11 | 李涛 | Handset device used for improving personal safety and early warning method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008108788A2 (en) * | 2006-05-31 | 2008-09-12 | Trx Systems, Inc. | Method and system for locating and monitoring first responders |
| CN105197134A (en) * | 2015-08-20 | 2015-12-30 | 深圳市比奇诺科技有限公司 | Bicycle riding safety monitoring device, as well as bicycle and riding safety monitoring method using bicycle riding safety monitoring device |
| US10867501B2 (en) * | 2017-06-09 | 2020-12-15 | Ibiquity Digital Corporation | Acoustic sensing and alerting |
| CN113593175B (en) * | 2021-06-15 | 2023-06-16 | 深圳市广和通无线股份有限公司 | Remote monitoring method, child watch, terminal device and storage medium |
| CN118736526A (en) * | 2024-06-04 | 2024-10-01 | 重庆工业职业技术学院 | Vehicle collision warning method, device, equipment and medium based on image recognition |
| CN119037415B (en) * | 2024-10-30 | 2025-01-28 | 深圳市平方科技股份有限公司 | Multimodal large model driving risk judgment method, system, medium and program product |
-
2024
- 2024-12-13 CN CN202411841247.6A patent/CN119694070B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104994218A (en) * | 2015-05-14 | 2015-10-21 | 努比亚技术有限公司 | Mobile terminal and method thereof for achieving use prompt |
| CN105049589A (en) * | 2015-05-20 | 2015-11-11 | 李涛 | Handset device used for improving personal safety and early warning method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119694070A (en) | 2025-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kumar et al. | An IoT-based vehicle accident detection and classification system using sensor fusion | |
| KR102306064B1 (en) | Activity recognition using accelerometer data | |
| EP4211601B1 (en) | Method for electronic trip familiarity detection | |
| EP2769574B1 (en) | Tracking activity, velocity, and heading using sensors in mobile devices or other systems | |
| US10504031B2 (en) | Method and apparatus for determining probabilistic context awareness of a mobile device user using a single sensor and/or multi-sensor data fusion | |
| Reddy et al. | Determining transportation mode on mobile phones | |
| US10614693B2 (en) | Dangerous situation notification apparatus and method | |
| KR20120043845A (en) | User equipment and method for cogniting user state thereof | |
| Jiang et al. | A multi-scale attributes attention model for transport mode identification | |
| US10104494B2 (en) | Marker based activity transition models | |
| US20140309964A1 (en) | Internal Sensor Based Personalized Pedestrian Location | |
| US20190200169A1 (en) | System and method to localise an entity | |
| Nirmal et al. | Transportation mode detection using crowdsourced smartphone data | |
| Genoud et al. | Soft fall detection using machine learning in wearable devices | |
| EP3382570A1 (en) | Method for characterizing driving events of a vehicle based on an accelerometer sensor | |
| CN119694070B (en) | Remote monitoring methods and children's watches | |
| Yu et al. | Deep learning-based vibration signal personnel positioning system | |
| CN107203259B (en) | Method and apparatus for determining probabilistic content awareness for mobile device users using single and/or multi-sensor data fusion | |
| Shafique et al. | Improving the Accuracy of Travel Mode Detection for Low Data Collection Frequencies | |
| de Cossío et al. | A Robust Approach for Transportation Mode Detection Using Smartphone-Based GPS Sensors and Road Network Information | |
| Shi et al. | On-Device Multi-Sensor Fusion with Machine Learning for Continuous Transit Navigation | |
| CN120599532B (en) | Cross-region AI track tracking method based on multi-information fusion | |
| Atashi | Multiple Model-based Indoor Localization via Bluetooth Low Energy and Inertial Measurement Unit Sensors | |
| CN120831122B (en) | A method for dynamic adjustment of ambulance positioning based on environmental perception | |
| EP4629239A1 (en) | Systems and methods for contextual scene analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |