CN108307049A - Electronic device falls model update method and Related product - Google Patents

Abstract

The application provides a method for updating a drop model of an electronic device and related products. The method includes: collecting acceleration data of the electronic device; collecting the distance between the electronic device and the ground; The acceleration value and distance value in the falling state are composed of the acceleration value and distance value in the falling state as the first input data, and the first input data is input into the first drop model for calculation to obtain the first output result, and the first input data is used as The training data retrains the first drop model to obtain the second drop model, and the first input data is input to the second drop model to obtain the second output result, and the first output result is compared with the second output result, such as the second output result greater than the first output result, the first drop model is replaced by the second drop model. The technical solution provided by this application has the advantage of high user experience.

Description

Method for updating drop model of electronic device and related products

technical field

The present application relates to the technical field of terminal equipment, in particular to a method for updating a drop model of an electronic device and related products.

Background technique

In the prior art, mobile terminals (such as mobile phones, tablet computers, etc.) have become the first choice and most frequently used electronic devices for users. For mobile terminals, the fragile screen is an unavoidable problem for manufacturers or users. After the screen is broken, the terminal's The residual value is greatly reduced, because the price of repairing and replacing the screen by most manufacturers has almost exceeded the residual value of the terminal. Moreover, 2.5D glass is currently popular in the industry as a screen, which is more likely to be broken and broken. All mainstream manufacturers are spending a lot of research and development costs to improve the drop resistance of the whole machine.

There are many ways to calculate the existing drop data, such as calculation through the drop model, but the parameters of the existing drop model will not change after the factory settings, but the parameters of the electronic device change with time, so now The calculation results of some drop models are not accurate, which affects the customer experience.

application content

Embodiments of the present application provide a method for updating a drop model of an electronic device and related products, which can restore a dropped scene, allow users to watch intuitively, and improve user experience.

In the first aspect, an embodiment of the present application provides an electronic device, the electronic device includes: an application processor AP, a touch display screen, a gravity sensor and a distance sensor, wherein,

The gravity sensor is used to collect acceleration data of the electronic device;

The distance sensor is used to collect the distance between the electronic device and the ground;

The AP is used to calculate an acceleration value based on the acceleration data, and determine the state of the electronic device according to the acceleration value, and the state includes: a normal state and a falling state;

The AP is used to extract the acceleration value and the distance value in the falling state, compose the acceleration value and the distance value in the falling state into the first input data, and input the first input data into the first drop model for calculation to obtain the first Output the result, use the first input data as training data to retrain the first drop model to obtain the second drop model, input the first input data to the second drop model to obtain the second output result, and combine the first output result with the second drop model Comparing the two output results, if the second output result is greater than the first output result, the first drop model is replaced by the second drop model.

In a second aspect, a method for calculating drop data based on artificial intelligence is provided, the method is applied in an electronic device, and the electronic device includes: an application processor AP, a touch display screen, a gravity sensor and a distance sensor, the method include:

collecting acceleration data of the electronic device;

Collect the distance between the electronic device and the ground;

The acceleration value is calculated according to the acceleration data, and the state of the electronic device is determined according to the acceleration value, and the state includes: a normal state and a falling state;

Extracting the acceleration value and distance value in the falling state, forming the first input data with the acceleration value and the distance value in the falling state, inputting the first input data into the first drop model for calculation to obtain the first output result, and using the first input data One input data is used as training data to retrain the first drop model to obtain the second drop model, and the first input data is input to the second drop model to obtain the second output result, and the first output result is compared with the second output result, such as The second output result is greater than the first output result, and the first drop model is replaced by the second drop model.

In a third aspect, an electronic device is provided, and the electronic device includes: a processing unit, a touch display screen, a gravity sensor, a circuit, and a distance sensor,

The gravity sensor is used to collect acceleration data of the electronic device;

The distance sensor is used to collect the distance between the electronic device and the ground;

The processing unit is used to calculate an acceleration value according to the acceleration data, and determine the state of the electronic device according to the acceleration value, and the state includes: a normal state and a falling state;

The processing unit is configured to extract the acceleration value and the distance value in the falling state, compose the acceleration value and the distance value in the falling state into first input data, and input the first input data into the first drop model for calculation to obtain the first input data. One output result, the first input data is used as the training data to retrain the first drop model to obtain the second drop model, the first input data is input to the second drop model to obtain the second output result, and the first output result is combined with Comparing the second output result, if the second output result is greater than the first output result, the first drop model is replaced by the second drop model.

In a fourth aspect, a computer-readable storage medium is provided, which stores a computer program for electronic data exchange, wherein the computer program causes a computer to execute the method provided in the second aspect.

In a fifth aspect, a computer program product is provided, the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute the method provided in the second aspect.

Implementing the embodiment of the present application has the following beneficial effects:

It can be seen that after the technical solution provided by the application collects acceleration data, the acceleration value is calculated according to the acceleration data, and the distance between the electronic device and the ground is collected. When it is determined to be in a falling state, the acceleration value and distance value of the falling state are extracted, and the The acceleration value and the distance value form the first input data, and input the first input data into the first drop model for calculation to obtain the first output result, and then use the first input data as training data to retrain the first drop model to obtain The second drop model, the first input data is input to the second drop model to obtain the second output result, if the second output result is greater than the first output result, the second drop model is used to replace the first drop model, so that it can be based on the actual The detection data updates the parameters of the existing drop model, so as to be more adaptable to the current state of the electronic device, improve calculation accuracy, and improve user experience.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings that need to be used in the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present application. For Those of ordinary skill in the art can also obtain other drawings based on these drawings without making creative efforts.

FIG. 1 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

Fig. 1a is a schematic diagram of a parallel plate capacitor provided by an embodiment of the present application.

Fig. 1b is a schematic diagram of another parallel plate capacitor provided by an embodiment of the present application.

Fig. 1c is a schematic diagram of another parallel plate capacitor provided by the embodiment of the present application.

Fig. 1d is a schematic diagram of the acceleration provided by the embodiment of the present application.

Fig. 2 is a schematic diagram of an electronic device disclosed in an embodiment of the present application.

Fig. 3a is a schematic diagram of a convolution disclosed in an embodiment of the present application.

Fig. 3b is a schematic diagram of movement of input data according to an embodiment of the present application.

Fig. 3c is a schematic diagram of input data according to the embodiment of the present application.

Fig. 3d is a schematic diagram of inserting input data according to the embodiment of the present application.

FIG. 4 is a schematic flowchart of a method for updating a drop model of an electronic device provided in an embodiment of the present application.

FIG. 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

FIG. 6 is a schematic structural diagram of a mobile phone disclosed in an embodiment of the present application.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.

The terms "first", "second", "third" and "fourth" in the specification and claims of the present application and the drawings are used to distinguish different objects, rather than to describe a specific order . Furthermore, the terms "include" and "have", as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes unlisted steps or units, or optionally further includes For other steps or units inherent in these processes, methods, products or apparatuses.

Reference herein to an "embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The occurrences of this phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is understood explicitly and implicitly by those skilled in the art that the embodiments described herein can be combined with other embodiments.

The electronic devices in this application may include smart phones (such as Android mobile phones, iOS mobile phones, WindowsPhone mobile phones, etc.), tablet computers, palmtop computers, notebook computers, mobile Internet devices (MID, Mobile InternetDevices) or wearable devices, etc., the above-mentioned electronic devices It is only an example, not exhaustive, including but not limited to the above-mentioned electronic device. For the convenience of description, the above-mentioned electronic device is referred to as user equipment (User equipment, UE) in the following embodiments. Of course, in practical applications, the above-mentioned user equipment is not limited to the above-mentioned realization forms, for example, it may also include: a smart vehicle terminal, a computer device, and the like.

In the electronic device provided in the first aspect, the electronic device further includes: a communication module;

The AP is further configured to control the communication module to send the weight data of the second drop model to the network side device.

In the electronic device provided in the first aspect, the AP is specifically used for traversing the n acceleration values in the order of the collection points if there are n acceleration values, and if m consecutive acceleration values are greater than the set threshold, it is determined to be in a falling state, Otherwise, it is determined as a normal state.

In the electronic device provided in the first aspect, the AP is specifically used to input the first input data as the training data into the first drop model to perform the V-layer forward operation to obtain the V-th layer forward operation result of the forward operation. The input data gradient of the Vth layer is obtained by processing the result of the forward operation of the Vth layer, and the input data gradient of the Vth layer is input to the Vth layer of the first drop model to perform the reverse operation of the V layer to obtain V weight gradients. The value gradient updates the weight of the V layer to obtain updated weight data, and the updated weight data is determined as the weight data of the second drop model.

In the method provided in the second aspect, the method also includes:

Sending the weight data of the second drop model to the network side device.

In the method provided in the second aspect, if there are n acceleration values, determining the state of the electronic device according to the acceleration values includes:

Traverse n acceleration values in the order of collection points. If m consecutive acceleration values are greater than the set threshold, it is determined to be in a falling state, otherwise it is determined to be in a normal state.

In the method provided in the second aspect, retraining the first drop model to obtain the second drop model by using the first input data as training data includes:

Inputting the first input data as the training data into the first drop model to perform the V-layer forward operation to obtain the V-th layer forward operation result of the forward operation, and processing the V-th layer forward operation result to obtain the V-layer input data gradient, Input the input data gradient of the Vth layer to the Vth layer of the first drop model to perform the reverse operation of the V layer to obtain V weight gradients, and use the V weight gradients to update the weights of the V layer to obtain the updated weight data. The updated weight data is determined as the weight data of the second drop model.

Please refer to FIG. 1. FIG. 1 is an electronic device provided by an embodiment of the present application. Please refer to FIG. 1. FIG. 1 is a schematic structural diagram of an electronic device 100 provided by an embodiment of the present invention. The above-mentioned electronic device 100 includes: a casing 110, a circuit board 120, a battery 130, a cover plate 140, a touch screen 150, a gravity sensor (English: Gravity Sensor, G-sensor for short) 170, a distance sensor 180, and the circuit board is arranged on the housing 110 120, the battery 130 and the cover plate 140, the circuit board 120 is also provided with a circuit connected to the touch display screen 150; the circuit board 120 may also include: an application processor AP190, a gravity sensor 170 and distance sensor 180 .

The above-mentioned touch display screen may specifically be a thin film transistor liquid crystal display (Thin Film Transistor-Liquid Crystal Display, TFT-LCD), a light emitting diode (Light Emitting Diode, LED) display screen, an organic light emitting diode (Organic Light-Emitting Diode, OLED) display screen etc.

The gravity sensor 170 is used to detect the direction and magnitude of the acceleration, which is equivalent to detecting the motion state of the electronic device. The function of the G-sensor is relatively simple to understand, mainly to perceive changes in acceleration force, such as shaking, falling, rising, falling and other movement changes, which can be converted into electrical signals by the G-sensor, and then calculated by the application processor AP190 After analysis, the acceleration value of the electronic device can be determined.

Optionally, the above-mentioned electronic device may further include: a geomagnetic sensor and a gyroscope, and the geomagnetic sensor and the gyroscope are respectively connected to the application processor AP190. On the electronic device, the G-sensor not only works alone, but also cooperates with the geomagnetic sensor 171 and the gyroscope 172 to provide more accurate and comprehensive motion sensing capabilities.

Specifically, in the electronic device, the gravity sensor 170 can actually be a parallel plate capacitor. The capacitance of the parallel plate capacitor is inversely proportional to the distance between the plates. By detecting the capacitance changes in the X, Y, and Z directions, it can Calculate the linear acceleration in each direction.

Taking the acceleration calculation method in the X direction as an example, the specific acceleration value can be:

As shown in Figure 1a, it is a schematic diagram of a parallel plate capacitor.

Referring to Figure 1a, the acceleration corresponding to Figure 1a is 0, as shown in Figure 1a, since there is no acceleration value at this time, the parallel plate in the middle is at the initial position, so the capacitance value C ₁ =C ₀ at this time, this C ₁ can is the capacitance value between the parallel plate and the lower electrode, and the C ₀ may be the initial capacitance value. The capacitance value at this time is C ₂ =C ₀ , and this C ₂ can be the capacitance value between the parallel plate and the upper electrode. At this time, the distance d ₁ =d ₀ corresponding to the capacitance C ₁ ; the distance d ₂ corresponding to the capacitance C ₂ =d ₀ ; wherein, d ₁ may be the distance between the parallel plate and the lower electrode, and d ₂ may be the distance between the parallel plate and the upper electrode. Since the acceleration value at this time is zero, C ₁ =C ₂ =C ₀ ; a _x =0 can be calculated according to the above formula.

Refer to Figure 1b. The acceleration corresponding to Figure 1b is a positive value. Due to the positive acceleration, the parallel plate will move to the lower electrode. Assuming that the moving distance is x, then the distance between the parallel plate and the upper electrode will increase by x , so at this time, d ₁ =d ₀ -x, d ₂ =d ₀ +x; according to the calculation formula of plate capacitance, it is shown in the following formula:

Among them, S can be the corresponding area between the two plates of the parallel plate capacitor, ε is the dielectric constant (it is determined by the material of the plate electrode), k is the electrostatic constant, and d is the distance between the two plates of the parallel plate capacitor the distance.

The capacitance values shown in Figure 1b are as follows:

Therefore, since the parallel plate of the parallel plate capacitor moves to the lower electrode, C ₁ >C ₂ , ie a _x >0.

Refer to Figure 1c. The acceleration corresponding to Figure 1c is negative. Due to the effect of negative acceleration, the parallel plate will move to the upper electrode. Assuming that the moving distance is x, then the distance between the parallel plate and the upper electrode will increase by x. So at this time, d ₁ =d ₀ +x, d ₂ =d ₀ -x; according to the calculation formula of plate capacitance, it is shown in the following formula:

The capacitance values shown in Figure 1c are as follows:

At this time, since the parallel plate of the parallel plate capacitor moves to the upper electrode, C ₁ <C ₂ , that is, a _x <0.

That is, through the test of the above-mentioned parallel plate capacitor, the specific acceleration value can be obtained, and this value can represent the direction of the acceleration.

Specifically, for an electronic device, the value of the acceleration tested has three directions, as shown in Figure 1d, which is a schematic diagram of the three directions tested by the electronic device. Specifically, it can be divided into the X-axis direction and the Y-axis direction. As well as the Z-axis direction, its specific display schematic diagram is shown in FIG. 1d.

Specifically, in an optional drop test, its corresponding acceleration value during the fall can be:

_ax = 0.049m/S ²

a _y =—0.026m/S ²

a _z = 9.800 m/S ² .

According to the above data, it can be determined that the electronic device is in a falling state.

As shown in FIG. 2, it is a schematic structural diagram of an electronic device provided by the present application. As shown in FIG. A sensor 260 and a circuit 240 , a camera 230 is arranged outside the housing, and the camera and the touch screen are connected to the application processor AP through at least one circuit. Wherein, the AP210 is connected to the gravity sensor 250 and the distance sensor through another circuit, wherein the circuit 240 may specifically include: a bus, a flexible circuit board, a connection chip, etc. Of course, in practical applications, the above circuit 240 may also be other The expression form of the above-mentioned circuit 240 is not limited by the specific embodiment of the present application. The above-mentioned electronic device 200 can also include: a geomagnetic sensor and a gyroscope, which can collect data in combination with a gravity sensor 250; the electronic device 200 can also include: an artificial intelligence processor, which can be set separately, It can also be integrated with the application processor AP210. For the convenience of description, in the embodiment shown in FIG. 2, the artificial intelligence processor is integrated in the AP210.

The gravity sensor 250 is used to collect the acceleration data of the electronic device, and transmit the acceleration data to the application processor AP;

The distance sensor 260 is used to collect the distance between the electronic device and the ground;

AP210 is used to calculate the acceleration value based on the acceleration data, and determine the state of the electronic device according to the acceleration value, and the state includes: normal state and falling state;

Optionally, the above acceleration data may be multiple capacitance values of parallel plate capacitors, specifically, may be the values of _C1 and _C2 as shown in Fig. 1a, Fig. 1b, and Fig. 1c. Of course, in practical applications, it is necessary to collect the acceleration data of the X, Y, and Z axes as shown in FIG. 1d. Of course, in practical applications, if other gravity sensors are used, the acceleration data may also be other types of data, and the specific embodiments of the present application do not limit the actual form of expression of the above acceleration data.

The number of the above acceleration values may be n acceleration values. Specifically, the AP 210 traverses n acceleration values in the order of collection points. If m consecutive acceleration values are greater than the set threshold, it is determined as a falling state, otherwise it is determined as a normal state. Wherein, n and m are integers greater than or equal to 2, and m<n.

AP210 is used to extract the acceleration value and distance in the falling state, compose the acceleration value and distance value into the first input data, input the first input data into the first drop model for calculation to obtain the first output result, and use the first input data The input data is used as training data to retrain the first drop model to obtain the second drop model, and the first input data is input to the second drop model to obtain the second output result, and the first output result is compared with the second output result, as in the first step The second output result is greater than the first output result, and the first drop model is replaced by the second drop model.

After the technical solution provided by the application collects the acceleration data, the acceleration value is calculated according to the acceleration data, and the distance between the electronic device and the ground is collected. When the falling state is determined, the acceleration value and the distance value of the falling state are extracted, and the acceleration value and the distance Values form the first input data, input the first input data into the first drop model for calculation to obtain the first output result, and then use the first input data as training data to retrain the first drop model to obtain the second drop model , input the first input data into the second drop model to get the second output result, if the second output result is greater than the first output result, the second drop model is used to replace the first drop model, so that the actual detection data can be used for the current The parameters of some drop models are updated, so as to be more suitable for the current state of the electronic device, improve the calculation accuracy, and improve the user experience.

Optionally, the aforementioned AP210 is further configured to delete the second drop model when it is determined that the second output result is less than or equal to the first output result.

For the technical solution where the second output result is less than or equal to the first output result, that is, the trained second drop model is inferior to the first drop model. At this time, there is no need to update the first drop model, so just delete the second drop model directly. .

The method for comparing the above-mentioned first output result with the second output result may be,

If both the first output result and the second output result are data blocks, the data blocks may specifically include: one of vectors, matrices, three-dimensional data, and four-dimensional data. Extract the maximum element value X _max1 of the first output result, extract the maximum element value X _max2 of the second output result; if X _max2 > X _max1 , determine that the second output result is greater than the first output result, such as X _max2 ≤ X _max1 , determine The second output result is less than or equal to the first output result. Of course, the comparison method can also adopt other methods, and the present application is not limited to the above-mentioned specific method of comparison.

Optionally, the electronic device above further includes: a communication module;

The AP is further configured to control the communication module to send the weight data of the second drop model to the network side device.

Optionally, the AP210 is specifically used to input the first input data as the training data into the first drop model to perform the V-layer forward operation to obtain the V-th layer forward operation result of the forward operation, and to perform the V-th layer forward operation result Process to obtain the input data gradient of the Vth layer, input the input data gradient of the Vth layer to the Vth layer of the first drop model, perform the reverse operation of the V layer to obtain V weight gradients, and use the V weight gradients to calculate the weight value of the V layer The updated weight data is obtained by updating, and the updated weight data is determined as the weight data of the second drop model.

Optionally, AP210, which is specifically used to obtain the input data gradient of the V-th layer by multiplying the result of the forward operation of the V-th layer and the empirical coefficient;

Or perform data type transformation on the forward operation result of the Vth layer to obtain the input data gradient of the Vth layer.

AP210 is specifically used to extract the number n of acceleration values and the number m of distance values, and extract the number of preset input data, namely CI*H*W; where H is the height value, W is the width value, and CI is the depth value, such as If n+m is less than CI*H*W, then the number of acceleration values and the number of pressure values are added according to a preset strategy such that n'+m'=CI*H*W after adding.

Optionally, the AP210 is specifically used to insert a row of insertion data every other row in the H direction when n+m=CI*H*W/2, and the insertion data is an average value of adjacent rows in the H square. Specifically, if the data inserted is the second row in the H direction, the inserted data is the average value of the first row and the third row in the H direction.

The following introduces the principle of the drop model. Most of the drop model belongs to the calculation of artificial intelligence. Most of the calculations of artificial intelligence use the calculation of neural network. For the calculation of neural network, although it has multi-layer calculations, the basic calculation for the convolution operation.

As shown in Figure 3a, it is a schematic diagram of a convolution operation. As shown in Figure 3a, the input data can be three-dimensional data of CI*H*W, and the weight of the convolution operation, that is, the convolution kernel can be CO*CI *3*3 convolution data, the output result can be: the output result of CO*(H-2)*(W-2), as shown in Figure 3a, each square is a value, and the value is specific It can be one of the number n of acceleration values or the number m of pressure values.

According to Figure 3a, introduce the calculation principle of the neural network. For the calculation of the neural network, that is, the artificial intelligence, the trained artificial intelligence model obtains the weight data through the training operation through the preset defined input data, and the trained artificial intelligence The intelligent model is a certain convolution kernel, that is, CO*CI*3*3. For the kernel kernel, it has the specifications (SIZE) of 3*3 and 5*5. Of course, in practical applications, the above weights are still Other specifications may be used, and the present application does not limit the specific form of the above specifications.

For the training operation, that is to say, multiple pieces of input data CI*H*W, in actual training, the values of multiple pieces of input data CI, H, and W can be different, and it executes the multi-layer forward operation of the neural network to obtain the output result , according to the output result to obtain the output result gradient, and then perform multi-layer reverse operation on the output result gradient to obtain the weight gradient of each layer, and then update the weight value of each layer through the weight gradient. After multiple iterations, the calculation is The final weight value is obtained, and the neural network model at this time is the trained neural network model. For this trained neural network model, the output result obtained by inputting the collected input data again and performing forward operation is CO*(H-2)*(W-2), through CO*(H-2)*(W- 2) The corresponding classification can be obtained by analyzing, and when applied to this application, the final cause of the drop can be obtained by analyzing CO*(H-2)*(W-2).

For the convolution operation, the convolution kernel, namely CO*CI*3*3, cannot be directly convolved with the input data, namely CI*H*W, the operation method can be, the convolution kernel is CO *CI*3*3 is cut into a kernel【3】【3】; then the convolution operation is performed with the input data CI*H*W using the kernel【3】【3】, that is, the kernel【3】【3】 ] is the basic granularity of moving on the input data. A schematic diagram of moving in a specific way is shown in Figure 3b, where the box in 3b is the cut data after moving. Through the applicant's experiments, it was found that for different neural network models, the closer the size of the input data, that is, the value of CI*H*W and the number of preset input data in the trained model, the more accurate the calculation of the output results , the better the user experience.

To illustrate with a practical example, assuming that the preset input data of the trained neural network model can be: H=50, W=50, CI=64, then if the collected input data gets too few values, it is assumed that the composed The three-dimensional data is: H=20, W=20, CI=12, so no matter how many times of training, the more accurate the weight value is, the accuracy of the output result of the calculation is very low, and it is found through experiments that The greater the deviation between the number of convolution cuts and the number of cuts of the preset input data, the lower the accuracy of the output results, for example, H=50, W=50, H, W in a layer of CI in CI=64 The number of cuts is: 48*48; the number of cuts for H and W in a layer of CI of the collected input data is: 10*10, and the calculated numbers are also very different, so in order to solve this problem, the technology provided by this application The scheme adds element values (that is, the number of squares) through a preset strategy. The specific preset strategy can be to add data by adding zeros, so that the corresponding value can be achieved by adding zeros. Specifically, as shown in the figure As shown in 3c, the original input data is: H=9, W=7, CI=4, and the preset input data is H=18, W=7, CI=4; then the way of adding zeros can be as follows: The interlaced zero interpolation method is inserted into the original input data. The specific inserted data is shown in FIG. 3d, and the black interval in FIG. 3d is the position of the inserted zero value.

Of course, in practical applications, the above preset strategy can also be added in the form of an average value. Taking Figure 3d as an example, the black interval is the position of the inserted average value, where the average value is the value between two adjacent values in the H direction. For example, the 7 values in the second row in the H direction can be the average value of the 7 values in the first row in the H direction and the 7 values in the third row, corresponding to the value of the last row inserted, that is The value of the 18th row in the H direction may be the same as the value of the 17th row. It is found through experiments that the output data calculated by the average value method has higher precision than the output result calculated by the zero interpolation method.

Referring to FIG. 4, FIG. 4 provides a method for updating a drop model of an electronic device, the method is applied in an electronic device, and the electronic device includes: an application processor AP, a touch screen, a gravity sensor and a distance sensor, the The touch display screen is connected with the application processor through at least one circuit; as shown in Figure 4, the method includes:

Step S401, collecting acceleration data of the electronic device;

Step S402, collecting the distance between the electronic device and the ground;

Step S403, calculating the acceleration value according to the acceleration data, and determining the state of the electronic device according to the acceleration value, and the state includes: normal state and dropped state;

Step S404, extracting the acceleration value and the distance value in the falling state, compose the acceleration value and the distance value in the falling state into the first input data, and input the first input data into the first drop model for calculation to obtain the first output result, The first input data is used as training data to retrain the first drop model to obtain the second drop model, and the first input data is input to the second drop model to obtain the second output result, and the first output result and the second output result For comparison, if the second output result is greater than the first output result, the first drop model is replaced by the second drop model.

Referring to FIG. 5, FIG. 5 provides an electronic device, which includes: a housing, a circuit board, a battery, a cover, a gravity sensor 504, a touch screen 503, a distance sensor 501 and a processing unit 502, wherein,

a gravity sensor 504, configured to collect acceleration data of the electronic device;

The pressure sensor 501 is used to collect the distance between the electronic device and the ground;

The processing unit 502 is configured to calculate an acceleration value based on the acceleration data, and determine the state of the electronic device according to the acceleration value, and the state includes: a normal state and a dropped state;

The processing unit 502 is configured to extract the acceleration value and the distance value in the falling state, compose the acceleration value and the distance value in the falling state into first input data, and input the first input data into the first drop model for calculation to obtain the first Output the result, use the first input data as training data to retrain the first drop model to obtain the second drop model, input the first input data to the second drop model to obtain the second output result, and combine the first output result with the second drop model Comparing the two output results, if the second output result is greater than the first output result, the first drop model is replaced by the second drop model.

After the technical solution provided by the application collects the acceleration data, the acceleration value is calculated according to the acceleration data, and the pressure value of the casing is collected. The input data is composed, and the input data is input into the artificial neural network model for calculation to obtain an output result, so that the cause of the drop of the electronic device can be obtained according to the output result.

FIG. 6 shows a block diagram of a partial structure of a mobile phone related to the mobile terminal provided by the embodiment of the present application. Referring to FIG. 6, the mobile phone includes: a radio frequency (Radio Frequency, RF) circuit 910, a memory 920, an input unit 930, a sensor 950, an audio circuit 960, a wireless fidelity (Wireless Fidelity, WiFi) module 970, an application processor AP980, and a power supply 990 and other components. Those skilled in the art can understand that the structure of the mobile phone shown in FIG. 6 does not constitute a limitation to the mobile phone, and may include more or less components than shown in the figure, or combine some components, or arrange different components.

The following is a specific introduction to each component of the mobile phone in conjunction with Figure 6:

The input unit 930 can be used to receive input numbers or character information, and generate key signal input related to user settings and function control of the mobile phone. Specifically, the input unit 930 may include a touch screen 933 , a fingerprint recognition device 931 , a face recognition device 936 , an iris recognition device 937 and other input devices 932 . The input unit 930 may also include other input devices 932 . Specifically, other input devices 932 may include but not limited to one or more of physical keys, function keys (such as volume control keys, switch keys, etc.), trackball, mouse, joystick, and the like. in,

The sensor 950 is used to collect the acceleration data of the electronic device and the distance between the mobile phone and the ground, and transmit the acceleration data and the distance value to the AP980.

AP980 is used to calculate the acceleration value based on the acceleration data, and determine the state of the electronic device according to the acceleration value. The state includes: normal state and falling state; extract the acceleration value and distance value in the falling state, and convert the The acceleration value and the distance value form the first input data, input the first input data into the first drop model for calculation to obtain the first output result, and use the first input data as training data to retrain the first drop model to obtain the first drop model The second drop model, input the first input data to the second drop model to obtain the second output result, compare the first output result with the second output result, if the second output result is greater than the first output result, use the second drop model to replace The first drop model.

AP980 is the control center of the mobile phone. It uses various interfaces and lines to connect various parts of the entire mobile phone. By running or executing software programs and/or modules stored in the memory 920, and calling data stored in the memory 920, the mobile phone is executed. Various functions and processing data, so as to monitor the mobile phone as a whole. Optionally, the AP980 can include one or more processing units; optionally, the AP980 can integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and application programs, etc. The tuner processor mainly handles wireless communication. It can be understood that the above-mentioned modem processor may not be integrated into the AP980.

In addition, the memory 920 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

RF circuitry 910 may be used for the reception and transmission of information. Generally, the RF circuit 910 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (Low Noise Amplifier, LNA), a duplexer, and the like. In addition, RF circuitry 910 may also communicate with networks and other devices via wireless communications. The above wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile Communication (Global System of Mobilecommunication, GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access, CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Messaging Service (Short Messaging Service, SMS), etc.

The handset may also include at least one sensor 950, such as a light sensor, motion sensor, and other sensors. Specifically, the light sensor can include an ambient light sensor and a proximity sensor, wherein the ambient light sensor can adjust the brightness of the touch display screen according to the brightness of the ambient light, and the proximity sensor can turn off the touch display screen when the mobile phone is moved to the ear. and/or backlighting. As a kind of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in various directions (generally three axes), and can detect the magnitude and direction of gravity when it is stationary, and can be used to identify the application of mobile phone posture (such as horizontal and vertical screen switching, related Games, magnetometer attitude calibration), vibration recognition related functions (such as pedometer, tap), etc.; as for other sensors such as gyroscope, barometer, hygrometer, thermometer, infrared sensor, etc. repeat.

The audio circuit 960, the speaker 961, and the microphone 962 can provide an audio interface between the user and the mobile phone. The audio circuit 960 can transmit the electrical signal converted from the received audio data to the loudspeaker 961, and the loudspeaker 961 converts it into a sound signal for playback; After being received, it is converted into audio data, and then the audio data is processed by the playback AP980, and then sent to another mobile phone through the RF circuit 910, or the audio data is played to the memory 920 for further processing.

WiFi is a short-distance wireless transmission technology. The mobile phone can help users send and receive emails, browse web pages, and access streaming media through the WiFi module 970. It provides users with wireless broadband Internet access. Although FIG. 6 shows a WiFi module 970, it can be understood that it is not an essential component of the mobile phone, and can be completely omitted according to needs without changing the essence of the application.

The mobile phone also includes a power supply 990 (such as a battery) for supplying power to various components. Optionally, the power supply can be logically connected to the AP980 through the power management system, so that functions such as charging, discharging, and power consumption management can be realized through the power management system.

Although not shown, the mobile phone may also include a camera, a Bluetooth module, a supplementary light device, a light sensor, etc., which will not be repeated here.

It can be seen that through the embodiment of the present application, after the acceleration data is collected, the state of the electronic device is determined according to the acceleration data. When it is determined to be in a falling state, the first picture of the ground is collected through the camera, and then the electronic device is obtained according to the acceleration value and the collection time. The distance to the ground of the device is extracted from a second picture of the electronic device (specifically, an outline picture), so that a 3D animation with the electronic device falling to the ground can be generated, which improves user experience.

An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables the computer to execute any drop model of an electronic device as described in the method embodiments above. Update some or all steps of the method.

The embodiment of the present application also provides a computer program product, the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable the computer to execute the method described in the above method embodiments Some or all steps of the method for updating the drop model of any electronic device.

It should be noted that for the foregoing method embodiments, for the sake of simple description, they are expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the described action sequence. Depending on the application, certain steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the application.

In the foregoing embodiments, the descriptions of each embodiment have their own emphases, and for parts not described in detail in a certain embodiment, reference may be made to relevant descriptions of other embodiments.

In the several embodiments provided in this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or can be Integrate into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented not only in the form of hardware, but also in the form of software program modules.

The integrated units may be stored in a computer-readable memory if implemented in the form of a software program module and sold or used as an independent product. Based on this understanding, the technical solution of the present application is essentially or part of the contribution to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory. Several instructions are included to make a computer device (which may be a personal computer, server or network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above-mentioned embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash disk , Read-only memory (English: Read-Only Memory, abbreviated: ROM), random access device (English: Random Access Memory, abbreviated: RAM), magnetic disk or optical disk, etc.

The embodiments of the present application have been introduced in detail above, and specific examples have been used in this paper to illustrate the principles and implementation methods of the present application. The descriptions of the above embodiments are only used to help understand the methods and core ideas of the present application; meanwhile, for Those skilled in the art will have changes in specific implementation methods and application scopes based on the ideas of the present application. In summary, the contents of this specification should not be construed as limiting the present application.

Claims (11)

1. An electronic device, the electronic device comprising: an application processor AP, a touch display screen, a gravity sensor and a distance sensor,

the gravity sensor is used for acquiring acceleration data of the electronic device;

the distance sensor is used for acquiring the distance between the electronic device and the ground;

the AP is used for calculating an acceleration value according to the acceleration data, and determining the state of the electronic device according to the acceleration value, wherein the state comprises the following steps: a normal state and a fall state;

the AP is used for extracting an acceleration value and a distance value under a falling state, forming first input data by the acceleration value and the distance value under the falling state, inputting the first input data into a first falling model to calculate to obtain a first output result, retraining the first falling model by taking the first input data as training data to obtain a second falling model, inputting the first input data into the second falling model to obtain a second output result, comparing the first output result with the second output result, and if the second output result is greater than the first output result, replacing the first falling model by the second falling model.

2. The electronic device of claim 1, further comprising: a communication module;

and the AP is also used for controlling a communication module to send the weight data of the second drop model to network side equipment.

3. The electronic device of claim 1,

the AP is specifically used for traversing n acceleration values according to the sequence of the acquisition points if the n acceleration values exist, determining the AP to be in a falling state if the m continuous acceleration values are larger than a set threshold value, and otherwise determining the AP to be in a common state.

4. The electronic device of claim 3,

the AP is specifically used for inputting the first input data as training data into the first falling model to execute V-layer forward operation to obtain a V-layer forward operation result of the forward operation, processing the V-layer forward operation result to obtain a V-layer input data gradient, inputting the V-layer input data gradient into the V-layer of the first falling model to execute V-layer reverse operation to obtain V weight gradients, updating the V-layer weight by adopting the V weight gradients to obtain updated weight data, and determining the updated weight data as the weight data of the second falling model.

5. A drop model updating method of an electronic device is applied to the electronic device, and the electronic device comprises the following steps: the method comprises the following steps that an application processor AP, a touch display screen, a gravity sensor and a distance sensor are adopted, and the method comprises the following steps:

acquiring acceleration data of the electronic device;

collecting the distance between the electronic device and the ground;

calculating an acceleration value according to the acceleration data, and determining the state of the electronic device according to the acceleration value, wherein the state comprises the following steps: a normal state and a fall state;

the method comprises the steps of extracting an acceleration value and a distance value under a falling state, forming first input data by the acceleration value and the distance value under the falling state, inputting the first input data into a first falling model to calculate to obtain a first output result, retraining the first falling model by taking the first input data as training data to obtain a second falling model, inputting the first input data into the second falling model to obtain a second output result, comparing the first output result with the second output result, and replacing the first falling model by the second falling model if the second output result is greater than the first output result.

6. The method of claim 5, further comprising:

and sending the weight data of the second drop model to network side equipment.

7. The method of claim 5, wherein determining the state of the electronic device according to the n acceleration values comprises:

and traversing the n acceleration values according to the sequence of the acquisition points, if the continuous m acceleration values are larger than a set threshold value, determining the falling state, and otherwise, determining the falling state as the common state.

8. The method of claim 7, wherein retraining the first fall model using the first input data as training data to obtain a second fall model comprises:

inputting the first input data as training data into a first falling model to execute V-layer forward operation to obtain a V-layer forward operation result of forward operation, processing the V-layer forward operation result to obtain a V-layer input data gradient, inputting the V-layer input data gradient into a V-layer of the first falling model to execute V-layer reverse operation to obtain V weight gradients, updating the V-layer weight by adopting the V weight gradients to obtain updated weight data, and determining the updated weight data as the weight data of a second falling model.

9. An electronic device, the electronic device comprising: a processing unit, a touch display screen, a gravity sensor, a circuit and a distance sensor,

the gravity sensor is used for acquiring acceleration data of the electronic device;

the distance sensor is used for acquiring the distance between the electronic device and the ground;

the processing unit is used for calculating an acceleration value according to the acceleration data, and determining the state of the electronic device according to the acceleration value, wherein the state comprises the following steps: a normal state and a fall state;

the processing unit is used for extracting an acceleration value and a distance value under a falling state, forming first input data by the acceleration value and the distance value under the falling state, inputting the first input data into a first falling model to calculate to obtain a first output result, retraining the first falling model by taking the first input data as training data to obtain a second falling model, inputting the first input data into the second falling model to obtain a second output result, comparing the first output result with the second output result, and if the second output result is greater than the first output result, replacing the first falling model by the second falling model.

10. A computer-readable storage medium, characterized in that it stores a computer program for electronic data exchange, wherein the computer program causes a computer to perform the method according to any one of claims 5-8.

11. A computer program product, characterized in that the computer program product comprises a non-transitory computer readable storage medium storing a computer program operable to cause a computer to perform the method according to any of claims 5-8.