Key quality detection method and apparatus

Abstract

The present disclosure relates to a key quality detection method and device, belonging to the technical field of key detection. The method comprises: collecting pressing force data from a sensor and displacement data of a key when the key is forced to move; obtaining first pressing force time series data and first displacement time series data for quality detection of the key according to the pressing force data and the displacement data; establishing a first network model, and generating first input data suitable for the first network model according to the first pressing force time series data and the first displacement time series data, and establishing a second network model, and generating second input data suitable for the second network model according to the first pressing force time series data and the first displacement time series data; inputting the first input data and the second input data into the first network model and the second network model respectively for training to obtain first training results and second training results; and obtaining a detection result of the key quality according to a set judgment strategy and the first training results and the second training results.

Description

Technical Field

[0001] This disclosure relates to the field of button detection technology, and more specifically, to a button quality detection method and apparatus. Background Technology

[0002] Currently, with the rapid development of electronic devices, users can output corresponding control commands by pressing buttons on electronic devices such as remote controls, game controllers, and smartwatches. The tactile feedback provided by the buttons during button presses affects the user's actual operating experience, thus necessitating button quality testing. Button quality testing typically involves setting thresholds for the peak and trough values ​​of the pressure applied to the button. However, changes in materials, environment, and other factors can alter the button's response to pressure, causing corresponding changes in the peak and trough values ​​of the applied pressure. Summary of the Invention

[0003] One objective of this disclosure is to provide a new technical solution for a method and apparatus for detecting button quality.

[0004] According to a first aspect of this disclosure, a method for detecting button quality is provided. The method includes: when a button is subjected to force, acquiring pressure data and displacement data from a sensor; obtaining first pressure timing data and first displacement timing data for quality detection of the button based on the pressure data and displacement data; wherein the first pressure timing data reflects the change in pressure applied to the button over time within a detection period, and the first displacement timing data reflects the change in displacement generated by the button over time within the detection period; establishing a first network model and generating first input data adapted to the first network model based on the first pressure timing data and the first displacement timing data; establishing a second network model and generating second input data adapted to the second network model based on the first pressure timing data and the first displacement timing data; inputting the first input data and the second input data into the first network model and the second network model respectively for training to obtain a first training result and a second training result; and obtaining a button quality detection result using the first training result and the second training result according to a set judgment strategy.

[0005] Optionally, the first input data reflects the correspondence between the pressure applied to the button, the displacement generated by the button, and the button movement stages, wherein the button movement stages include a pressing stage and a lifting stage.

[0006] Optionally, the first network model is a deep network model, and the first network model further includes a first-dimensional channel, a second-dimensional channel, and a third-dimensional channel; the step of generating first input data adapted to the first network model based on the first pressing pressure timing data and the first displacement timing data includes: inputting the first pressing pressure timing data, the first displacement timing data, and the values ​​representing the key movement stage into the first-dimensional channel, the second-dimensional channel, and the third-dimensional channel respectively, to generate image data adapted to the first network model as the first input data; wherein, the pressing stage corresponds to the first value, and the lifting stage corresponds to the second value.

[0007] Optionally, the second input data reflects at least one extreme value of the pressure applied to the button and the displacement generated by the button, as well as the average value of the pressure applied to the button and the displacement generated by the button over multiple compression periods.

[0008] Optionally, the second network model is an ensemble learning model; the step of generating second input data adapted to the second network model based on the first pressing pressure time-series data and the first displacement time-series data includes: extracting at least one extreme value information for the first pressing pressure time-series data and the first displacement time-series data, and extracting mean value information for the first pressing pressure time-series data and the first displacement time-series data; constructing a first sequence data by combining at least one extreme value information and mean value information for the first pressing pressure time-series data in time order, and constructing a second sequence data by combining at least one extreme value information and mean value information for the first displacement time-series data in time order; and using the first sequence data and the second sequence data as second input data suitable for the second network model.

[0009] Optionally, the at least one extreme value includes the global extreme value of the pressure applied to the button and the displacement generated by the button within the detection cycle, and the local extreme value within different stages of the detection cycle.

[0010] Optionally, obtaining first pressure timing data and first displacement timing data for quality detection of the button based on the pressure data and displacement data includes: dividing the detection period into multiple sampling periods; obtaining the pressure value of the button in each of the multiple sampling periods as second pressure timing data based on the pressure data; obtaining the displacement value of the button in each of the multiple sampling periods as second displacement timing data based on the displacement data; and obtaining the first pressure timing data and first displacement timing data based on the second pressure timing data and the second displacement timing data.

[0011] Optionally, obtaining the key quality detection result using the first training result and the second training result according to the set judgment strategy includes: filtering out the maximum probability value among the probability values ​​represented by the first training result and the second training result respectively through the set judgment strategy; comparing the maximum probability value with a set threshold, and obtaining the detection result based on the comparison result.

[0012] According to a second aspect of this disclosure, a button quality detection device is also provided. The device includes: a data acquisition module for acquiring pressure data and displacement data from sensors when the button is subjected to force; a data acquisition module for acquiring first pressure timing data and first displacement timing data for quality detection of the button based on the pressure data and displacement data; wherein the first pressure timing data reflects the change in pressure applied to the button over time within the detection period, and the first displacement timing data reflects the change in displacement generated by the button over time within the detection period; a data generation module for establishing a first network model and generating first input data adapted to the first network model based on the first pressure timing data and the first displacement timing data, and establishing a second network model and generating second input data adapted to the second network model based on the first pressure timing data and the first displacement timing data; a training result obtaining module for inputting the first input data and the second input data into the first network model and the second network model respectively for training to obtain a first training result and a second training result; and a detection result obtaining module for obtaining a button quality detection result based on a set judgment strategy using the first training result and the second training result.

[0013] According to a third aspect of this disclosure, a key quality detection device is also provided, comprising a memory and a processor, the memory being used to store a computer program; the processor being used to execute the computer program to implement the method according to a first aspect of this disclosure.

[0014] According to a fourth aspect of this disclosure, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the method described according to a first aspect of this disclosure.

[0015] One beneficial effect of this embodiment is that by performing quality detection on the button to obtain first pressure timing data and first displacement timing data, and then processing the first pressure timing data and first displacement timing data respectively to obtain first input data and second input data, the first input data can be input into a first network model to obtain a first training result, and the second input data can be input into a second network model to obtain a second training result, thereby obtaining a detection result regarding the button quality. By using the first network model and the second network model to detect the button's pressure and displacement, the accuracy of button quality detection can be improved.

[0016] Other features and advantages of the embodiments of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments of the present disclosure and, together with their description, serve to explain the principles of the embodiments of the present disclosure.

[0018] Figure 1 This is a schematic diagram of the composition structure of a key quality detection system capable of applying a key quality detection method according to an embodiment.

[0019] Figure 2 This is a flowchart illustrating a button quality testing method according to another embodiment;

[0020] Figure 3 This is a block schematic diagram of an electronic device according to another embodiment;

[0021] Figure 4 This is a schematic diagram of the hardware structure of an electronic device according to another embodiment. Detailed Implementation

[0022] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention.

[0023] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0024] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0025] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0026] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0027] <System Implementation Example>

[0028] Figure 1 This is a schematic diagram of the structural composition of a key quality detection system capable of applying a key quality detection method according to one embodiment. For example... Figure 1 As shown, the system includes a button quality detection device, a controller 200, a motor 300, a force sensor 400, a displacement sensor 500, and a pressing head 600. The button quality detection device can be a host computer 100, the motor 300 can be a voice coil motor, and the displacement sensor 500 can be a grating sensor. This system can be applied in human-computer interaction scenarios.

[0029] The host computer 100 can be, but is not limited to, various smartphones, personal computers, laptops, and tablets. The host computer 100 can control the controller 200 to turn on, off, and reset, and can also send operation commands to the controller 200 to cause the motor 300 to turn on, off, and reset.

[0030] The controller 200 can be directly electrically connected to the host computer 100, or it can be connected wirelessly via Bluetooth, which is not limited here. The output of the controller 200 is connected to the motor 300 and drives the motor 300 to move in order to control the pressing pressure of the motor 300 on the button 1000.

[0031] Motor 300 can be a voice coil motor. The moving end of the voice coil motor has a good response speed. The moving end of motor 300 can move to the corresponding position in response to the command output by controller 200.

[0032] The press head 600 is located between the force sensor 400 and the button 1000 and is fixedly connected to the force sensor 400. The press head 600 is in contact with the button 1000, and when the button 1000 is subjected to a force applied by the user, the press head 600 can transmit its own deformation to the force sensor 400.

[0033] Force sensor 400 is electrically connected to controller 200. The end of force sensor 400 away from the pressing head can be installed on the moving end of motor 300. Force sensor 400 can convert the force transmitted by pressing head 600 into an electrical signal and output the electrical signal to controller 200 through amplifier 410.

[0034] The displacement sensor 500 is electrically connected to the controller 200. The displacement sensor 500 can be a grating sensor. The grating sensor can obtain the movement distance of the button 1000 and output it to the controller 200 in the form of an electrical signal.

[0035] The system may also include a limit switch 700, which may be as follows: Figure 1 The photoelectric switch shown detects whether the moving end of the motor 300 has moved to its limit position. When the host computer 100 detects that the moving end of the motor 300 has reached its limit position, it sends feedback to the controller 200, which then controls the motor 300 to stop moving or to retract in the opposite direction, thus limiting the movement of the motor 300.

[0036] In the embodiments of this disclosure, the memory of the key quality detection device is used to store a computer program that controls the processor of the key quality detection device to operate in order to implement the key quality detection method according to any embodiment. Those skilled in the art can design the computer program based on the solutions of the embodiments of this disclosure. How the computer program controls the processor to operate is well known in the art and will not be described in detail here.

[0037] <Method Implementation>

[0038] Figure 2 This is a flowchart illustrating a button quality detection method according to one embodiment. In this embodiment, the aforementioned button quality detection device is used as the execution entity.

[0039] like Figure 2 As shown, the button quality detection method of this embodiment may include the following steps S201 to S205:

[0040] Step S201: When the button is subjected to force and moves, collect the pressing force data and the button displacement data from the sensor.

[0041] The force sensor deployed on the button can convert the pressing force into data, i.e., pressing force data, and the displacement sensor can convert the distance the button moves into data, i.e., displacement data.

[0042] Specifically, when a user presses a button, the button moves due to the pressure applied by the user. The force sensor can collect the pressing force data, and the displacement sensor can collect the displacement data and output them to the button quality detection device.

[0043] Step S202: Based on the pressing pressure data and displacement data, obtain the first pressing pressure timing data and the first displacement timing data for quality detection of the button; wherein, the first pressing pressure timing data is data reflecting the change of pressing pressure on the button over time during the detection cycle, and the first displacement timing data is data reflecting the change of displacement generated by the button over time during the detection cycle.

[0044] Specifically, the entire process of the button's movement change when the user presses the aforementioned press head can be considered as a detection cycle. During this detection cycle, the press head transmits the force received to the force sensor, while the displacement sensor obtains the distance the press head moves. This allows the corresponding button quality detection device to obtain data reflecting the change in the pressing force of the button over time within this detection cycle; this data is the first pressing force timing data. The button quality detection device can also obtain data reflecting the change in the displacement of the button over time within this detection cycle; this data is the first displacement timing data.

[0045] In one embodiment, step S202 specifically includes the following: dividing the detection period into multiple sampling periods; obtaining the pressing force value of the button in each sampling period of the multiple sampling periods as second pressing force timing data based on the pressing force data; obtaining the displacement value of the button in each sampling period of the multiple sampling periods as second displacement timing data based on the displacement data; and obtaining first pressing force timing data and first displacement timing data based on the second pressing force timing data and the second displacement timing data.

[0046] Specifically, the detection cycle is divided into multiple sampling periods, each of equal length. The pressure and displacement values ​​for each sampling period within the detection cycle are determined from the pressure and displacement data. If multiple pressure or displacement values ​​exist within a single sampling period, the average of these values ​​is taken, making the sampling period and the corresponding average a second pressure timing data set or a second displacement timing data set. If no pressure or displacement value exists within a single sampling period, the average of the pressure or displacement values ​​from two adjacent sampling periods is taken, making the average and the corresponding sampling period a second pressure timing data set or a second displacement timing data set. By selecting the second pressure and second displacement timing data from the same sampling period, the average pressure and displacement values ​​across multiple sampling periods can be obtained, effectively reducing the amount of data that the button quality detection device needs to process while ensuring data integrity.

[0047] Similarly, the second pressure time-series data or the second displacement time-series data can also be normalized. Specifically, for example, the maximum and minimum values ​​of the button's pressure and displacement can be set. Based on the maximum and minimum pressure values, the mean pressure value in the second pressure time-series data is normalized using the maximum-minimum normalization method; similarly, based on the maximum and minimum displacement values, the mean displacement value in the second displacement time-series data is normalized using the maximum-minimum normalization method. This results in normalized pressure and displacement data, which correspond one-to-one with the corresponding data collection periods. In other words, the normalized data has a more uniform specification, reducing the difficulty of subsequently constructing the first or second network model.

[0048] Step S203: Establish a first network model and generate first input data adapted to the first network model based on the first pressing pressure time series data and the first displacement time series data; establish a second network model and generate second input data adapted to the second network model based on the first pressing pressure time series data and the first displacement time series data.

[0049] Specifically, a first network model and a second network model are pre-established. These two network models can be different probability calculation models to obtain corresponding probability data in two different ways. Accordingly, the first pressing pressure time series data and the first displacement time series data need to be processed for different network models to adapt to the corresponding network models.

[0050] Specifically, the first pressure timing data and the first displacement timing data are processed. This processing may involve classifying the pressure and displacement based on different button movement stages and time intervals, allowing the processed first pressure and displacement timing data to be input into the first network model. The first pressure and displacement timing data are then processed further, specifically by extracting feature values ​​such as initial value, final value, maximum value, and minimum value. These extracted feature values ​​from the first pressure and displacement timing data are then input into the second network model.

[0051] Step S204: Input the first input data and the second input data into the first network model and the second network model respectively to train and obtain the first training result and the second training result.

[0052] Specifically, the probability data obtained by outputting the first input data into the first network model is used as the first training result. The probability data obtained by outputting the second input data into the second network model is used as the second training result. In other words, using probability data output from two different network models can effectively improve the accuracy of subsequent button quality judgments.

[0053] In one embodiment, the first input data can reflect the correspondence between the pressure applied to the key, the displacement generated by the key, and the key movement stages, wherein the key movement stages include a pressing stage and a lifting stage.

[0054] Since the force exerted on a button during the pressing process is different from that during the rebound process, the button movement phase can include a pressing phase and a lifting phase.

[0055] Specifically, the first pressing pressure time-series data is divided into different data points according to the pressing and releasing phases, and the first displacement time-series data is also divided into different data points according to the pressing and releasing phases. This allows us to obtain the correspondence between the pressing pressure, the displacement generated by the button, and different button movement phases, enabling us to classify and process the first pressing pressure and first displacement time-series data for different button movement phases. In other words, based on the different trends in pressing pressure and displacement during different button movement phases, the first model can process the first input data in different cases, resulting in a higher degree of realism in the output first training result.

[0056] In one embodiment, the first network model is a deep neural network model, which further includes a first-dimensional channel, a second-dimensional channel, and a third-dimensional channel. The process of generating first input data adapted to the first network model specifically includes: inputting first pressure timing data, first displacement timing data, and values ​​representing the button movement phases into the first-dimensional channel, the second-dimensional channel, and the third-dimensional channel, respectively, to generate image data adapted to the first network model as the first input data. The pressing phase corresponds to the first value, and the releasing phase corresponds to the second value.

[0057] Specifically, the detection period can be divided into N sampling periods, where N > 0 and N is an integer. Each sampling period corresponds to a pressure value and a displacement value. Based on the changing trends of the pressure and displacement values ​​in each sampling period, the button movement stage corresponding to each sampling period is determined. The two button movement stages correspond to different values; for example, the first value corresponding to the pressing stage could be 1.2, and the second value corresponding to the releasing stage could be 1. Therefore, a three-dimensional matrix is ​​constructed based on the fact that each of the N sampling periods corresponds to a pressure value, a displacement value, and a button movement stage value. The first input data can include the values ​​corresponding to the first dimension, the second dimension, and the third dimension. The first dimension can represent the pressure value, the second dimension can represent the displacement value, and the third dimension can represent the value of the pressing movement stage. The first input data is input into the first model, which is a deep neural network model. The first network model includes a first-dimensional channel inputting the value corresponding to the first dimension, a second-dimensional channel inputting the value corresponding to the second dimension, and a third-dimensional channel inputting the value corresponding to the third dimension. The first network model assigns different colors to different key movement stages input to the third-dimensional channel; for example, blue corresponds to the pressing stage, and red corresponds to the releasing stage. It constructs an N×N planar image using the pressing force value as the ordinate, the displacement value as the abscissa, and setting the corresponding key movement stage value at each coordinate position. All sampling stages are then sampled and input into the first network model for training, yielding the first training result, i.e., the confidence matrix data. The specific expression for generating the confidence matrix data can be as follows:

[0058] D image =f matrix (D sensor ) Formula (1)

[0059] In formula (1), D sensor D represents the pressure and displacement values ​​at different sampling times. image f represents the confidence matrix data. matrix () indicates a matrix transformation algorithm, which is an existing technology and will not be elaborated here.

[0060] Optionally, the first network model is a pre-trained deep neural network model, which can be built based on a convolutional neural network. Inputting the three-dimensional matrix data into the deep neural network model yields a first training result, which can be represented as a confidence matrix P1. The confidence matrix P1 can include the prediction probability of the deep neural network model for excellent button quality. In other words, the confidence matrix data output by the deep neural network model can effectively improve the accuracy of judging the quality of buttons.

[0061] In one embodiment, the second input data may reflect at least one extreme value of the pressure applied to the button and the displacement generated by the button, as well as the average value of the pressure applied to the button and the displacement generated by the button over multiple compression periods.

[0062] Specifically, extreme values ​​of the pressure applied to the button and the displacement generated by the button can be determined during the detection cycle. These extreme values ​​include extreme values ​​of pressure and displacement during the detection cycle. Extreme values ​​of pressure can include the maximum and minimum values ​​of pressure, and extreme values ​​of displacement can include the maximum and minimum values ​​of displacement. Several pressure and displacement values ​​between each extreme value are determined, and these values ​​are compressed using a data compression algorithm to obtain their average values. This yields the average values ​​of the pressure applied to the button and the displacement generated by the button over multiple compression periods. In other words, extreme values ​​of pressure and displacement can largely reflect the trend of pressure and displacement changes, while the average values ​​of pressure and displacement can reflect the degree of change in pressure and displacement during the process. Therefore, obtaining the corresponding second input data using the selected extreme values ​​and average values ​​can reduce memory usage while ensuring data integrity.

[0063] In one embodiment, the second network model is an ensemble learning model; the process of generating the second input data specifically includes the following: extracting at least one extreme value information for the first pressing pressure time series data and the first displacement time series data, and extracting the mean value information for the first pressing pressure time series data and the first displacement time series data; constructing a first sequence data by combining at least one extreme value information and the mean value information for the first pressing pressure time series data in time sequence, and constructing a second sequence data by combining at least one extreme value information and the mean value information for the first displacement time series data in time sequence; and using the first sequence data and the second sequence data as the second input data applicable to the second network model.

[0064] Specifically, the extreme value information and mean value information of the first pressing force time series data are concatenated using a function and output to obtain the first sequence data. Similarly, the extreme value information and mean value information of the first displacement time series data are concatenated using a function and output to obtain the second sequence data. The first and second sequence data are then used as the second input data. The function can be a `concatenate` function. The expression for obtaining the sequence data can be as follows:

[0065] D sequence =concatenate(f feature (D sensor ),f compress (D sensor )) Formula (2)

[0066] In formula (2), D sensor D represents the extreme values ​​and mean values ​​selected above. sequence f represents sequence data feature () represents the feature extraction algorithm, f compress () indicates a data compression algorithm. The feature extraction algorithm and data compression algorithm here are existing technologies and will not be described in detail.

[0067] In one embodiment, the at least one extreme value may include the global extreme value of the pressure applied to the button and the displacement generated by the button during the detection cycle, and the local extreme value during different stages of the detection cycle.

[0068] Specifically, at least one extreme value may include the pressing force and displacement value when the pressing force reaches its maximum value within the detection cycle, the local maxima and local minima corresponding to the pressing force and displacement during the pressing phase within the detection cycle, and the local maxima and local minima corresponding to the pressing force and displacement during the releasing phase within the detection cycle. In other words, filtering extreme values ​​for the overall detection cycle and different key movement phases can increase the number of extreme value points obtained, thereby further improving the accuracy of the second input data.

[0069] Furthermore, the second network model is an ensemble learning model.

[0070] The second input data is fed into the ensemble learning model. The ensemble learning model can employ either the serial method XGBoost or the parallel method Random Forest. The training parameters of the ensemble learning model can be optimized using Bayesian optimization methods to find the optimal configuration.

[0071] Specifically, inputting the second input data into the ensemble learning model yields a second training result, which can be represented as a confidence matrix P2. The confidence matrix P2 can include the ensemble learning model's prediction probability of excellent button quality. In other words, by inputting sequence data into the ensemble learning model, the accuracy of judging excellent button quality can be effectively improved.

[0072] More notably, to improve the accuracy of the training results output by the deep neural network model and the ensemble learning model, pressure timing data and displacement timing data of the key detection process can be collected, and key quality can be determined based on these data. Data labels are generated based on the quality determination results. The process of training the deep neural network model and the ensemble learning model using data labels is existing technology and will not be elaborated here. Through the well-trained deep neural network model and the ensemble learning model, the accuracy of the first and second training results output by the deep neural network model and the ensemble learning model, respectively, can be improved.

[0073] Step S205: Based on the set judgment strategy, obtain the detection result of the button quality using the first training result and the second training result.

[0074] Specifically, when the button quality detection device obtains the first training result and the second training result, it can obtain the detection result of the button quality by averaging the two or comparing the two, so that the user can judge whether the button is of good quality through the detection result.

[0075] In one embodiment, step S205 specifically includes the following: selecting the maximum probability value among the probability values ​​represented by the first training result and the second training result respectively through a set judgment strategy; comparing the maximum probability value with a set threshold, and obtaining the detection result based on the comparison result.

[0076] First, a judgment strategy can be predefined, and the expression for this judgment strategy is as follows:

[0077]

[0078] In formula (3), R represents the maximum probability value, max() represents extracting the maximum value, .ndex represents extracting the corresponding sampling period, P1 represents the first training result, and P2 represents the second training result.

[0079] Accordingly, the first training result output by the first network model can be a probability value, and the second training result output by the second network model can also be a probability value. The first training result and the second training result are input into formula (3). When the probability values ​​represented by the first training result and the second training result are equal, P1 is taken, which is the first training result. When the probability values ​​represented by the first training result and the second training result are not equal, the larger value is taken, and the value output by formula (3) is the maximum probability value obtained. The maximum probability value obtained is then compared with a preset threshold to obtain a comparison result, and the detection result is obtained based on the comparison result. For example: when the maximum probability value is less than the preset threshold, the comparison result can be "0", and the corresponding detection result can be that the button is unqualified. When the maximum probability value is not less than the preset threshold, the comparison result can be "1", and the corresponding detection result can be that the button is qualified.

[0080] <Equipment Example 1>

[0081] Figure 3 This is a schematic block diagram of an electronic device according to one embodiment. For example... Figure 4 As shown, the button quality detection device 310 may include a data acquisition module 311, a data acquisition module 312, a data generation module 313, a training result acquisition module 314, and a detection result acquisition module 315.

[0082] The data acquisition module 311 is used to acquire pressing force data and displacement data of the button from the sensor when the button is subjected to force and moves; the data acquisition module 312 is used to acquire first pressing force timing data and first displacement timing data for quality detection of the button based on the pressing force data and displacement data; wherein, the first pressing force timing data reflects the change of pressing force on the button over time during the detection period, and the first displacement timing data reflects the change of displacement generated by the button over time during the detection period; the data generation module 313 is used to establish a first network model and generate first input data adapted to the first network model based on the first pressing force timing data and the first displacement timing data, and to establish a second network model and generate second input data adapted to the second network model based on the first pressing force timing data and the first displacement timing data; the training result obtaining module 314 is used to input the first input data and the second input data into the first network model and the second network model respectively for training to obtain a first training result and a second training result; the detection result obtaining module 315 is used to obtain the detection result of the button quality based on the first training result and the second training result according to the set judgment strategy.

[0083] Optionally, the data generation module 313 is further configured to input the first pressing pressure timing data, the first displacement timing data, and the values ​​representing the key movement stages into the first dimension channel, the second dimension channel, and the third dimension channel, respectively, to generate image data adapted to the first network model as the first input data; wherein, the pressing stage corresponds to the first value, and the lifting stage corresponds to the second value.

[0084] Optionally, the data acquisition module 312 is further configured to divide the detection period into multiple sampling periods; obtain the pressing force value of the button in each sampling period in the multiple sampling periods as second pressing force timing data based on the pressing force data; obtain the displacement value of the button in each sampling period in the multiple sampling periods as second displacement timing data based on the displacement data; and obtain first pressing force timing data and first displacement timing data based on the second pressing force timing data and the second displacement timing data.

[0085] Optionally, the detection result obtaining module 315 is further configured to obtain the maximum probability value of judging the button quality as qualified based on the set judgment strategy, using the first training result and the second training result; compare the maximum probability value with the set threshold, and obtain the detection result based on the comparison result.

[0086] The button quality detection device 310 can be Figure 1 The host computer 100 in the middle.

[0087] <Equipment Example 2>

[0088] Figure 4 This is a schematic diagram of the hardware structure of a button quality detection device according to another embodiment.

[0089] like Figure 4 As shown, the button quality detection device 420 includes a processor 421 and a memory 422. The memory 422 is used to store an executable computer program, and the processor 421 is used to execute the method as described in any of the above method embodiments under the control of the computer program.

[0090] The button quality detection device 420 can be Figure 1 The host computer 100 in the middle.

[0091] Each module of the button quality detection device 310 described above can be implemented by the processor 421 executing the computer program stored in the memory 422 in this embodiment, or it can be implemented by other structures, which are not limited here.

[0092] This invention can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the invention.

[0093] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0094] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0095] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0096] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0097] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0098] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0099] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It will be known to those skilled in the art that implementation in hardware, implementation in software, and implementation using a combination of software and hardware are equivalent.

[0100] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of the invention is defined by the appended claims.

Claims

1. A method for detecting button quality, characterized in that, The method includes: When the button is pressed, the pressure data and displacement data of the button are collected from the sensor; Based on the pressure data and displacement data, first pressure timing data and first displacement timing data for quality detection of the button are obtained; wherein, the first pressure timing data is data reflecting the change of the pressure applied to the button over time during the detection period, and the first displacement timing data is data reflecting the change of the displacement generated by the button over time during the detection period. A first network model is established, and first input data adapted to the first network model is generated based on the first pressing pressure time series data and the first displacement time series data; a second network model is established, and second input data adapted to the second network model is generated based on the first pressing pressure time series data and the first displacement time series data. The first input data and the second input data are respectively input into the first network model and the second network model for training to obtain the first training result and the second training result; Based on the established judgment strategy, the detection result of the key quality is obtained using the first training result and the second training result.

2. The method according to claim 1, characterized in that, The first input data reflects the correspondence between the pressure applied to the button, the displacement generated by the button, and the button movement stages, wherein the button movement stages include a pressing stage and a lifting stage.

3. The method according to claim 2, characterized in that, The first network model is a deep neural network model, and the first network model also includes a first-dimensional channel, a second-dimensional channel, and a third-dimensional channel; The step of generating first input data adapted to the first network model based on the first pressing pressure time series data and the first displacement time series data includes: The first pressing pressure timing data, the first displacement timing data, and the value representing the key movement stage are respectively input into the first dimension channel, the second dimension channel, and the third dimension channel to generate image data adapted to the first network model as the first input data; The pressing phase corresponds to the first value, and the lifting phase corresponds to the second value.

4. The method according to claim 1, characterized in that, The second input data reflects at least one extreme value of the pressure applied to the button and the displacement generated by the button, as well as the average value of the pressure applied to the button and the displacement generated by the button over multiple compression periods.

5. The method according to claim 4, characterized in that, The second network model is an ensemble learning model; The step of generating second input data adapted to the second network model based on the first pressing pressure time series data and the first displacement time series data includes: Extract at least one extreme value information for the first pressing force time series data and the first displacement time series data, and extract the mean value information for the first pressing force time series data and the first displacement time series data; The first sequence data is constructed by arranging at least one extreme value information and mean value information of the first pressing force time series data in time sequence, and the second sequence data is constructed by arranging at least one extreme value information and mean value information of the first displacement time series data in time sequence. The first sequence data and the second sequence data are used as the second input data for the second network model.

6. The method according to claim 4, characterized in that, The at least one extreme value includes the global extreme value of the pressure applied to the button and the displacement generated by the button within the detection cycle, and the local extreme value within different stages of the detection cycle.

7. The method according to any one of claims 1 to 6, characterized in that, The step of obtaining first pressure timing data and first displacement timing data for quality detection of the button based on the pressure data and displacement data includes: The detection period is divided into multiple sampling time periods; Based on the pressure data, the pressure value of the button in each of the multiple sampling periods is obtained as the second pressure timing data; Based on the displacement data, the displacement value of the button in each of the multiple sampling periods is obtained as the second displacement timing data; Based on the second pressing pressure timing data and the second displacement timing data, the first pressing pressure timing data and the first displacement timing data are obtained.

8. The method according to any one of claims 1 to 6, characterized in that, The step of obtaining the key quality detection result based on the first training result and the second training result according to the set judgment strategy includes: The maximum probability value among the probability values ​​represented by the first training result and the second training result is selected by the set judgment strategy. The maximum probability value is compared with a set threshold, and the detection result is obtained based on the comparison result.

9. A button quality testing device, characterized in that, The device includes: The data acquisition module is used to collect pressure data and displacement data of the button from the sensor when the button is subjected to force and moves. The data acquisition module is used to acquire first pressing pressure timing data and first displacement timing data for quality detection of the button based on the pressing pressure data and displacement data; wherein, the first pressing pressure timing data is data reflecting the change of pressing pressure on the button over time during the detection period, and the first displacement timing data is data reflecting the change of displacement generated by the button over time during the detection period. The data generation module is used to establish a first network model and generate first input data adapted to the first network model based on the first pressing pressure time series data and the first displacement time series data; and to establish a second network model and generate second input data adapted to the second network model based on the first pressing pressure time series data and the first displacement time series data. The training result acquisition module is used to input the first input data and the second input data into the first network model and the second network model respectively for training to obtain the first training result and the second training result; The detection result acquisition module is used to obtain the key quality detection result based on the first training result and the second training result according to the set judgment strategy.

10. A button quality testing device, characterized in that, The device includes a memory and a processor, wherein the memory is used to store a computer program; and the processor is used to execute the computer program to implement the key quality detection method according to any one of claims 1-8.

Images