TW202105131A - Method, computer program product, and electronic device for setting function of a button of a mobile device - Google Patents

Abstract

A method for setting function of a button of a mobile device, including capturing a screen shot from a mobile device; performing a situation recognition operation on the screen shot to determine whether the screen shot is in a situation where the upper and lower page function can be used; and when the screen shot is in the situation where the upper and lower page functions can be use, setting a physical volume button of the mobile device as a function key that can adjust the upper and lower pages. The situation recognition operation is performed by the mobile device according to a training model, which is established by using a plurality of screen shots as training data and is operated to determine whether a target screen shot is in the situation where the upper and lower page functions can be used.

Description

Method for setting button function of mobile device, computer program product and electronic device

The present invention relates to a method for setting button functions, in particular to a method for setting button functions of a mobile device, a computer program product and an electronic device.

In the context of using a mobile device, such as in a browser window or in a conversation window of an instant messaging software, when a user is surfing the Internet or performing a text conversation with others, he often needs to click the action of the next page . As the screen of the mobile device becomes larger and larger, if the user is in a situation where it is not easy to use both hands at this time, it is difficult to perform the action of clicking the next page.

In order to solve the above-mentioned problems, the present invention proposes a method for setting button functions of mobile devices, computer program products, and electronic devices. Artificial Intelligence (AI) is used to automatically identify whether the user is using the context of the upper and lower pages. The volume key of the mobile device is set to perform the function of switching the upper and lower pages.

A method for setting key functions of a mobile device according to an embodiment of the present invention includes: capturing a screen displayed on the screen of the mobile device; performing a context recognition operation on the screen to determine whether the screen is in the context of using the next page function ; When it is determined that the screen is in the context of using the page up and down function, perform the function key setting to set the physical volume key of the mobile device as a function key that can adjust the up and down pages. The context recognition operation is executed by the mobile device according to a training model, which is established by using plural screenshots as training data, and is operated to determine whether a target screen is in the context of the upper and lower page function.

Such as the button function setting method mentioned above, wherein the establishment of the training model includes: collecting the screenshots; detecting objects on the screenshots; classifying the screenshots according to whether they match specific objects A context group with upper and lower pages and a context group with no upper and lower pages; and based on whether the target screen has the specific object included in the context group with upper and lower pages, context identification is performed on the target screen.

Such as the above-mentioned button function setting method, wherein, when the target screen has the specific object contained in the context group with upper and lower pages, it is determined that the target screen is in the context of using the function of the upper and lower pages; or as the target screen If there is no specific object included in the context group with the upper and lower pages, it is determined that the target screen is not in the context of using the upper and lower pages function.

Such as the button function setting method mentioned above, in which, when the target screen is in the context of using the upper and lower page functions, the physical volume keys of the mobile device are set as the function keys that can adjust the upper and lower pages; or when the target screen is not in use In the context of the page up and down function, the physical volume key of the mobile device maintains its original function.

According to an embodiment of the present invention, a computer program product is used to set the key function of a mobile device, and the computer program product is loaded by the mobile device to execute a retrieval command, a detection command, and a setting command. The capture command enables a processor of the mobile device to capture a frame displayed on the screen of the mobile device. The detection instruction enables the processor to perform a context recognition operation on the screen to determine whether the screen is in the context of using the page up and down function. When it is determined that the screen is in the context of using the page up and down function, the processor is made to perform function key setting, so as to set the physical volume key of the mobile device as a function key that can adjust the page up and down. The context recognition operation is executed by the mobile device according to a training model, which is established by using plural screenshots as training data, and is operated to determine whether a target screen is in the context of the upper and lower page function.

For the computer program product mentioned above, the establishment of the training model includes: a screenshot collection command, an object detection command, a context classification command, and a context recognition command. The screenshot collection command is used to collect these screenshots. The object detection command is used to perform object detection on the screenshots. The context classification command is used to classify the screen shots into a context group with upper and lower pages and a context group without upper and lower pages according to whether there is a specific object. The context recognition command is used to perform context recognition on the target screen according to whether the target screen has the specific object included in the context group with upper and lower pages.

Such as the above-mentioned computer program product, in which, when the target screen has the specific object contained in the context group with upper and lower pages, it is determined that the target screen is in the context of using the function of the upper and lower pages; or when the target screen is in the context of using the upper and lower pages function; If it does not have the specific object included in the context group with the upper and lower pages, it is determined that the target screen is not in the context of using the upper and lower pages function.

Such as the above-mentioned computer program product, in which, when the target screen is in the context of using the upper and lower page function, the physical volume key of the mobile device is set as the function key that can adjust the upper and lower pages; or when the target screen is not in the context of using the upper and lower pages. In the context of the page function, the physical volume key of the mobile device maintains its original function.

An electronic device according to an embodiment of the invention includes a mobile device. The mobile device includes a display screen, a physical volume button, a processor, and a training model. The display screen is used to display a picture. The training model is stored in a memory of the mobile device, so that the processor can perform operations including: capturing the screen displayed on the display screen; performing a context recognition operation on the screen to determine whether the screen is in use. The context of the function; when it is determined that the screen is in the context of using the upper and lower page functions, the function key setting is performed to set the physical volume key of the mobile device as a function key that can adjust the upper and lower pages. The context recognition operation is executed by the mobile device according to the operation of the training model. The training model is established using plural screenshots as training data, and the operation is used to determine whether a target screen is in the context of the upper and lower page function.

Such as the above-mentioned electronic device, wherein the establishment of the training model includes: collecting the screenshots; performing object detection on the screenshots; classifying the screenshots into one according to whether there is a specific object An upper and lower page context group and a no upper and lower page context group; and based on whether the target screen has the specific object included in the upper and lower page context group, context identification is performed on the target screen.

Such as the above electronic device, wherein, when the target screen has the specific object contained in the context group with up and down pages, it is determined that the target screen is in the context of using the up and down page function; or when the target screen is not If there is the specific object included in the context group with the upper and lower pages, it is determined that the target screen is not in the context of using the upper and lower pages function.

For the above-mentioned electronic device, when the target screen is in the context of using the upper and lower page function, the physical volume key of the mobile device is set as a function key that can adjust the upper and lower pages; or when the target screen is not in the context of using the upper and lower pages In the context of the function, the physical volume key of the mobile device maintains the original function.

The button function setting method of the mobile device of the present invention is executed by a mobile device and a plurality of program codes running on a processor of the mobile device. These plural codes, such as a training model, can perform different functions, such as screen capture function, object detection function, image recognition function, or image classification function. For example, the training model can be a background program running on the mobile device to capture the screen at each time point in the mobile device. With the development of artificial intelligence and big data analysis technology, object detection technology has been applied in many fields, such as face detection, face recognition, license plate recognition, human figure and vehicle recognition, etc. Object detection technology allows a computer to analyze an object in a picture or a segment of image stream and mark the object. The object detection technology needs to input a large amount of object data to a neural network as training (learning) data of the neural network, and the neural network can recognize the object.

FIG. 1 is an application flowchart of a method for setting button functions of a mobile device according to an embodiment of the present invention. As shown in Figure 1, the button function setting method of the mobile device of the present invention captures a screen displayed on the screen of the mobile device (step S100).

Next, the button function setting method of the present invention performs a context recognition operation on the screen to determine whether the screen is in the context of using the upper and lower page function (step S102). In this embodiment, the context recognition operation is executed by the mobile device according to a training model. The training model is created using multiple screenshots as training data, and the operation is used to determine whether a target screen is in the upper and lower positions. The context of the page function. For example, the button function setting method of the present invention can use an external computer to collect a large number of mobile phone screenshots on the Internet to obtain the plural screenshots, and the mobile phone screenshots cover all usage scenarios, such as mobile phone desktop, settings The screen or the use screen of any application (application: APP), and then the captured plural screen shots are imported into the mobile device for the training model to learn. In another embodiment, a large number of mobile phone screenshots collected on the Internet can also be stored in a web server, so that the training model can read the stored in the web server via the Internet Wait for multiple screenshots for the training model to learn.

FIG. 2A is a schematic diagram of two screens with the context of using the upper and lower pages according to an embodiment of the present invention. As shown in Figure 2A, the training model of the mobile device determines through object detection that the screen 100 has a specific object 104 (for example, left and right arrows), and the screen 102 has another specific object 106 (for example, larger or smaller than symbol). FIG. 2B is a schematic diagram of two screens that do not have the context of using the upper and lower pages according to an embodiment of the present invention. As shown in FIG. 2B, the training model of the mobile device determines that the screen 108 and the screen 110 do not have the specific object. In other words, as long as there are more samples of the screenshots provided to the training model, that is, the more samples are used as training data, the training model is used to determine whether there are one or more specific images in the target screen. The ability of an object to determine whether it is in the context of using the next page function is stronger.

When the training model of the mobile device determines that the target screen is in the context of using the page up and down function, such as screen 100 and screen 102 in Figure 2A, step S104 in Figure 1 is entered. The target screen is the screen currently displayed on the mobile device when the user uses the mobile device. In step S104, the processor of the mobile device sets the physical volume key of the mobile device as a function key that can adjust the upper and lower pages. When the training model determines that the screen is not in the context of using the page up and down function, for example, the screen 108 and the screen 110 in Figure 2B, step S106 is entered. In step S106, the processor of the mobile device maintains the original function of the physical volume key of the mobile device.

For example, when a user needs to surf the Internet, and the screen displayed on the mobile device is switched from the original wallpaper screen (such as the screen 108 in Figure 2B) to a browser screen (such as the screen 100 in Figure 2A), At this time, the training model operating on the mobile device captured screen 100, and used a large number of historical screenshots previously collected as a sample, and detected that screen 100 has a specific context associated with the use of the next page function物件104. After that, the training model instructs the processor of the mobile device to have the context of the upper and lower page functions on the screen 100, and sets the physical volume key of the mobile device as a function key for adjusting the upper and lower pages. At this time, the user can adjust the upper and lower pages of the screen 100 through the physical volume keys of the mobile device. When the user switches the screen displayed on the mobile device from screen 100 back to screen 108, the training model determines that the specific object does not exist on the screen 108, that is, the situation where there is no page up and down function, the training model instructs the mobile device The processor, in the context of the screen 108, switches the physical volume key of the mobile device back to the original volume adjustment function.

Generally speaking, the hardware computing performance of computers (including desktop computers, notebook computers, or servers) is higher than that of mobile devices (such as mobile phones or PDAs). For example, the computing speed of a computer's processor is often higher than that of a mobile device's processor. Therefore, in this embodiment, the button function setting method of the present invention can establish the training model on another computer or a different platform of the mobile device. The button function setting method of the present invention can convert the training model originally running on the computer into the training model running on the mobile device, and the training model running on the mobile device and the training model running on the computer are Share the same training data sample (that is, a large number of captured historical screenshots).

FIG. 3 is a flowchart of establishing a training model of the method for setting button functions of a mobile device according to an embodiment of the present invention. As shown in FIG. 3, the method for setting the button function of the present invention builds the training model from step S300. In step S300, the training model collects multiple screenshots from the Internet or a storage. In this embodiment, the user can instruct the training model to set a specific object to be recognized through the computer or the mobile device. For example, the user can set "←", "→", "↑", "↓", "<", "＞", "", "", "Previous", "Next", "BACK", or "NEXT", etc. represent symbols or texts used to indicate that the upper and lower (front and back) page functions can be used, and are set to specific objects that the training model needs to recognize, but it is not limited to this. Then, in step S302, the training model performs object detection on the plurality of screenshots. For example, the training model started to perform the above-mentioned symbols "←", "→", "↑", "↓", "<", ">", "", or """ in these plural screenshots. Detection.

In step S304, the training model categorizes the plural screen shots into a context group with upper and lower pages and a context group without upper and lower pages according to whether they conform to a specific object. When there are specific objects in the screenshots, for example, the above symbols "←", "→", "↑", "↓", "<", ">", "", or """, then In step S306, the training model performs context recognition on a target screen according to whether the target screen has the specific object included in the context group with upper and lower pages. When the target screen has the specific object included in the context group with upper and lower pages, context recognition is performed on the target screen. When the screenshots do not match the specific object, step S308 is entered, and the training model classifies the screenshots into the context group without upper and lower pages. Take the screens 100 and 102 in Figure 2A and the screens 108 and 110 in Figure 2B as examples. Since the training model determines that there is a specific object 104 in the screen 100 and a specific object 106 in the screen 102, the training model will 100 and the screen 102 are classified into the context group with upper and lower pages. Conversely, since the training model determines that the pictures 108 and 110 do not have the specific object, the training model will classify the pictures 108 and 110 into the context group with no upper and lower pages.

After the training model completes the classification of a large number of historical screenshots, when the training model is run by, for example, a processor, it can determine whether the captured target image has the context of using the upper and lower page function. In order to optimize the execution performance on the mobile device, the key function setting method of the present invention originally runs the training model on the computer into the training model run on the mobile device, so that the training model can be based on a large number of historical screens collected Take a screenshot to determine whether the screen of the mobile device has a context where the upper and lower page functions can be used. In this embodiment, the training model running on the computer is written using Tensor Flow. The button function setting method of the present invention converts the training model running on the computer into the training model written by Tensor Flow Lite, and cooperates with the application programming interface (API) of Tensor Flow Lite to make the The training model can be executed in the environment of Android or iOS mobile phones and Chromebooks.

In this embodiment, the training model has completed training (learning) after performing steps S300 to S308, and can return to step S100 in Figure 1 to use the training model to start capturing the current screen of the mobile device. In more detail, when the current screen of the mobile device has the specific object contained in the context group with up and down pages, the training model determines that the current screen is in the context of using the up and down page function, and the mobile device The physical volume key of is set as a function key that can adjust the upper and lower pages. Conversely, when the current screen of the mobile device does not have the specific object contained in the context group with up and down pages, the training model determines that the current screen is not in the context of using the up and down page function, and the entity of the mobile device The volume keys maintain their original functions.

For example, referring to Figures 2A and 2B, suppose the user sets the "←", "→", "↑", "↓", "<", ">", "", and """ symbols as The specific object that the training model needs to identify, because the specific object "<" set above is the same as the specific object 106 of the screen 102, the training model classifies the screen 102 into the context group with the upper and lower pages, and displays it on the screen 102 In the context of, set the physical volume key of the mobile device as a function key that can adjust the upper and lower pages. Conversely, since no objects set above appear in the screen 100, the training model classifies the screen 100 into the context group with no upper and lower pages, and maintains the physical volume keys of the mobile device in the context of the screen 100. There are functions.

Although the screen 100 is actually in the context of using the page up and down function, the training model cannot identify the specific object 104. In order for the training model to recognize the specific object 104 in the screen 100, the screen 100 must be added to the training data (that is, the screenshots), and the steps S300~S308 in Figure 3 should be executed again, and the The symbols "⇦" and "⇨" are set to the specific objects that the training model needs to recognize. After re-training, the training model can recognize the specific object 104 in the screen 100, so that the screen 100 can be classified into the context group with upper and lower pages, and in the context of the screen 100, the action The physical volume keys of the device are set as function keys that can adjust the upper and lower pages.

The button function setting method of the present invention executes steps S100~S106 in Figure 1 to determine whether to set the physical volume button of the mobile device as a function button that can adjust the upper and lower pages, and then the user verifies the training model in step S102 Is the judgment correct? If the judgment of the training model is correct, the user can continue to use the training model to assist the user in setting the physical volume keys of the mobile device to be able to adjust the upper and lower pages in the context of the function of adjusting the upper and lower pages. Function keys. If the judgment of the training model is incorrect, the user can adjust the parameters of the training model or add new historical screen data to the training data to increase the number of samples of the training data, thereby achieving improvement of the training The accuracy of the model.

For example, generally speaking, the specific objects used to switch between the upper and lower pages of the screen of the mobile device are distributed on the upper and lower sides or edge corners of the screen. Therefore, the training model will first search for the screen when detecting the screen. The upper and lower sides or edge corners. The user can expand the search range of the training model in the screen by adjusting the parameters. For example, the user needs to search the center of the screen to improve the accuracy of the training model.

The present invention further discloses a computer program product for setting a key function of a mobile device. The computer program product loads the program through the mobile device to execute an acquisition command, a detection command, and a setting command. The retrieval instruction can cause a training model of a processor running on the mobile device to execute step S100 in FIG. 1. The detection instruction can cause the training model to perform step S102 in FIG. 1. The setting instruction can cause the processor of the mobile device to execute step S104 or step S106 in FIG. 1. The context recognition operation is executed by the mobile device according to a training model, which is established by using plural screenshots as training data, and is operated to determine whether a target screen is in the context of the upper and lower page function.

In this embodiment, the training model can also be run on a processor on an external computer to perform functions such as screen capture, object detection, and image classification. The training model execution includes a screenshot collection command, a context classification command, and a context judgment command. The screenshot collection command can cause a processor of the external computer or the mobile device to execute step S300 in FIG. 3. The object detection command can cause the processor of the external computer or the mobile device to execute step S302 in FIG. 3. The context classification command can cause the processor of the external computer or the mobile device to execute step S304 in FIG. 3. The context determination instruction can cause the processor of the external computer or the mobile device to execute step S306 or step S308 in FIG. 3.

The steps or actions performed by the computer program product of the present invention can directly correspond to the button function setting method for mobile devices disclosed in the present invention. The button function setting method for mobile devices disclosed in the present invention has been described in the present specification. The paragraphs [0017]~[0031] are described in detail, so they will not be repeated.

The invention further discloses an electronic device. FIG. 4 is a schematic diagram of an electronic device 400 according to an embodiment of the invention. As shown in Figure 4, the electronic device 400 includes a mobile device 402, a computer 412, and the mobile device 402 and the computer 412 can be connected via an Internet 420 or a connection line 422 (such as a USB cable or a Lightning cable, but not Limited to this) transmit data to each other. The mobile device 402 includes a display 404, a physical volume key 405, a processor 408, and a storage 411. The display screen 404 is used to display the screen 406. A training model 410 is stored in the memory 411 of the mobile device 402, so that the processor 408 executes the steps including: capturing the screen 406 displayed on the display screen 404 of the mobile device 402 (corresponding to step S100 in Figure 1); Context recognition operation to determine whether the screen 406 is in the context of using the page up and down function (corresponding to step S102 in Figure 1); when the screen 406 is in the context of using the page up and down function, set the physical volume key 405 as a function key that can adjust the page up and down (Corresponding to step S104 in Figure 1); when the screen 406 is not in the context of using the page up and down function, the physical volume key 405 of the mobile device 402 maintains its original function (corresponding to step S106 in Figure 1). The context recognition operation is executed by the mobile device according to the operation of the training model 410. The training model 410 is created by using multiple screenshots as training data and is operated to determine whether a target screen is in the context of the upper and lower page function.

The computer 412 includes a processor 414 and a storage 418. A training model 416 is stored in the storage 418 of the computer 412, so that the processor 414 can execute the following steps: collecting the plurality of screenshots from the Internet 420 or the storage 418 (corresponding to step S300 in Figure 3); Take screenshots for object detection (corresponding to step S302 in Fig. 3); these screenshots are classified into a context group with upper and lower pages and a context group without upper and lower pages according to whether they match specific objects (corresponding to step S302 in Fig. 3) S304); when the screenshots match a specific object and are classified into a context group with upper and lower pages, the target screen is contextualized according to whether a target screen has the specific object contained in the context group with upper and lower pages Recognition (corresponding to step S306 in Figure 3); when the screen sections do not match the specific object on the way, classify the screenshots into a context group with no upper and lower pages (corresponding to step S308 in Figure 3). The electronic device 400 of the present invention can convert the training model 416 suitable for the computer 412 into the training model 410 suitable for the mobile device 402, and transmit the converted training model 410 to the mobile device through the connection line 422 or the Internet 420 402, so that the training model 410 can perform context recognition on the screen 406 according to the screenshots.

The steps or actions performed by the components in the electronic device 400 of the present invention can directly correspond to the button function setting method for mobile devices disclosed in the present invention. The button function setting method for mobile devices disclosed in the present invention has been described in Paragraphs [0017]~[0031] of the specification of this case are described in detail, so I will not repeat them. The method, computer program product, and electronic device for the key function setting of the mobile device disclosed in the present invention allow the user to use the physical volume keys of the mobile device to perform the action of clicking on the upper and lower pages, so that the user can more easily operate the mobile device. For convenience.

Although the embodiments of the present invention are as described above, we should understand that what is presented above is only an example, not a limitation. According to this embodiment, many changes of the above exemplary embodiment can be implemented without violating the spirit and scope of the invention. Therefore, the breadth and scope of the present invention should not be limited by the embodiments described above. More precisely, the scope of the present invention should be defined by the following patented scope and its equivalents.

S100, S102, S104, S106: steps 100, 102, 108, 110: screen 104, 106: specific objects S300, S302, S304, S306, S308: steps 400: Electronic device 402: mobile device 404: Display 405: One or more physical volume keys 406: Screen 408: processor 410: training model 411: Storage 412: Computer 414: processor 416: training model 418: Storage 420: Internet 422: connection line

FIG. 1 is an application flowchart of a method for setting button functions of a mobile device according to an embodiment of the present invention. FIG. 2A is a schematic diagram of two screens with the context of using the upper and lower pages according to an embodiment of the present invention. FIG. 2B is a schematic diagram of two screens that do not have the context of using the upper and lower pages according to an embodiment of the present invention. FIG. 3 is a flowchart of establishing a training model of the method for setting button functions of a mobile device according to an embodiment of the present invention. FIG. 4 is a schematic diagram of an electronic device according to an embodiment of the invention.

S100, S102, S104, S106: steps

Claims (18)

A method for setting button functions of a mobile device includes: Capture a screen displayed on the screen of the mobile device; Perform context recognition operations on the screen to determine whether the screen is in the context of using the page up and down function; When it is determined that the screen is in the context of using the page up and down function, execute the function key setting to set the physical volume key of the mobile device as a function key that can adjust the up and down pages; Wherein, the context recognition operation is executed by the mobile device according to a training model. The training model is created by using multiple screenshots as training data, and the operation is used to determine whether a target screen is in the context of the next page function . The button function setting method described in item 1 of the scope of patent application, wherein the establishment of the training model includes: Collect screenshots of such screens; Perform object detection on these screenshots; Classify the screenshots into a context group with upper and lower pages and a context group without upper and lower pages according to whether there are any objects that match a specific object; and According to whether the target screen has the specific object included in the context group with upper and lower pages, context recognition is performed on the target screen. For example, the button function setting method described in item 2 of the scope of patent application, wherein, when the target screen has the specific object contained in the context group with up and down pages, it is determined that the target screen is using the up and down page function Situation. For example, the button function setting method described in item 2 of the scope of patent application, wherein, when the target screen does not have the specific object contained in the context group with up and down pages, it is determined that the target screen is not using the up and down page function Situation. For example, the button function setting method described in item 3 of the scope of patent application, wherein when the target screen is in the context of using the page up and down function, the physical volume key of the mobile device is set as a function key that can adjust the up and down page. For example, the button function setting method described in item 4 of the scope of patent application, wherein when the target screen is not in the context of using the upper and lower page function, the physical volume key of the mobile device maintains its original function. A computer program product used to set the key function of a mobile device. The computer program product is loaded by the mobile device to execute: A capture command to enable a processor of the mobile device to capture a screen displayed on the screen of the mobile device; A detection instruction to enable the processor to perform a context recognition operation on the screen to determine whether the screen is in a context where the next page function is used; A setting instruction, when it is determined that the screen is in the context of using the page up and down function, make the processor execute the function key setting, so as to set the physical volume key of the mobile device as a function key that can adjust the up and down pages; Wherein, the context recognition operation is executed by the mobile device according to a training model. The training model is created by using multiple screenshots as training data, and the operation is used to determine whether a target screen is in the context of the next page function . For example, the computer program product described in item 7 of the scope of patent application, wherein the establishment of the training model includes: A screenshot collection command to collect screenshots of such screens; An object detection command for object detection on the screenshots; A context classification command for classifying the screen shots into a context group with upper and lower pages and a context group without upper and lower pages according to whether there are specific objects that match; and A context recognition command is used to perform context recognition on the target screen according to whether the target screen has the specific object included in the context group with upper and lower pages. For example, the computer program product described in item 8 of the scope of patent application, in which, when the target screen has the specific object contained in the context group with up and down pages, it is determined that the target screen is in the context of using the up and down page function . For example, the computer program product described in item 8 of the scope of patent application, wherein, when the target screen does not have the specific object contained in the context group with up and down pages, it is determined that the target screen is not in the use of the up and down page function Situation. For example, in the computer program product described in item 9 of the scope of patent application, when the target screen is in the context of using the upper and lower page functions, the physical volume key of the mobile device is set as a function key that can adjust the upper and lower pages. For example, in the computer program product described in item 10 of the scope of patent application, when the target screen is not in the context of using the upper and lower page function, the physical volume key of the mobile device maintains its original function. An electronic device, including: A mobile device, including: A display screen to display a screen; A physical volume key; A processor; and A training model, stored in a memory of the mobile device, so that the processor can execute: Capture the screen displayed on the display; Perform context recognition operations on the screen to determine whether the screen is in the context of using the page up and down function; When it is determined that the screen is in the context of using the page up and down function, execute the function key setting to set the physical volume key of the mobile device as a function key that can adjust the up and down pages; Wherein, the context recognition operation is executed by the mobile device according to the operation of the training model, and the training model is created by using multiple screenshots as training data, and the operation is used to determine whether a target screen is in the context of the next page function . The electronic device described in item 13 of the scope of patent application, wherein the establishment of the training model includes: Collect screenshots of such screens; Perform object detection on these screenshots; Classify the screenshots into a context group with upper and lower pages and a context group without upper and lower pages according to whether there are any objects that match a specific object; and According to whether the target screen has the specific object included in the context group with upper and lower pages, context recognition is performed on the target screen. For example, the electronic device described in item 14 of the scope of patent application, wherein, when the target screen has the specific object included in the context group with up and down pages, it is determined that the target screen is in the context of using the up and down page function. For example, the electronic device described in item 14 of the scope of patent application, wherein, when the target screen does not have the specific object included in the context group with up and down pages, it is determined that the target screen is not in the context of using the up and down page function . For example, in the electronic device described in item 15 of the scope of patent application, when the target screen is in the context of using the page up and down function, the physical volume key of the mobile device is set as a function key that can adjust the up and down pages. For example, in the electronic device described in item 16 of the scope of patent application, when the target screen is not in the context of using the upper and lower page function, the physical volume key of the mobile device maintains its original function.

Images