TWI704804B - Electronic apparatus and automatic advertisement closing method thereof - Google Patents

Abstract

An electronic apparatus and automatic advertisement closing method thereof are provided. The method is adapted to the electronic apparatus including display screen and includes following steps. A previous frame and a current frame displayed by the display screen are captured. Whether the current frame includes an advertisement window is determined according to the previous frame and the current frame. If the current frame includes the advertisement window, a closing object of the advertisement window is detected through a machine learning model. The advertisement window is closed according to the position of the closing object.

Description

Electronic device and its advertisement automatic closing method

The present invention relates to an electronic device, and more particularly to an electronic device and a method for automatically closing advertisements.

With the popularization of personal computers and the rapid development of the Internet, modern people have become very accustomed to using personal computers to handle various tasks, and to browse various information on the Internet through the browser in the personal computer. Based on commercial considerations, most of the current commercial websites provide web pages with advertisements for various products or services related to web content or related to other businesses. Whenever users link to these webpages or at specific times, advertisements may pop up and appear in front of users, thereby achieving the effect of advertising marketing.

However, these advertisements often interrupt the user's operation and distract the user's attention, and make the user feel distressed. Generally speaking, the user needs to click on the display object that can close the advertisement to remove the advertisement from the display screen, which also makes the user feel very troublesome.

In view of this, the present invention provides an electronic device and an advertisement automatic closing method thereof, which can automatically close pop-up advertisements to improve the operating convenience of the electronic device.

The embodiment of the present invention provides a method for automatically closing advertisements, which is suitable for an electronic device including a display screen. The method includes the following steps. Capture the previous and current images displayed on the display screen. Based on the previous screen and the current screen, it is determined whether the current screen includes an advertisement window. If the current screen includes an advertising window, the closed object of the advertising window is detected through the machine learning model. Close the ad window according to the position of the closed object.

An embodiment of the present invention provides an electronic device, which includes a display screen, a memory, and a processor. The processor is coupled to the display screen and the memory, and is configured to perform the following steps. Capture the previous and current images displayed on the display screen. Based on the previous screen and the current screen, it is determined whether the current screen includes an advertisement window. If the current screen includes an advertising window, the closed object of the advertising window is detected through the machine learning model. Close the ad window according to the position of the closed object.

Based on the above, the embodiment of the present invention can determine whether an advertisement window appears based on the previous screen and the current screen displayed on the display screen. When the advertisement window is displayed on the display screen, the embodiment of the present invention can identify the closed object used to close the advertisement window according to the machine learning model. Based on this, in the case that the user does not give a manipulation instruction, the embodiment of the present invention can automatically close the advertisement window based on the location of the closed object. Thereby, the user's actions are not interrupted by the advertisement window, and the fluency and convenience of the user's operation of the electronic device are improved.

In order to make the above-mentioned features and advantages of the present invention more comprehensible, the following specific embodiments are described in detail in conjunction with the accompanying drawings.

Part of the embodiments of the present invention will be described in detail in conjunction with the accompanying drawings. The reference symbols in the following description will be regarded as the same or similar elements when the same symbol appears in different drawings. These embodiments are only a part of the present invention, and do not disclose all the possible implementation modes of the present invention. More precisely, these embodiments are just examples of methods and devices within the scope of the patent application of the present invention.

FIG. 1 is a schematic diagram of an electronic device according to an embodiment of the present invention, but this is only for convenience of description and is not intended to limit the present invention. Please refer to FIG. 1, the electronic device 10 includes a display screen 101, a memory 102, an input device 103, and a processor 104. The electronic device 10 is, for example, a desktop computer, a notebook computer, a smart phone, a tablet computer, a game console, an e-book, or a smart TV, etc. The present invention is not limited thereto.

The display screen 101 is, for example, a Liquid Crystal Display (LCD), a Light-Emitting Diode (LED) display, a Field Emission Display (FED), or an Organic Light-Emitting Diode (Organic Light-Emitting Diode) display. Diode, OLED) or other types of displays, the present invention is not limited to this.

The memory 102 is used to store images, program codes, software modules, and other data. It can be, for example, any type of fixed or removable random access memory (RAM), read-only memory (read -only memory, ROM), flash memory (flash memory), hard disk or other similar devices, integrated circuits and combinations thereof.

The input device 103 is used to receive user operations issued by the user, such as a mouse, a keyboard, a touch device, or a remote control.

The processor 104 is, for example, a central processing unit (Central Processing Unit, CPU), or other programmable general-purpose or special-purpose microprocessor (Microprocessor), digital signal processor (Digital Signal Processor, DSP), programmable Controllers, Application Specific Integrated Circuits (ASIC), Programmable Logic Device (PLD), Graphics Processing Unit (GPU or other similar devices) or a combination of these devices. The processor 104 can execute program codes, software modules, instructions, etc. recorded in the memory 102 to implement the method for automatically closing advertisements in the embodiment of the present invention.

However, in addition to the display screen 101, the memory 102, the input device 103, and the processor 104, the electronic device 10 may also include other components not shown in FIG. 1, such as speakers, microphones, cameras, communication modules, etc. The present invention is not limited to this.

FIG. 2 is a flowchart of a method for automatically closing advertisements according to an embodiment of the present invention. Please refer to FIG. 2, the method of this embodiment is applicable to the electronic device 10 in the above-mentioned embodiment. The detailed steps of the method for automatically closing advertisements in this embodiment are described below in conjunction with various components in the electronic device 10.

In step S201, the processor 104 captures the previous frame and the current frame displayed on the display screen 101. In one embodiment, the processor 104 can obtain the previous screen and the current screen displayed on the display screen 101 through a desktop capture technology such as the "Desktop Duplication API" of the Windows operating system. The display time of the previous screen is earlier than the current screen. In other words, the processor 104 first captures the previous frame and then captures the current frame.

In step S202, the processor 104 determines whether the current picture includes an advertisement window based on the previous picture and the current picture. In the embodiment of the present invention, the advertisement window is a pop-up window. It should be noted that the ad window is often reflected when the user opens a new window, a new web page, or launches an application. Based on this, in one embodiment, the processor 104 determines whether the current screen includes an advertisement window in response to opening a new window, a new web page, or an application.

In addition, the pop-up advertisement window will cover the screen content that the user wants to browse, that is, the current screen formed by covering part of the content of the previous screen. Based on this, in one embodiment, the processor 104 can determine whether the advertisement window appears by comparing the image content of the previous screen with the image content of the current screen. Alternatively, the appearance timing of the pop-up advertisement window is related to the screen content displayed when the user operates the electronic device 10. Specifically, pop-up advertisement windows usually pop up when the user browses the web or executes specific applications (such as game programs, etc.) using the browser, and the pop-up advertisement windows will obscure parts of the previous screen. The content forms the current screen. Based on this, in one embodiment, the processor 104 can use a machine learning model to predict whether the current screen includes an advertising window. The machine learning model is, for example, a time recurrent neural network (RNN) model or long short-term memory ( Long Short Term Memory (LSTM) models and other models that can handle training data about time series. Specifically, the processor 104 may input at least one previous frame and the current frame as an input image sequence to the machine learning model, and the machine learning model will determine whether an advertisement window appears on the current frame. It can be seen that if a machine learning model is to be used to predict whether an advertisement will appear on the current screen, the training of the machine learning model is necessary. When training this machine learning model, it can be trained based on multiple training image sequences and corresponding classification labels. The above-mentioned training image sequence includes at least two screen images, and the classification label can be "including advertising window" and "not including advertising window". Based on this, the machine learning model can classify the input image sequence including the previous image and the current image as including or not including the advertisement window to predict whether the current picture includes the advertisement window.

Next, in step S203, if the processor 104 determines that the current screen includes an advertisement window, the processor 104 detects the closed object of the advertisement window through a machine learning model. In detail, the advertisement window has a closing object for closing the advertisement window. For example, the closed object can be an icon option "X" displayed in the corner of the advertisement window, a text option "close" displayed in the advertisement window, or implemented as other virtual button/icon options. The closed object may be displayed anywhere in the ad window. In response to the user clicking the close object through the input device 103, the advertisement window will be closed.

In one embodiment, the processor 104 may use a machine learning model to identify closed objects in the advertisement window. The above-mentioned machine learning model may be a CNN model, a Yolo model, a model using the NMS algorithm in deep learning, or other machine learning models that can be used for object detection, and the present invention is not limited thereto. Here, the machine learning model used to identify the closed object is trained in advance from many sample advertising window images and the marked sample closed object. In the training process of the machine learning model, a large number of sample advertisement window images must be collected first, and the sample closed object is marked on the sample advertisement window image by marking the sample advertisement window image. After determining a suitable machine learning model, input a large number of sample advertising window images and marked sample closed objects into the machine learning model to gradually train a set of rules that can be used to predict closed objects (that is, the parameters of the machine learning model) ) To finally establish a machine learning model that can be used to detect closed objects. In one embodiment, the machine learning algorithm can automatically extract the features used to detect the closed object to create a machine learning model for detecting the closed object. In this way, the processor 104 can use the machine learning model to predict the location of the closed object based on the advertisement window image.

After recognizing the closed object in the advertisement window, in step S204, the processor 104 closes the advertisement window according to the position of the closed object. Specifically, the processor 104 can automatically generate a click signal associated with the closed object to the operating system of the electronic device 10 according to the position of the closed object. Based on this, the operating system of the electronic device 10 can determine that it receives a click signal for clicking the object to be closed, and then closes the advertisement window.

FIG. 3 is a schematic diagram of detecting an advertisement window according to an embodiment of the invention. Please refer to FIG. 3, the previous screen P1 is a display screen of a user browsing a webpage using a browser. Then, a pop-up advertisement window W1 appears on the current screen C1. In one embodiment, the processor 104 can compare the previous frame P1 with the current frame C1 and recognize that the current frame C1 includes an image block different from the previous frame P1, and the image block Z1 can be determined as an advertisement window. Then, the processor 104 can use the machine learning model to detect the closing object Obj1 in the image block Z1, and generate a click signal for clicking the closing object Obj1, so that the advertisement window W1 is closed.

However, the machine learning model used to detect closed objects may be misjudged, such as failing to detect closed objects, etc. At this time, the user still manually closes the advertisement window. Therefore, in one embodiment, the user behavior of the user closing the advertisement window can be used as a basis for updating the machine learning model. Examples will be listed below to illustrate.

FIG. 4 is a flowchart of a method for automatically closing advertisements according to an embodiment of the present invention. Please refer to FIG. 4, the method of this embodiment is applicable to the electronic device 10 in the above-mentioned embodiment. The detailed steps of the method for automatically closing advertisements in this embodiment are described below in conjunction with various components in the electronic device 10.

First, in step S401, the processor 104 captures the previous frame and the current frame displayed on the display screen 101. In step S402, the processor 104 compares the previous frame with the current frame to determine whether the current frame includes an image block different from the previous frame. In this embodiment, step S402 can be implemented as sub-steps S4021 to S4022. Specifically, in step S4021, based on the previous screen and the current screen, the processor 104 determines whether the current screen includes an advertisement window and compares the previous screen with the current screen to determine whether the current screen includes a first image block that is different from the previous screen. If the current frame includes a first image block that is different from the previous frame, in step S4022, the processor 104 performs image feature identification on the first image block to determine whether the first image block includes advertising content. Specifically, the processor 104 can compare the previous screen with the current screen according to the image subtraction method, and the first image block in the current screen that is different from the previous screen may be an advertisement window or another application opened by the user. Windows. Therefore, in this embodiment, the processor 104 can perform text recognition or image feature recognition on the first image block to determine whether the first image block contains a specific keyword or a specific image related to the advertisement content. For example, specific keywords are, for example, "advertisement", "special price", "discount", "limited period" and so on. The specific signs are, for example, brand trademarks or store trademarks.

If the current screen includes a first image block different from the previous screen and the first image block includes advertising content, in step S403, the processor 104 detects the closed object of the advertising window through the machine learning model. This step is similar to step S203, and will not be repeated here. In step S404, the processor 104 closes the advertisement window according to the position of the closed object. In this embodiment, step S404 can be implemented as sub-steps S4041 to S4042. In detail, in step S4041, the processor 104 generates a click signal for closing the object according to the position of the closing object. In step S4042, in response to the click signal associated with the closed object, the processor 104 closes the advertisement window.

Next, in step S405, the processor 104 determines whether the advertisement window is closed. Specifically, the processor 104 may compare the current frame with the next frame, such as performing image subtraction on the current frame and the next frame, to determine whether the advertisement window is closed. Generally speaking, if the pop-up advertisement window is not successfully closed, the current screen and the next screen are roughly the same, and the two screens do not have obvious different blocks.

It is worth mentioning that if the advertisement window is not successfully closed, it means that the machine learning model cannot successfully detect the closed object. Therefore, users generally close the advertisement window manually, that is, click on the advertisement window through the input device 103 Close the object. Therefore, if the advertisement window is not closed, in step S406, the processor 104 detects the user behavior input through the input device 103. For example, the processor 104 can detect mouse input events, keyboard input events, touch input events, and so on. In step S407, the processor 104 obtains the click position of the user's behavior on the display screen 101. This click position is where the object is closed. Based on this, in step S408, the processor 104 captures the image content around the selected location to obtain the second image block. This second image block must cover part or all of the closed object. In step S409, the processor 104 updates the machine learning model according to the second image block associated with the user behavior in the advertisement window, thereby improving the accuracy of the machine learning model. For example, FIG. 5 is a schematic diagram illustrating an example of obtaining an image block associated with a closed object according to an embodiment of the invention. Please refer to Figure 5, if the advertisement window W2 cannot be closed through the machine learning model, the user will close the object Obj2 by clicking the mouse. The processor 104 responds to detecting the click signal of the mouse, and captures the image block Z2 around the click position L1, and updates the parameters of the machine learning model according to the image block Z2 and the advertising window W2. In this way, the user behavior of closing the advertisement window can be used to continue training the machine learning model, and the accuracy of detecting the closed object can be improved.

In summary, in the embodiment of the present invention, it can be determined whether an advertisement window appears based on the previous screen and the current screen displayed on the display screen. When the advertisement window is displayed on the display screen, the closing object used to close the advertisement window can be identified according to the machine learning model. In the embodiment of the present invention, the purpose of automatically closing the advertisement window can be achieved only based on the screen image data. In the case that the user does not give a manipulation instruction, the embodiment of the present invention can automatically close the advertisement window based on the location of the closed object. In this way, the user's actions are not interrupted by the advertisement window, and the fluency and convenience of the user's operation of the electronic device are improved. In addition, the embodiment of the present invention can update the machine learning model according to the user behavior of the user closing the advertisement window, so as to improve the accuracy of the machine learning model.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention. Anyone with ordinary knowledge in the technical field can make some changes and modifications without departing from the spirit and scope of the present invention. The scope of protection of the present invention shall be determined by the scope of the attached patent application.

10: Electronic device

101: display screen

102: memory

103: input device

104: processor

W1, W2: Advertising window

Obj1, Obj2: close objects

L1: Click location

Z1, Z2: image block

P1: Previous screen

C1: Current screen

S201~S204, S401~S409, S4021~S4022, S4041~S4042: steps

FIG. 1 is a schematic diagram of an electronic device according to an embodiment of the invention. FIG. 2 is a flowchart of a method for automatically closing advertisements according to an embodiment of the present invention. FIG. 3 is a schematic diagram of detecting an advertisement window according to an embodiment of the invention. FIG. 4 is a flowchart of a method for automatically closing advertisements according to an embodiment of the present invention. FIG. 5 is a schematic diagram illustrating an example of obtaining an image block associated with a closed object according to an embodiment of the present invention.

S201~S204: steps

Claims (12)

A method for automatically closing advertisements is suitable for an electronic device that includes a display screen. The method includes: capturing a previous screen and a current screen displayed on the display screen; judging the current screen according to the previous screen and the current screen Whether an advertisement window is included; if it is determined that the current screen includes the advertisement window, a machine learning model is used to detect a closed object of the advertisement window; and the advertisement window is closed according to the position of the closed object. For example, in the method for automatically closing advertisements as described in item 1 of the scope of patent application, the step of judging whether the current picture includes the advertisement window based on the previous picture and the current picture includes: comparing the previous picture with the current picture to determine the Whether the current frame includes a first image block different from the previous frame. For example, the method for automatically closing advertisements according to the second item of the scope of patent application, wherein the step of judging whether the current screen includes the advertisement window based on the previous screen and the current screen further includes: if the current screen includes a screen different from the previous screen Performing image feature identification on the first image block to determine whether the first image block includes advertising content. For example, the method for automatically closing advertisements as described in item 1 of the scope of patent application, wherein the method further includes: determining whether the advertisement window is closed; if the advertisement window is not closed, detecting a user behavior; The second image block associated with the user behavior updates the machine learning model. According to the method for automatically closing advertisements according to item 4 of the scope of patent application, wherein the step of updating the machine learning model according to the second image block associated with the user behavior in the advertisement window includes: obtaining the user behavior points Select the click position of the display screen; and capture the image content around the click position to obtain the second image block. For example, in the method for automatically closing advertisements described in claim 1, wherein the step of closing the advertisement window according to the position of the closing object includes: generating a click signal of the closing object according to the position of the closing object; and responding to associating the closed object Close the click signal of the object and close the advertisement window. An electronic device, including: a display screen; a memory: and The processor is coupled to the input device, the memory and the display screen, and is configured to: capture a previous frame and a current frame displayed on the display screen; determine the current frame based on the previous frame and the current frame Whether the screen includes an advertisement window; if the current screen includes the advertisement window, detect the closed object of the advertisement window through a machine learning model; and close the advertisement window according to the position of the closed object. The electronic device according to claim 7, wherein the processor is configured to: compare the previous frame with the current frame to determine whether the current frame includes a first image block that is different from the previous frame . The electronic device according to claim 8, wherein the processor is configured to: if the current frame includes the first image block that is different from the previous frame, perform image characteristics on the first image block Identify and determine whether the first image block includes advertising content. For example, the electronic device described in item 7 of the scope of patent application, wherein the electronic device further includes an input device, and the processor is further configured to: determine whether the advertisement window is closed; if the advertisement window is not closed, detect input through the input device A user of Behavior; and update the machine learning model according to the second image block associated with the user behavior in the advertisement window. For example, the electronic device described in claim 10, wherein the processor is configured to: obtain the click position of the user behavior clicking the display screen; and capture the image content around the click position to obtain the The second image block. For example, the electronic device described in claim 7, wherein the processor is configured to: generate a click signal of the closed object according to the position of the closed object; and turn off in response to the click signal associated with the closed object The ad window.

Images