Shooting game assistance method

Abstract

The application discloses a shooting game auxiliary method, and relates to the technical field of game auxiliary, and the method comprises the following steps: recording a screen and intercepting game images at a preset frequency; cropping the game images to obtain to-be-recognized images containing small map regions; inputting the to-be-recognized images into a trained image recognition model to recognize target object information and small map identification positions; cropping the to-be-recognized images according to the small map identification positions to obtain small map images and calculating positions of the target objects in the small map images; and generating voice prompts according to the positions of the target objects in the small map images. The application improves the accuracy of target object recognition.

Description

Technical Field

[0001] This application relates to the field of game assistance technology, and in particular to a method for assisting shooting games. Background Technology

[0002] In shooting games, a minimap is usually placed in a specific area to help players observe certain game information, such as gunshots represented by gun icons or footsteps. However, novice players often find it difficult to effectively utilize the minimap information, leading to a quick game ending and preventing them from fully experiencing the game's enjoyment.

[0003] In related technologies, there are already methods based on image recognition technology to identify small maps and provide assistance to game players. However, due to the influence of the input image of the model, the recognition results are distorted, resulting in poor assistance effect. Summary of the Invention

[0004] The main purpose of this application is to provide an auxiliary method for shooting games, which aims to solve the technical problem of how to improve the accuracy of target object recognition.

[0005] To achieve the above objectives, this application proposes a method for assisting shooting games, comprising: Record the screen and capture game images at a preset frequency; The game image is cropped to obtain an image to be identified that includes the minimap area; The image to be identified is input into a trained image recognition model to identify target object information and small map location markers. The target object information includes the type of the target object and its location in the image to be identified. Based on the location marked on the minimap, the image to be identified is cropped to obtain a minimap image, and the position of the target object in the minimap image is calculated; A voice prompt is generated based on the location of the target object in the minimap image.

[0006] In one embodiment, after the step of recording the screen and capturing game images at a preset frequency, the method further includes: The game image is preprocessed, the preprocessing including: Obtain the deflection direction of the gaming device, and perform deflection recovery on the game image based on the deflection direction of the gaming device; and / or, Set the red channel of the game image to zero; and / or, The area at the edge of the game image with a preset ratio is cut off.

[0007] In one embodiment, cropping the game image to obtain an image to be identified that includes the minimap region includes: The game image is scaled to a valid image size, and the area in the upper right corner of the scaled game image is cropped according to a predetermined ratio to obtain an image to be identified that includes the minimap area.

[0008] In one embodiment, scaling the game image to an effective image size includes: Calculate the scaling factor based on the ratio of the shorter side size to the reference size of the game image; The game image is scaled according to the scaling factor to obtain a game image of effective image size.

[0009] In one embodiment, the method further includes: The confidence level of the target object type is compared with the corresponding confidence threshold to determine whether the target object type identification result is reliable. The step of generating voice prompts based on the location of the target object in the minimap image includes: When the identification result of the target object type is reliable, a voice prompt is generated based on the location of the target object in the minimap image.

[0010] In one embodiment, the method further includes: Test the hardware performance of gaming devices; The step of inputting the image to be recognized into the trained image recognition model includes: When the hardware performance of the gaming device exceeds a preset standard, the image to be recognized is input into a trained standard image recognition model; When the hardware performance of the gaming device is lower than a preset standard, the image to be recognized is input into a trained lightweight image recognition model, wherein the recognition accuracy of the lightweight image recognition model is lower than that of the standard image recognition model.

[0011] In one embodiment, generating a voice prompt based on the location information of the target object includes: Based on the location information of the target object and the coordinate system constructed with the center point of the minimap image, the azimuth angle of the target object relative to the player character is calculated using the azimuth angle calculation model. Based on the azimuth angle of the target object relative to the player character, and the type of the target character, a voice prompt is generated.

[0012] One or more technical solutions proposed in this application have at least the following technical effects: A method for assisting in shooting games is proposed. This method involves recording the screen and cropping the game image at a preset frequency; cropping the game image to obtain a target image containing a minimap region; inputting the target image into a trained image recognition model to identify target object information and minimap marker locations; cropping the target image based on the minimap marker locations to obtain a minimap image and calculating the position of the target object within the minimap image; and generating voice prompts based on the target object's position within the minimap image, thereby improving the accuracy of target object recognition. Attached Figure Description

[0013] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 A flowchart illustrating the first embodiment of the shooting game assistance method provided in this application; Figure 2 This is a comparison chart of azimuth angles at different cutting precisions.

[0016] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application. To better understand the technical solutions of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0018] In related technologies, there are already methods based on image recognition technology to identify small maps and provide assistance to game players. However, due to the influence of the input image of the model, the recognition results are distorted, resulting in poor assistance effect.

[0019] For example, using the entire image as input to the model results in too many invalid interfering elements, increasing the computational load for feature extraction and easily leading to misidentification. Another example is using a cropped minimap as input, but if the cropping direction is off or the resolution is inconsistent, the cropped minimap will be incomplete (e.g., ...). Figure 2 (As shown) or includes redundant elements, which leads to lower accuracy in subsequent location identification and calculation.

[0020] To address the aforementioned issues, the solution provided in this application involves recording the screen and capturing game images at a preset frequency; cropping the game images to obtain an image to be recognized that includes a minimap area; inputting the image to be recognized into a trained image recognition model to identify target object information and minimap marker locations; cropping the image to be recognized based on the minimap marker locations to obtain a minimap image and calculating the position of the target object within the minimap image; and generating voice prompts based on the position of the target object within the minimap image, thereby improving the accuracy of target object recognition.

[0021] This application provides an auxiliary method for shooting games.

[0022] In the first embodiment of the shooting game assistance method of this application, refer to Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the shooting game assistance method of this application. The shooting game assistance method may include steps S10 to S50: Step S10: Record the screen and capture game images at a preset frequency.

[0023] In this embodiment, the game image is captured at a frequency of 120Hz.

[0024] It should be noted that in this step, the game image refers to the complete game image, which includes at least all elements of the game interface. The size of the game image will vary depending on the gaming device. For example, if the gaming device is a mobile phone, the size may be 720*1080. Different screen ratios of mobile phones (such as 16:9 / 18:9 / 19.5:9, etc.) will also result in different sizes. When the gaming device is a tablet, the size of the game image may be 1120*980. These are not listed one by one here.

[0025] Step S20: Crop the game image to obtain an image to be identified that includes the minimap area.

[0026] In one feasible implementation, step S20 may include the following steps: The game image is scaled to a valid image size, and the area in the upper right corner of the scaled game image is cropped according to a predetermined ratio to obtain an image to be identified that includes the minimap area.

[0027] Preferably, cropping the game image to obtain an image to be identified that includes the minimap region includes: Calculate the scaling factor based on the ratio of the shorter side size to the reference size of the game image; scale the game image according to the scaling factor to obtain a game image of effective image size.

[0028] It should be noted that the short side dimension refers to the length of the shorter side of the game image. For example, if the game image is 720*1080, then the short side dimension is 720. The reference dimension is the length of the shorter side of the effective image size. In this embodiment, the reference dimension is configured as 1080, so the scaling factor is 1.69 (1080 / 640), which enlarges the game image by 1.69 times to obtain the game image of the effective image size.

[0029] Step S30: Input the image to be identified into the trained image recognition model to identify the target object information and the location marked on the minimap. The target object information includes the type of the target object and its location in the image to be identified.

[0030] It should be noted that the minimap markers are used to indicate the location of the minimap. For example, if the minimap is positioned below the main minimap area, its precise location within the image to be identified can be determined after its location is determined. The target object is a marker on the minimap, such as a gunshot marker or a footsteps marker.

[0031] Step S40: Based on the location marked on the minimap, crop the image to be identified to obtain a minimap image, and calculate the position of the target object in the minimap image.

[0032] It should be noted that the center point of the minimap image is the location of the player in the game, and the position of the target object in the minimap image can be represented by a coordinate system constructed with the center point of the minimap as the origin.

[0033] Step S50: Generate a voice prompt based on the location of the target object in the minimap image.

[0034] In one feasible implementation, step S50 may include the following steps: Based on the location information of the target object and the coordinate system constructed with the center point of the minimap image, the azimuth angle of the target object relative to the player character is calculated using the azimuth angle calculation model. Based on the azimuth angle of the target object relative to the player character, and the type of the target character, a voice prompt is generated.

[0035] In this embodiment, the expression for the azimuth calculation model is: θ=arctan2(Δx,Δy)×180 / π+90 In the formula, θ represents the azimuth angle, arctan2(Δx,Δy) represents the arctangent function in the four quadrants, and (Δx,Δy) is the coordinate difference between the target object and the game player.

[0036] In one feasible implementation, the following steps are included after step S10: The game image is preprocessed, the preprocessing including: obtaining the deflection direction of the game device, and performing deflection recovery on the game image according to the deflection direction of the game device; and / or, zeroing the red channel of the game image; and / or, cropping a region of the game image edge by a preset proportion.

[0037] It should be noted that by deflecting and restoring the game image, the accuracy of subsequent image recognition can be improved; zeroing the red channel of the game image can eliminate red interference in the image; and cutting off a pre-proportioned area at the edge of the game image can avoid interference from the phone's status bar.

[0038] In one feasible implementation, the method further includes: The confidence level of the target object type is compared with the corresponding confidence threshold to determine whether the target object type identification result is reliable.

[0039] Accordingly, step S50 may include the following steps: When the identification result of the target object type is reliable, a voice prompt is generated based on the location of the target object in the minimap image.

[0040] In this embodiment, the confidence threshold is configured to 0.8 when the target object is an opponent; the confidence threshold is configured to 0.75 when the target object is a safe zone; and the confidence threshold is configured to 0.8 when the target object is a sound source. In one feasible implementation, the method further includes: Test the hardware performance of gaming devices; Accordingly, step S30 may include the following steps: When the hardware performance of the gaming device exceeds a preset standard, the image to be recognized is input into a trained standard image recognition model; When the hardware performance of the gaming device is lower than a preset standard, the image to be recognized is input into a trained lightweight image recognition model, wherein the recognition accuracy of the lightweight image recognition model is lower than that of the standard image recognition model.

[0041] This embodiment provides an auxiliary method for shooting games. It involves recording the screen and capturing game images at a preset frequency; cropping the game images to obtain an image to be recognized that includes a minimap area; inputting the image to be recognized into a trained image recognition model to identify target object information and minimap marker positions; cropping the image to be recognized based on the minimap marker positions to obtain a minimap image and calculating the position of the target object within the minimap image; and generating voice prompts based on the target object's position within the minimap image, thereby improving the accuracy of target object recognition.

[0042] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for assisting in shooting games, characterized in that, The method includes: Record the screen and capture game images at a preset frequency; The game image is cropped to obtain an image to be identified that includes the minimap area; The image to be identified is input into a trained image recognition model to identify target object information and small map location markers. The target object information includes the type of the target object and its location in the image to be identified. Based on the location marked on the minimap, the image to be identified is cropped to obtain a minimap image, and the position of the target object in the minimap image is calculated; Based on the location of the target object in the minimap image, a voice prompt is generated; The step of generating voice prompts based on the location information of the target object includes: Based on the location information of the target object and the coordinate system constructed with the center point of the minimap image, the azimuth angle of the target object relative to the player character is calculated using the azimuth angle calculation model. Based on the azimuth angle of the target object relative to the player character, and the type of the target character, a voice prompt is generated.

2. The shooting game assistance method as described in claim 1, characterized in that, After the step of recording the screen and capturing game images at a preset frequency, the method further includes: The game image is preprocessed, the preprocessing including: Obtain the deflection direction of the gaming device, and perform deflection recovery on the game image based on the deflection direction of the gaming device; and / or, Set the red channel of the game image to zero; and / or, The area at the edge of the game image with a preset ratio is cut off.

3. The shooting game assistance method as described in claim 1, characterized in that, The step of cropping the game image to obtain the image to be identified, which includes the minimap region, includes: The game image is scaled to a valid image size, and the area in the upper right corner of the scaled game image is cropped according to a predetermined ratio to obtain an image to be identified that includes the minimap area.

4. The shooting game assistance method as described in claim 3, characterized in that, The step of scaling the game image to a valid image size includes: Calculate the scaling factor based on the ratio of the shorter side size to the reference size of the game image; The game image is scaled according to the scaling factor to obtain a game image of effective image size.

5. The shooting game assistance method as described in claim 1, characterized in that, The method further includes: The confidence level of the target object type is compared with the corresponding confidence threshold to determine whether the target object type identification result is reliable. The step of generating voice prompts based on the location of the target object in the minimap image includes: When the identification result of the target object type is reliable, a voice prompt is generated based on the location of the target object in the minimap image.

6. The shooting game assistance method as described in claim 1, characterized in that, The method further includes: Test the hardware performance of gaming devices; The step of inputting the image to be recognized into the trained image recognition model includes: When the hardware performance of the gaming device exceeds a preset standard, the image to be recognized is input into a trained standard image recognition model; When the hardware performance of the gaming device is lower than a preset standard, the image to be recognized is input into a trained lightweight image recognition model, wherein the recognition accuracy of the lightweight image recognition model is lower than that of the standard image recognition model.

Images