Image processing method, device, edge computing device and computer storage medium

Abstract

The disclosed embodiments propose an image processing method, apparatus, edge computing device, and computer storage medium, the method comprising: determining a recognition result of each frame of a game platform image in a plurality of frames of game platform images, the recognition result comprising at least information on items replacing funds; in the case of using a sliding window to determine N frames of game platform images in the plurality of frames of game platform images each time, re-determining the information on items replacing funds in a target frame image according to the recognition results of the N frames of game platform images in the sliding window, wherein the sliding order of the sliding window is the frame order of the plurality of frames of game platform images, N is an integer greater than 1, and the target frame image is a frame image in the N frames of game platform images in the sliding window.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This disclosure claims priority to Singapore patent application number 10202106600X filed with the Intellectual Property Office of Singapore on June 18, 2021, the entire contents of which are incorporated by reference into this application. Technical Field

[0003] The present disclosure relates to computer vision processing technology, and is related to but not limited to an image processing method, apparatus, edge computing device, and computer storage medium. Background Art

[0004] At present, the game platform image can be captured by an image acquisition device, thereby detecting the money substitute items in the game platform image. However, factors such as the occlusion between the money substitute items, the occlusion of the money substitute items by the players, and the on-site lighting will affect the accuracy of the money substitute item detection results, thereby reducing the recognition accuracy of the money substitute item detection to a certain extent. Summary of the invention

[0005] The embodiments of the present disclosure can provide an image processing method, an apparatus, an edge computing device, and a computer storage medium, which can more accurately obtain the detection results of money substitute items.

[0006] The present disclosure provides an image processing method, the method comprising:

[0007] Determine a recognition result of each frame of the game platform image in the plurality of frames of the game platform image, wherein the recognition result at least includes information of a fund replacement item;

[0008] In the case where N frames of the game platform image are determined each time by using a sliding window, the information on funds replacing items in the target frame image is re-determined according to the recognition results of the N frames of the game platform image in the sliding window, wherein the sliding order of the sliding window is the frame order of the multiple frames of the game platform image, N is an integer greater than 1, and the target frame image is a frame image among the N frames of the game platform image in the sliding window.

[0009] In some embodiments, the re-determining the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window includes:

[0010] For each of the N frames of game platform images within the sliding window, determine the attribute data of the same money substitute item in the recognition result of each of the N frames of game platform images by target tracking;

[0011] Among the attribute data of the same money substitute item, the attribute data with the largest number of occurrences is determined; and the money substitute item information in the target frame image is determined as: the attribute data with the largest number of occurrences.

[0012] It is understandable that the information on the money substitute items in the target frame image may be inaccurate due to the influence of the brightness of the on-site lighting, the occlusion of the money substitute items, the occlusion of the money substitute items by the players during the game, the movement of the money substitute items and other factors. In the embodiment of the present disclosure, the attribute data that appears most frequently in the attribute data of the N-frame game platform image is determined to determine the information on the money substitute items in the target frame image, which can improve the accuracy of the information on the money substitute items in the target frame image to a certain extent.

[0013] In some embodiments, the re-determining the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window includes:

[0014] For each of the N frames of game platform images within the sliding window, determine the attribute data of the same money substitute item in the recognition result of each of the N frames of game platform images by target tracking;

[0015] Determine the attribute data whose confidence is greater than or equal to the confidence threshold from the attribute data as the target attribute data of the funds replacement item;

[0016] Among the target attribute data of the same money substitute item, the target attribute data with the largest number of occurrences is determined; and the money substitute item information in the target frame image is determined as: the target attribute data with the largest number of occurrences.

[0017] It can be understood that in the embodiments of the present disclosure, attribute data with a confidence level greater than or equal to a confidence threshold is determined from the attribute data of N frames of game platform images, that is, the target attribute data is determined, thereby determining the information on funds substitute items in the target frame image based on the target attribute data that appears the most times, which can improve the accuracy of the information on funds substitute items in the target frame image to a certain extent.

[0018] In some embodiments, the re-determining the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window includes:

[0019] For each of the N frames of game platform images in the sliding window, determine the attribute data of at least two fund replacement items in the recognition results of the N frames of game platform images;

[0020] Among the attribute data of the at least two money substitute items, the attribute data of each of the at least two money substitute items is determined by using the tracking identifiers of each of the at least two money substitute items;

[0021] For each of the at least two money substitute items, determining the attribute data that appears most frequently;

[0022] The money substitute item information in the target frame image is determined as: the attribute data with the largest number of occurrences determined for each of the money substitute items.

[0023] It can be understood that in the embodiments of the present disclosure, the attribute data of different funds substitute items in N frames of game platform images can be determined based on the tracking identifiers of the funds substitute items. Thus, the attribute data with the highest number of occurrences can be determined for the attribute data of at least two funds substitute items, thereby determining the funds substitute item information in the target frame image, which can improve the accuracy of the funds substitute item information in the target frame image to a certain extent.

[0024] In some embodiments, the attribute data of the fund substitute item includes at least one of the following: the denomination of the fund substitute item, the type of the fund substitute item, the quantity of the fund substitute item, and the information of the owner of the fund substitute item.

[0025] It can be seen that the embodiments of the present disclosure can accurately determine the denomination, quantity, type and other information of the money substitute item.

[0026] In some embodiments, the N frames of game platform images are N consecutive frames of images in the multiple frames of game platform images.

[0027] The disclosed embodiment can more accurately obtain the money substitute item information in the target frame image based on the money substitute item information in N consecutive frames of images.

[0028] In some embodiments, the target frame image is the earliest frame image captured among the N frames of game platform images.

[0029] In this way, the embodiment of the present disclosure can obtain the money substitute item information of the earliest frame image collected among N frames of images, thereby obtaining accurate money substitute item information in a timely manner.

[0030] In some embodiments, the method further comprises:

[0031] According to the money substitute item information in the target frame image, the business detection logic related to the money substitute item is executed.

[0032] The disclosed embodiment can accurately execute the business detection logic related to the money substitute items on the basis of accurately obtaining the money substitute item information in the target frame image, and can reduce the probability of false alarms and the like caused by erroneous money substitute item information to a certain extent.

[0033] In some embodiments, executing the business detection logic related to the money substitute item according to the money substitute item information in the target frame image includes:

[0034] Determine a human hand detection frame and a money substitute item detection frame in the target frame image;

[0035] In the case where the human hand detection frame does not overlap with the money substitute item detection frame, the business detection logic related to the money substitute item is executed according to the money substitute item information in the target frame image.

[0036] It can be seen that the embodiments of the present disclosure can execute the business detection logic related to the funds substitute items for the funds substitute items where there is no overlap between the human hand and the funds substitute items, thereby executing the business detection logic related to the funds substitute items when it is determined that the human hand does not block the funds substitute items, thereby reducing the probability of business detection logic errors caused by the blocking of the funds substitute items by the human hand to a certain extent.

[0037] In some embodiments, executing the business detection logic related to the money substitute item according to the money substitute item information in the target frame image includes:

[0038] Determine change information of the money substitute item information in the plurality of target frame images arranged in chronological order according to the re-determined money substitute item information in the plurality of target frame images;

[0039] The change information is pushed to the management device of the game platform.

[0040] It can be seen that the disclosed embodiment can push the change information of the funds replacing items in a timely manner, which is conducive to the subsequent processing of the change information of the funds replacing items on the management device side.

[0041] In some embodiments, determining the recognition result of each frame of the game platform image in the multiple frames of the game platform image includes:

[0042] The target object in each frame of the image is identified, and the identified target object is mapped to a predetermined area division map to obtain a recognition result of each preset area in each frame of the image.

[0043] In the disclosed embodiment, the predetermined area division map can accurately represent the target objects in each area of ​​the game platform. Therefore, the identified target objects are mapped to the predetermined area division map, and the recognition results of each frame image can be obtained more accurately.

[0044] The present disclosure also provides an image processing device, the device comprising:

[0045] A determination module, used to determine the recognition result of each frame of the game platform image in the multiple frames of the game platform image, wherein the recognition result at least includes the information of the fund replacement item;

[0046] A processing module is used to re-determine the information of funds replacing items in the target frame image according to the recognition results of the N frames of game platform images in the sliding window each time when the sliding window is used to determine N frames of game platform images in the multiple frames of game platform images, wherein the sliding order of the sliding window is the frame order of the multiple frames of game platform images, N is an integer greater than 1, and the target frame image is a frame image of the N frames of game platform images in the sliding window.

[0047] In some embodiments, the processing module is used to re-determine the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window, including:

[0048] For each of the N frames of game platform images within the sliding window, determine the attribute data of the same money substitute item in the recognition result of each of the N frames of game platform images by target tracking;

[0049] Among the attribute data of the same money substitute item, determine the attribute data that appears the most times;

[0050] The money substitute item information in the target frame image is determined as: the attribute data with the largest number of appearances.

[0051] In some embodiments, the processing module is used to re-determine the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window, including:

[0052] For each of the N frames of game platform images within the sliding window, determine the attribute data of the same money substitute item in the recognition result of each of the N frames of game platform images by target tracking;

[0053] Determine the attribute data whose confidence is greater than or equal to the confidence threshold from the attribute data as the target attribute data of the funds replacement item;

[0054] Among the target attribute data of the same fund replacement item, determine the target attribute data that appears the most times;

[0055] The fund substitute item information in the target frame image is determined as: the target attribute data with the largest number of appearances.

[0056] In some embodiments, the processing module is used to re-determine the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window, including:

[0057] For each of the N frames of game platform images in the sliding window, determine the attribute data of at least two fund replacement items in the recognition results of the N frames of game platform images;

[0058] Among the attribute data of the at least two money substitute items, the attribute data of each of the at least two money substitute items is determined by using the tracking identifiers of each of the at least two money substitute items;

[0059] For each of the at least two money substitute items, determining the attribute data that appears most frequently;

[0060] The money substitute item information in the target frame image is determined as: the attribute data with the largest number of occurrences determined for each of the money substitute items.

[0061] In some embodiments, the attribute data of the fund substitute item includes at least one of the following: the denomination of the fund substitute item, the type of the fund substitute item, the quantity of the fund substitute item, and the information of the owner of the fund substitute item.

[0062] In some embodiments, the N frames of game platform images are N consecutive frames of images in the multiple frames of game platform images.

[0063] In some embodiments, the target frame image is the earliest frame image captured among the N frames of game platform images.

[0064] In some embodiments, the processing module is further used to execute business detection logic related to the money substitute item according to the money substitute item information in the target frame image.

[0065] In some embodiments, the processing module is used to execute the business detection logic related to the money substitute item according to the money substitute item information in the target frame image, including:

[0066] Determine a human hand detection frame and a money substitute item detection frame in the target frame image;

[0067] In the case where the human hand detection frame does not overlap with the money substitute item detection frame, the business detection logic related to the money substitute item is executed according to the money substitute item information in the target frame image.

[0068] In some embodiments, the processing module is used to execute the business detection logic related to the money substitute item according to the money substitute item information in the target frame image, including:

[0069] Determine change information of the money substitute item information in the plurality of target frame images arranged in chronological order according to the re-determined money substitute item information in the plurality of target frame images;

[0070] The change information is pushed to the management device of the game platform.

[0071] In some embodiments, the determination module is used to determine the recognition result of each frame of the game platform image in the multiple frames of the game platform image, including:

[0072] The target object in each frame of the image is identified, and the identified target object is mapped to a predetermined area division map to obtain a recognition result of each preset area in each frame of the image.

[0073] The embodiment of the present disclosure further provides an edge computing device, the edge computing device is used to receive a plurality of frames of game platform images sent by an image acquisition device; the plurality of frames of game platform images are images acquired by the image acquisition device;

[0074] The edge computing device includes a processor and a memory for storing a computer program that can be run on the processor; wherein,

[0075] The processor is used to run the computer program to execute any one of the above-mentioned image processing methods.

[0076] The embodiment of the present disclosure further provides a computer storage medium on which a computer program is stored. When the computer program is executed by a processor, any one of the above-mentioned image processing methods is implemented.

[0077] In the image processing method, apparatus, edge computing device and computer storage medium proposed in the embodiments of the present disclosure, the recognition result of each frame of a game platform image in a plurality of frames of game platform images is determined, and the recognition result includes at least information on items replacing funds; when a sliding window is used to determine N frames of game platform images in a plurality of frames of game platform images each time, the information on items replacing funds in a target frame image is re-determined based on the recognition results of the N frames of game platform images in the sliding window, wherein the sliding order of the sliding window is the frame order of the plurality of frames of game platform images, N is an integer greater than 1, and the target frame image is a frame image in the N frames of game platform images in the sliding window.

[0078] It can be seen that in the embodiment of the present disclosure, the recognition results of N frames of game platform images in the sliding window can be used to determine the money substitute item information of the target frame image. Since the recognition results of N frames of game platform images contain more and more accurate money substitute item information, the embodiment of the present disclosure can more accurately obtain the money substitute item information of the target frame image, thereby effectively improving the recognition accuracy of money substitute items.

[0079] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used to illustrate the technical solutions of the present disclosure together with the specification.

[0081] Figure 1 is a flowchart of an image processing method according to an embodiment of the present disclosure;

[0082] Figure 2A is a reference schematic diagram of a game platform image in an embodiment of the present disclosure;

[0083] Figure 2B For Figure 2A A division map of the areas where funds can be substituted for other items is obtained on the basis of

[0084] Figure 3 is a schematic diagram of the composition structure of an image processing device according to an embodiment of the present disclosure;

[0085] Figure 4 A schematic diagram of the structure of an edge computing device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0086] The present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the embodiments provided herein are only used to explain the present disclosure and are not intended to limit the present disclosure. In addition, the embodiments provided below are partial embodiments for implementing the present disclosure, rather than providing all embodiments for implementing the present disclosure. In the absence of conflict, the technical solutions recorded in the embodiments of the present disclosure can be implemented in any combination.

[0087] It should be noted that, in the embodiments of the present disclosure, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a method or apparatus including a series of elements includes not only the elements explicitly recorded, but also includes other elements not explicitly listed, or also includes elements inherent to the implementation of the method or apparatus. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other related elements (such as steps in a method or units in a device, for example, a unit may be a part of a circuit, a part of a processor, a part of a program or software, etc.) in the method or apparatus including the element.

[0088] For example, the image processing method provided by the embodiment of the present disclosure includes a series of steps, but the image processing method provided by the embodiment of the present disclosure is not limited to the recorded steps. Similarly, the image processing device provided by the embodiment of the present disclosure includes a series of modules, but the device provided by the embodiment of the present disclosure is not limited to including the modules explicitly recorded, and may also include modules that need to be set up to obtain relevant information or perform processing based on information.

[0089] The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set consisting of A, B, and C.

[0090] The disclosed embodiments can be applied to edge computing devices in gaming scenarios and can operate with many other general or special computing system environments or configurations. Here, edge computing devices can be thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, small computer systems, etc.

[0091] Edge computing devices can execute instructions through program modules. Generally, program modules can include routines, programs, target programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types. Computer systems / servers can be implemented in a distributed cloud computing environment, where tasks are performed by remote processing devices linked through a communication network. In a distributed cloud computing environment, program modules can be located on local or remote computing system storage media including storage devices.

[0092] In the relevant technology, in the game scene, the calculation of game results, the alarm detection of money substitute items (such as the minimum denomination of money substitute items, the maximum denomination of money substitute items), etc. all need to use the money substitute item information. Only by obtaining accurate money substitute item information can the business detection logic related to money substitute items be accurately executed to meet the requirements of the game scene; however, in the actual game scene, there are a large number of objective factors that affect the recognition results of money substitute items, resulting in errors in the money substitute item information. For example, the influence of the brightness of the scene lights, the occlusion between money substitute items (the distance between money substitute items is close, and different money substitute items are at different heights), the occlusion of money substitute items by players during the game, the movement of money substitute items, etc. will cause the jump of money substitute item detection results. Therefore, how to improve the accuracy of money substitute item detection results is a technical problem that needs to be solved urgently.

[0093] In response to the above technical problems, in some embodiments of the present disclosure, a technical solution for image processing is proposed, which can be applied to game scenarios.

[0094] The application scenarios of the embodiments of the present disclosure are exemplarily described below.

[0095] In the game scenario, the running status of various games can be monitored through computer vision processing technology. Here, the operation of each game is related to money substitute items.

[0096] In some embodiments, the game in the game scene may be a poker game or other games on a game platform, and the embodiments of the present disclosure are not limited to this.

[0097] In the disclosed embodiments, computer vision is a science that studies how to make machines "see", which means using cameras and computers to replace human eyes to identify, track and measure targets, and further perform image processing. During the game, three cameras can be used to detect what happens on the game platform for further analysis; the game platform can be a physical desktop platform or other physical platform.

[0098] Figure 1 is a flowchart of an image processing method according to an embodiment of the present disclosure, as shown in Figure 1 As shown, the process may include:

[0099] Step 101: Determine the recognition result of each frame of the game platform image in the multiple frames of the game platform image, and the recognition result at least includes the information of the fund replacement item.

[0100] In the embodiments of the present disclosure, at least one camera may be used to photograph the game platform to obtain video data or image data, and then multiple frames of game platform images may be obtained from the video data or image data; in some embodiments, the camera that photographs the game platform may be a camera located directly above the game platform for photographing a bird's-eye view of the game platform, or may be a camera that photographs the game platform from other angles; accordingly, each frame of the game platform image may be an image of the game platform from a bird's-eye view or other perspectives; in other embodiments, each frame of the game platform image may also be an image obtained by fusing a bird's-eye view with an image of the game platform from another perspective.

[0101] After obtaining each frame of the image, each frame of the game platform image can be processed by computer vision processing technology to obtain the recognition result of each frame of the game platform image; in some embodiments, each frame of the game platform image can be subjected to target recognition to obtain the target object in each frame of the game platform image, and the target object at least includes a fund substitute item. For example, the target object can also include a human body and a playing card. The human body in the target object can include the entire human body, or can include parts of the human body such as a human hand and a human face; the playing card in the target object can be a type of playing card such as a spade, a heart, a diamond, or a club. After obtaining the target object in each frame of the image, the corresponding recognition result can be determined based on the target object in each frame of the image, and here, the recognition result can be the information of the target object.

[0102] In some embodiments, the information of the money substitute item may include the attribute data of the money substitute item, and the attribute data of the money substitute item may include at least one of the following: the denomination of the money substitute item, the type of the money substitute item, the quantity of the money substitute item, and the information of the owner of the money substitute item; in practical applications, the information of the money substitute item may be determined for one money substitute item, in which case the number of the money substitute item is 1; multiple money substitute items in contact with each other may also be regarded as the same money substitute item, so as to analyze the information of the same money substitute item, in which case the number of the money substitute items is greater than 1, for example, for a stack of money substitute items, the stack of money substitute items may be regarded as the same money substitute item, and the number of the money substitute items in the stack of money substitute items is greater than 1. The information of the owner of the money substitute item may include the identity information of the owner of the money substitute item.

[0103] In some embodiments, the denomination, type, quantity, and other information of the fund substitute item can be determined by analyzing the image of the fund substitute item; the information of the owner of the fund substitute item can be determined based on the image of the human body in contact with the fund substitute item.

[0104] Step 102: When using the sliding window to determine N frames of game platform images in multiple frames of game platform images each time, re-determine the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window, wherein the sliding order of the sliding window is the frame order of the multiple frames of game platform images, N is an integer greater than 1, and the target frame image is a frame image in the N frames of game platform images in the sliding window.

[0105] In some embodiments, the size of the sliding window can be pre-set through a configuration file, that is, the value of N can be configured through a configuration file. After obtaining the recognition results of each frame of the game platform image, the recognition results of each frame of the game platform image can be stored in the storage area corresponding to the sliding window in chronological order. In some embodiments, the recognition results of each frame of the game platform image are located in a message queue, and the recognition results of each frame in the message queue can be read, and the recognition results of each frame of the image read can be stored in the storage area corresponding to the sliding window in sequence. When the number of recognition results in the sliding window reaches N, if the recognition result of a new frame of the image in the message queue is to be read, the sliding window moves, thereby pushing out a frame of the image, and the recognition result of the pushed out frame of the image can be obtained. The recognition result of the target frame image represents: after the sliding window moves, the recognition result of a frame of the image changes from inside the sliding window to outside the sliding window.

[0106] In some embodiments, an edge computing device can be used to receive multiple frames of game platform images sent by an image acquisition device; the multiple frames of game platform images are images captured by the image acquisition device; the image acquisition device may include at least one of the cameras described above.

[0107] Accordingly, the edge computing device can be used to detect and identify each frame of the multiple frames of the game platform image to obtain the identification result of each frame of the multiple frames of the game platform image; the edge computing device can be used to determine the information of the funds replacement items in the target frame image based on the identification results of the N frames of the game platform image in the sliding window.

[0108] In practical applications, steps 101 to 102 can be implemented using a processor in an edge computing device, and the processor can be at least one of an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field-programmable gate array (FPGA), a central processing unit (CPU), a controller, a microcontroller, and a microprocessor.

[0109] It can be seen that in the embodiment of the present disclosure, the recognition results of multiple frames of game platform images in the sliding window can be used to determine the money substitute item information of the target frame image. Since the recognition results of multiple frames of game platform images contain more and more accurate money substitute item information, the embodiment of the present disclosure can more accurately obtain the money substitute item information of the target frame image, thereby effectively improving the recognition accuracy of money substitute items.

[0110] Furthermore, the disclosed embodiments can be applied not only to poker game scenarios, but can also be reused in various scenarios where money is used as a substitute for items. Thus, the cost of use can be reduced, and accurate identification of money substitute item information can be achieved at a faster speed, which is easy to implement.

[0111] In some embodiments, the above-mentioned determination of the recognition result of each frame of the game platform image in multiple frames of game platform images may include: identifying the target object in each frame of the image, mapping the identified target object to a predetermined area division map, and obtaining the recognition result of each preset area of ​​each frame of the image.

[0112] The target object may include the above-mentioned money substitute items, and may also include other objects such as game props in the game platform image. Exemplarily, the target object in each frame of the image may be one or more; the region division map may be used to indicate the correct region where various types of target objects are located. After mapping any target object to a predetermined region division map, if the target object is in the corresponding correct region in the region division map, the information of the target object is retained in the recognition result of each frame of the game platform image; if the target object is not in the corresponding correct region in the region division map, the information of the target object may be filtered out in the recognition result of each frame of the game platform image.

[0113] For example, Figure 2A is a reference schematic diagram of a game platform image in an embodiment of the present disclosure, Figure 2A In the figure, D1, D2, D3, D4, D5, D6 and D7 represent different areas in the game platform.

[0114] Figure 2B For Figure 2A Based on the division diagram of the capital substitution area, it can be seen that Figure 2B The money substitute item area 201 is shown in FIG. Figure 2A Middle area D1 to area D7.

[0115] Reference Figure 2B , in the case where the target object is a capital substitute, if the capital substitute is located Figure 2B The fund substitute item area 201 in the Figure 2B If the funds substitute item is outside the fund substitute item area 201 shown, the information of the fund substitute item is deleted.

[0116] It can be seen that the predetermined area division map can accurately represent the target objects in each area of ​​the game platform. Therefore, the identified target objects are mapped to the predetermined area division map, and the recognition results of each preset area in each frame image can be obtained more accurately.

[0117] In some embodiments, the above-mentioned N-frame game platform images can be non-continuous N-frame images in a multi-frame game platform image, or can be continuous N-frame images in a multi-frame game platform image; in the case where the above-mentioned N-frame game platform images are continuous N-frame images, the embodiment of the present disclosure can more accurately derive the fund replacement item information in the target frame image based on the fund replacement item information of the continuous N-frame images.

[0118] In some embodiments, the above-mentioned target frame image is the earliest frame image collected among the N frames of game platform images; in this way, the embodiment of the present disclosure can obtain the funds replacement item information of the earliest frame image collected among the N frames of images, thereby obtaining accurate funds replacement item information in a timely manner.

[0119] In one implementation, the value of N is 5, and the recognition results of the first to fifth frames of images can be stored in the sliding window in sequence. When the number of recognition results in the sliding window reaches 5, if the recognition result of the sixth frame of image is to be stored in the sliding window, the right edge of the sliding window moves to the right, so that the recognition result of the first frame of image is pushed out of the sliding window. At this time, the first frame of image is the above-mentioned target frame of image; when the recognition result of the first frame of image is pushed out of the sliding window, the data in the sliding window are the recognition results of the second to sixth frames of image. Afterwards, if the recognition results of seven frames of image are to be stored in the sliding window, the right edge of the sliding window moves to the right, so that the recognition result of the second frame of image is pushed out of the sliding window. At this time, the second frame of image is the above-mentioned target frame of image; when the recognition result of the second frame of image is pushed out of the sliding window, the data in the sliding window are the recognition results of the third to seventh frames of image. By analogy, through the sliding mechanism of the sliding window, multiple target frame images and the recognition results of each target frame image can be determined in turn.

[0120] In some embodiments, the above-mentioned re-determining the information of the money replacement item in the target frame image according to the recognition results of the N frames of game platform images in the sliding window may include:

[0121] For each of the N frames of game platform images within a sliding window, the attribute data of the same money substitute item in the recognition results of each frame of the N frames of game platform images are determined through target tracking; among the attribute data of the same money substitute item, the attribute data with the largest number of occurrences is determined; and the money substitute item information in the target frame image is determined as: the attribute data with the largest number of occurrences.

[0122] In the disclosed embodiment, the same money substitute item includes at least one money substitute item. For the recognition results of N frames of game platform images, a target tracking method can be used to determine the attribute data of the same money substitute item in the recognition results of N frames of game platform images; in some embodiments, the tracking identifier of the money substitute item in the target frame image is determined by a target detection method, and then the tracking ID of the money substitute item in the target frame image is used as a reference to perform target tracking in the recognition results of N frames of game platform images in the sliding window, thereby determining the attribute data of the same money substitute item in the recognition results of N frames of game platform images. Exemplarily, the tracking identifier can be a tracking ID (track identity document).

[0123] In the disclosed embodiment, after the attribute data of the same money substitute item is determined, the attribute data with the largest number of occurrences can be determined through election among the attribute data of the same money substitute item.

[0124] In some embodiments, when the attribute data includes the denomination of the money substitute item, the denomination data that appears most frequently can be determined by election in the attribute data of the same money substitute item; for example, the value of N is 5, the data in the sliding window is the recognition results of the 1st frame image to the 5th frame image, the target frame image is the 1st frame image, and according to the recognition results of the 1st frame image to the 5th frame image, the denomination of the same money substitute item in the 1st frame image to the 5th frame image can be identified; when the denominations of the money substitute items with the same tracking ID in the 1st frame image, the 2nd frame image, the 3rd frame image, the 4th frame image and the 5th frame image are respectively identified as 200, 200, 300, 300 and 300, Through election, it can be determined that the denomination that appears most times is 300. In this way, it can be considered that the denomination of the corresponding fund substitute item in the first frame image is inaccurate, and the denomination of the corresponding fund substitute item in the first frame image can be updated to 300; when the denominations of fund substitute items with the same tracking ID in the first frame image, the second frame image, the third frame image, the fourth frame image and the fifth frame image are respectively identified as 400, 400, 400, 400 and 300, it can be determined through election that the denomination that appears most times is 400. In this way, it can be considered that the denomination of the fund substitute item in the first frame image is accurate, and the denomination of the fund substitute item in the first frame image can be kept unchanged at 400.

[0125] In this way, after re-determining the denomination of the money substitute item in the first frame image, the sliding window can be moved backwards, the first frame image can be pushed out of the sliding window, and then the second frame image becomes the first frame in the sliding window. At this time, the money substitute item information in the second frame image is updated according to the recognition results of the second to seventh frames in the sliding window. Thus, by moving the sliding window backwards in sequence according to the frame order, the money substitute item information in the first frame, the second frame, the third frame, ... can be updated successively to obtain a reliable recognition result for the video frame sequence.

[0126] In some embodiments, when the attribute data includes the number of fund substitute items, the information on the number of fund substitute items that appear the most times can be determined by election in the attribute data of the same fund substitute item; for example, the value of N is 5, the data in the sliding window is the recognition results of the 1st frame image to the 5th frame image, the target frame image is the 1st frame image, and based on the recognition results of the 1st frame image to the 5th frame image, the number of the same fund substitute item in the 1st frame image to the 5th frame image can be identified; in the 1st frame image, the 2nd frame image, the 3rd frame image, the 4th frame image and the 5th frame image, the number of fund substitute items with the same tracking ID is identified as 3, 3, 4, 4 and 4 respectively. In the following example, the number of fund substitute items that appear the most times can be determined as 4 through election. In this case, it can be considered that the number of fund substitute items in the first frame image is inaccurate, and the number of fund substitute items in the first frame image can be updated to 4; when the numbers of fund substitute items with the same tracking ID in the first frame image, the second frame image, the third frame image, the fourth frame image and the fifth frame image are identified as 4, 4, 4, 4 and 3 respectively, it can be determined by election that the number of fund substitute items that appear the most times is 4. In this case, it can be considered that the number of fund substitute items in the first frame image is accurate, and the number of fund substitute items in the first frame image can be kept unchanged at 4.

[0127] It is understandable that the information on the funds substitute items in the target frame image may be inaccurate due to the influence of the brightness of the on-site lighting, the occlusion between the funds substitute items, the occlusion of the funds substitute items by the players during the game, the movement of the funds substitute items and other factors. In the disclosed embodiment, the funds substitute item information with the largest number of occurrences can be determined in the funds substitute item information of the N-frame game platform image, which can improve the accuracy of the funds substitute item information in the target frame image to a certain extent.

[0128] In some embodiments, the above-mentioned re-determining the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window may include:

[0129] For each of the N frames of game platform images within a sliding window, the attribute data of the same money substitute item in the recognition results of each frame of the N frames of game platform images are determined through target tracking; the attribute data with a confidence level greater than or equal to a confidence level threshold is determined from the attribute data as the target attribute data of the money substitute item; among the target attribute data of the same money substitute item, the target attribute data with the largest number of occurrences is determined; the money substitute item information in the target frame image is determined as: the target attribute data with the largest number of occurrences.

[0130] Here, the confidence threshold can be preset according to actual application requirements. For example, the confidence threshold can be 0.9 or 1. The implementation method of determining the attribute data with the most occurrences by election has been described in the above-mentioned contents and will not be repeated here.

[0131] It can be understood that in the embodiments of the present disclosure, attribute data with a confidence level greater than or equal to a confidence threshold is determined from the attribute data of N frames of game platform images, that is, the target attribute data is determined, thereby determining the information on funds substitute items in the target frame image based on the target attribute data that appears the most times, which can improve the accuracy of the information on funds substitute items in the target frame image to a certain extent.

[0132] In some embodiments, the above-mentioned re-determining the information of the funds substitute items in the target frame image based on the recognition results of the N frames of game platform images in the sliding window may include: for each of the N frames of game platform images in the sliding window, determining the attribute data of at least two funds substitute items in the recognition results of the N frames of game platform images; among the attribute data of the at least two funds substitute items, determining the attribute data of each of the at least two funds substitute items through the tracking identifiers of each of the at least two funds substitute items; for each of the at least two funds substitute items, determining the attribute data with the largest number of occurrences; and determining the information of the funds substitute items in the target frame image as: the attribute data with the largest number of occurrences determined for each of the funds substitute items.

[0133] In some embodiments, the attribute data includes the denomination of the money substitute item, and the tracking ids of different money substitute items are id-1 and id-2 respectively. Based on id-1, target tracking is performed in the recognition results of N frames of game platform images within the sliding window, so as to determine the attribute data of the money substitute item with the tracking id of id-1 in the recognition results of N frames of game platform images; then, in the attribute data of the money substitute item with the tracking id of id-1 corresponding to the N frames of game platform images, the attribute data with the most occurrences is determined by the above-described election method, and the information of the money substitute item with the tracking id of id-1 in the target frame image is determined as: the determined attribute data with the most occurrences. Similarly, for the money substitute item with the tracking id of id-2, the information of the money substitute item with the tracking id of id-2 in the target frame image can also be determined.

[0134] It can be understood that in the embodiments of the present disclosure, the attribute data of at least two fund substitute items in N frames of game platform images can be determined based on the tracking identifiers of the fund substitute items. Thus, for each of the at least two fund substitute items, the attribute data with the highest number of occurrences can be determined, thereby determining the fund substitute item information in the target frame image, which can improve the accuracy of the fund substitute item information in the target frame image to a certain extent.

[0135] In some embodiments, after the money substitute item information in the target frame image is determined, the business detection logic related to the money substitute item can be executed according to the money substitute item information in the target frame image.

[0136] In the embodiments of the present disclosure, the business detection logic related to the funds substitute items may be: determining whether the game on the current game platform allows a specific type of funds substitute items, determining whether the face value of the funds substitute items is less than the lower face value limit specified by the game, determining whether the face value of the funds substitute items is greater than the upper face value limit specified by the game, obtaining the funds substitute items information used by the owner of the funds substitute items in the game, etc.; it should be noted that the above-mentioned recorded content is merely an exemplary description of the business detection logic related to the funds substitute items, and the embodiments of the present disclosure are not limited to this.

[0137] In some embodiments, the above-mentioned execution of business detection logic related to the money substitute item based on the money substitute item information in the target frame image may include: determining a human hand detection frame and a money substitute item detection frame in the target frame image; when the human hand detection frame and the money substitute item detection frame do not overlap, executing business detection logic related to the money substitute item based on the money substitute item information in the target frame image.

[0138] In the disclosed embodiments, by performing hand detection and money substitute item detection on the target frame image, the hand detection frame and the money substitute item detection frame of the target frame image can be determined; in some embodiments, the target frame image frame can be respectively input into a first neural network for performing hand detection and a second neural network for performing money substitute item detection, and the target frame image is processed by the first neural network and the second neural network to obtain the hand detection frame and the money substitute item detection frame of the target frame image; the disclosed embodiments do not limit the network structure of the first neural network and the second neural network, for example, the first neural network and the second neural network can be SingleShot MultiBox Detector (SSD), You Only Look Once (You Only Look Once), Faster Region-Convolutional Neural Networks (Faster RCNN) or other neural networks based on deep learning.

[0139] In the disclosed embodiments, the degree of overlap between the human hand detection frame and the cash substitute item detection frame in the target frame image can be calculated based on the human hand detection frame and the cash substitute item detection frame in the target frame image. When the degree of overlap is greater than 0, it can be considered that the human hand detection frame and the cash substitute item detection frame in the target frame image overlap; when the degree of overlap is equal to 0, it can be considered that the human hand detection frame and the cash substitute item detection frame in the target frame image do not overlap.

[0140] In some embodiments, when there is overlap between the human hand detection frame and the money substitute item detection frame, the money substitute item information corresponding to the money substitute item detection frame can be obtained in the recognition result of the target frame image, that is, relevant business detection logic processing is not performed based on the money substitute item information corresponding to the money substitute item detection frame.

[0141] In an implementation method of executing business detection logic related to money substitute items, when the human hand detection frame and the money substitute item detection frame do not overlap, it can be determined that the corresponding money substitute item in the target frame image is an identifiable stable money substitute item. At this time, the type of money substitute items currently allowed by the game platform can be determined based on a pre-configured configuration file, and then, based on the type of identifiable stable money substitute item, it can be judged whether the game platform allows the identifiable stable money substitute item to play the game.

[0142] It can be seen that the embodiments of the present disclosure can execute the business detection logic related to the funds substitute items for the funds substitute items where there is no overlap between the human hand and the funds substitute items, thereby executing the business detection logic related to the funds substitute items when it is determined that the human hand does not block the funds substitute items, thereby reducing the probability of business detection logic errors caused by the blocking of the funds substitute items by the human hand to a certain extent.

[0143] In some embodiments, the change information of the fund replacement item information in multiple target frame images can be determined based on the re-determined fund replacement item information in multiple target frame images arranged in chronological order; and the above change information can be pushed to the management device of the game platform.

[0144] In actual applications, for money substitute items with the same tracking ID in multiple target frame images, information such as the quantity or denomination of the money substitute items may change. In order to facilitate management personnel to promptly understand the change information of the money substitute items, in the embodiment of the present disclosure, after determining that the information of the money substitute items has changed in multiple target frame images, the corresponding change information can be pushed, which is conducive to subsequent processing of the change information of the money substitute items on the management device side.

[0145] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.

[0146] Based on the image processing method proposed in the above-mentioned embodiment, the embodiment of the present disclosure proposes an image processing device.

[0147] Figure 3 FIG. 1 is a schematic diagram of the structure of the image processing device according to an embodiment of the present disclosure. Figure 3 As shown, the device may include:

[0148] A determination module 301 is used to determine a recognition result of each frame of a game platform image in a plurality of frames of game platform images, wherein the recognition result at least includes information of a fund replacement item;

[0149] The processing module 302 is used to re-determine the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window each time when the sliding window is used to determine N frames of game platform images in the multiple frames of game platform images, wherein the sliding order of the sliding window is the frame order of the multiple frames of game platform images, N is an integer greater than 1, and the target frame image is a frame image in the N frames of game platform images in the sliding window.

[0150] In some embodiments, the processing module 302 is used to re-determine the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window, including:

[0151] For each of the N frames of game platform images within the sliding window, determine the attribute data of the same money substitute item in the recognition result of each of the N frames of game platform images by target tracking;

[0152] Among the attribute data of the same money substitute item, determine the attribute data that appears the most times;

[0153] The money substitute item information in the target frame image is determined as: the attribute data with the most occurrences.

[0154] In some embodiments, the processing module 302 is used to re-determine the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window, including:

[0155] For each of the N frames of game platform images within the sliding window, determine the attribute data of the same money substitute item in the recognition result of each of the N frames of game platform images by target tracking;

[0156] Determine the attribute data whose confidence is greater than or equal to the confidence threshold from the attribute data as the target attribute data of the funds replacement item;

[0157] Among the target attribute data of the same fund replacement item, determine the target attribute data that appears the most times;

[0158] The fund substitute item information in the target frame image is determined as: the target attribute data with the largest number of appearances.

[0159] In some embodiments, the processing module 302 is used to re-determine the information of the money replacement item in the target frame image according to the recognition results of the N frames of game platform images in the sliding window, including:

[0160] For each of the N frames of game platform images in the sliding window, determine the attribute data of at least two fund replacement items in the recognition results of the N frames of game platform images;

[0161] Among the attribute data of the at least two money substitute items, the attribute data of each of the at least two money substitute items is determined by using the tracking identifiers of each of the at least two money substitute items;

[0162] For each of the at least two money substitute items, determining the attribute data that appears most frequently;

[0163] The money substitute item information in the target frame image is determined as: the attribute data with the largest number of occurrences determined for each of the money substitute items.

[0164] In some embodiments, the attribute data of the fund substitute item includes at least one of the following: the denomination of the fund substitute item, the type of the fund substitute item, the quantity of the fund substitute item, and the information of the owner of the fund substitute item.

[0165] In some embodiments, the N frames of game platform images are N consecutive frames of images in a plurality of frames of game platform images.

[0166] In some embodiments, the target frame image is the earliest frame image captured among the N frames of game platform images.

[0167] In some embodiments, the processing module 302 is further configured to execute business detection logic related to the money substitute item according to the money substitute item information in the target frame image.

[0168] In some embodiments, the processing module 302 is used to execute business detection logic related to the money substitute item according to the money substitute item information in the target frame image, including:

[0169] Determine a human hand detection frame and a money substitute object detection frame in the target frame image;

[0170] In the case where the human hand detection frame does not overlap with the money substitute item detection frame, the business detection logic related to the money substitute item is executed according to the money substitute item information in the target frame image.

[0171] In some embodiments, the processing module 302 is used to execute business detection logic related to the money substitute item according to the money substitute item information in the target frame image, including:

[0172] Determine change information of the money substitute item information in the multiple target frame images according to the re-determined money substitute item information in the multiple target frame images arranged in chronological order;

[0173] The change information is pushed to the management device of the game platform.

[0174] In some embodiments, the determination module 301 is used to determine the recognition result of each frame of the game platform image in the multiple frames of the game platform image, including:

[0175] The target object in each frame of the image is identified, and the identified target object is mapped to a predetermined area division map to obtain a recognition result for each preset area of ​​each frame of the image.

[0176] In practical applications, both the determination module 301 and the processing module 302 can be implemented using a processor in an edge computing device, and the processor can be at least one of an ASIC, a DSP, a DSPD, a PLD, a FPGA, a CPU, a controller, a microcontroller, and a microprocessor.

[0177] In addition, each functional module in this embodiment can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or software functional modules.

[0178] If the integrated unit is implemented in the form of a software function module and is not sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this embodiment is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) or a processor to perform all or part of the steps of the method described in this embodiment. The aforementioned storage medium includes: U disk, mobile hard disk, read only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0179] Specifically, a computer program instruction corresponding to an image processing method in this embodiment can be stored on a storage medium such as a CD, a hard disk, or a USB flash drive. When the computer program instruction corresponding to an image processing method in the storage medium is read or executed by an electronic device, any image processing method in the aforementioned embodiments is implemented.

[0180] Based on the same technical concept as the aforementioned embodiment, the embodiment of the present disclosure also provides an edge computing device, which is used to receive a multi-frame game platform image sent by an image acquisition device; the multi-frame game platform image is an image acquired by the image acquisition device.

[0181] See also Figure 4 , which shows an edge computing device 4 provided by an embodiment of the present disclosure, which may include: a memory 401 and a processor 402; wherein,

[0182] The memory 401 is used to store computer programs and data;

[0183] The processor 402 is configured to execute the computer program stored in the memory to implement any one of the image processing methods in the foregoing embodiments.

[0184] In practical applications, the memory 401 may be a volatile memory, such as RAM; or a non-volatile memory, such as ROM, flash memory, hard disk drive (HDD) or solid-state drive (SSD); or a combination of the above types of memory, and provide instructions and data to the processor 402.

[0185] The processor 402 may be at least one of ASIC, DSP, DSPD, PLD, FPGA, CPU, controller, microcontroller, and microprocessor. It is understandable that for different devices, the electronic device used to implement the processor function may also be other, which is not specifically limited in the embodiments of the present disclosure.

[0186] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0187] The above description of the various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other. For the sake of brevity, this article will not repeat them.

[0188] The methods disclosed in the various method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0189] The features disclosed in the various product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0190] The features disclosed in the various method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0191] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0192] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. An image processing method, the method comprising: Determine a recognition result of each frame of the game platform image in the plurality of frames of the game platform image, wherein the recognition result at least includes information of a fund replacement item; In the case where N frames of the game platform image are determined each time by using a sliding window, the fund replacement item information in the target frame image is re-determined according to the recognition results of the N frames of the game platform image in the sliding window, wherein the sliding order of the sliding window is the frame order of the multiple frames of the game platform image, N is an integer greater than 1, and the target frame image is one frame image of the N frames of the game platform image in the sliding window; The step of determining the recognition result of each frame of the game platform image in the multiple frames of the game platform image includes: Identify the target object in each frame of the image, map the identified target object to a predetermined region division map, and obtain a recognition result of each preset region of each frame of the image; the region division map is used to indicate the correct region where the target object is located; The method further comprises: After determining the money substitute item information in the target frame image, executing the business detection logic related to the money substitute item according to the money substitute item information in the target frame image; Among them, the business detection logic related to the funds substitute items at least includes: determining whether the game on the current game platform allows a specific type of funds substitute items, determining whether the face value of the funds substitute items is less than the lower face value limit specified by the game, determining whether the face value of the funds substitute items is greater than the upper face value limit specified by the game, and obtaining the funds substitute items information used by the owner of the funds substitute items in the game.

2. The method according to claim 1, wherein: The method of re-determining the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window includes: For each of the N frames of game platform images within the sliding window, determine the attribute data of the same money substitute item in the recognition result of each of the N frames of game platform images by target tracking; Among the attribute data of the same money substitute item, determine the attribute data that appears the most times; The money substitute item information in the target frame image is determined as: the attribute data with the largest number of appearances.

3. The method according to claim 1, wherein: The method of re-determining the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window includes: For each of the N frames of game platform images within the sliding window, determine the attribute data of the same money substitute item in the recognition result of each of the N frames of game platform images by target tracking; Determine the attribute data whose confidence is greater than or equal to the confidence threshold from the attribute data as the target attribute data of the funds replacement item; Among the target attribute data of the same fund replacement item, determine the target attribute data that appears the most times; The fund substitute item information in the target frame image is determined as: the target attribute data with the largest number of appearances.

4. The method according to claim 1, wherein: The method of re-determining the fund replacement item information in the target frame image according to the recognition results of the N frames of game platform images in the sliding window includes: For each of the N frames of game platform images in the sliding window, determine the attribute data of at least two fund replacement items in the recognition results of the N frames of game platform images; Among the attribute data of the at least two money substitute items, the attribute data of each of the at least two money substitute items is determined by using the tracking identifiers of each of the at least two money substitute items; For each of the at least two money substitute items, determining the attribute data that appears most frequently; The money substitute item information in the target frame image is determined as: the attribute data with the largest number of occurrences determined for each of the money substitute items.

5. The method according to any one of claims 2 to 4, wherein: The attribute data of the fund substitute item includes at least one of the following: the denomination of the fund substitute item, the type of the fund substitute item, the quantity of the fund substitute item, and the information of the owner of the fund substitute item.

6. The method according to claim 1, wherein: The N frames of game platform images are N consecutive frames of images in the multiple frames of game platform images.

7. The method according to claim 1, wherein: The target frame image is the earliest frame image captured among the N frames of game platform images.

8. The method according to claim 1, wherein: The executing of the business detection logic related to the money substitute item according to the money substitute item information in the target frame image includes: Determine a human hand detection frame and a money substitute item detection frame in the target frame image; In the case where the human hand detection frame does not overlap with the money substitute item detection frame, the business detection logic related to the money substitute item is executed according to the money substitute item information in the target frame image.

9. The method according to claim 1, characterized in that: The executing of the business detection logic related to the money substitute item according to the money substitute item information in the target frame image includes: Determine change information of the money substitute item information in the plurality of target frame images arranged in chronological order according to the re-determined money substitute item information in the plurality of target frame images; The change information is pushed to the management device of the game platform.

10. An image processing device, comprising: A determination module, used to determine the recognition result of each frame of the game platform image in the multiple frames of the game platform image, wherein the recognition result at least includes the information of the fund replacement item; A processing module, configured to re-determine the information of the money substitute item in the target frame image according to the recognition results of the N frames of the game platform image in the sliding window when the sliding window is used to determine N frames of the game platform image in the multiple frames of the game platform image each time, wherein the sliding order of the sliding window is the frame order of the multiple frames of the game platform image, N is an integer greater than 1, and the target frame image is one frame image in the N frames of the game platform image in the sliding window; The determination module is further used to identify the target object in each frame of the image, and map the identified target object to a predetermined region division map to obtain a recognition result of each preset region of each frame of the image; the region division map is used to indicate the correct region where the target object is located; The processing module is further used to execute business detection logic related to the money substitute items according to the money substitute item information in the target frame image after determining the money substitute item information in the target frame image, wherein the business detection logic related to the money substitute items at least includes: determining whether the game on the current game platform allows a specific type of money substitute item, determining whether the face value of the money substitute item is less than the lower face value limit specified by the game, determining whether the face value of the money substitute item is greater than the upper face value limit specified by the game, and obtaining the money substitute item information used by the owner of the money substitute item in the game.

11. An edge computing device, characterized in that: The edge computing device is used to receive a plurality of frames of game platform images sent by an image acquisition device; the plurality of frames of game platform images are images acquired by the image acquisition device; The edge computing device includes a processor and a memory for storing a computer program that can be run on the processor; wherein, The processor is configured to execute the computer program to: Determine a recognition result of each frame of the game platform image in the plurality of frames of the game platform image, wherein the recognition result at least includes information of a fund replacement item; In the case where N frames of the game platform image are determined each time by using a sliding window, the fund replacement item information in the target frame image is re-determined according to the recognition results of the N frames of the game platform image in the sliding window, wherein the sliding order of the sliding window is the frame order of the multiple frames of the game platform image, N is an integer greater than 1, and the target frame image is one frame image of the N frames of the game platform image in the sliding window; After determining the money substitute item information in the target frame image, executing the business detection logic related to the money substitute item according to the money substitute item information in the target frame image; The business detection logic related to the money substitute items at least includes: determining whether the game on the current game platform allows a specific type of money substitute item, determining whether the face value of the money substitute item is less than the lower face value limit specified by the game, determining whether the face value of the money substitute item is greater than the upper face value limit specified by the game, and obtaining the money substitute item information used by the owner of the money substitute item in the game; The step of determining the recognition result of each frame of the game platform image in the multiple frames of the game platform image includes: Identify the target object in each frame of the image, map the identified target object to a predetermined area division map, and obtain the recognition result of each preset area of ​​each frame of the image; the area division map is used to indicate the correct area where the target object is located.

12. The edge computing device according to claim 11, wherein: The processor is configured as follows: For each of the N frames of game platform images within the sliding window, determine the attribute data of the same money substitute item in the recognition result of each of the N frames of game platform images by target tracking; Among the attribute data of the same money substitute item, determine the attribute data that appears the most times; The money substitute item information in the target frame image is determined as: the attribute data with the largest number of appearances.

13. The edge computing device according to claim 11, wherein: The processor is configured as follows: For each of the N frames of game platform images within the sliding window, determine the attribute data of the same money substitute item in the recognition result of each of the N frames of game platform images by target tracking; Determine the attribute data whose confidence is greater than or equal to the confidence threshold from the attribute data as the target attribute data of the funds replacement item; Among the target attribute data of the same fund replacement item, determine the target attribute data that appears the most times; The fund substitute item information in the target frame image is determined as: the target attribute data with the largest number of appearances.

14. The edge computing device according to claim 11, wherein: The processor is configured as follows: For each of the N frames of game platform images in the sliding window, determine the attribute data of at least two fund replacement items in the recognition results of the N frames of game platform images; Among the attribute data of the at least two money substitute items, the attribute data of each of the at least two money substitute items is determined by using the tracking identifiers of each of the at least two money substitute items; For each of the at least two money substitute items, determining the attribute data that appears most frequently; The money substitute item information in the target frame image is determined as: the attribute data with the largest number of occurrences determined for each of the money substitute items.

15. The edge computing device according to any one of claims 12 to 14, wherein: The attribute data of the fund substitute item includes at least one of the following: the denomination of the fund substitute item, the type of the fund substitute item, the quantity of the fund substitute item, and the information of the owner of the fund substitute item.

16. The edge computing device according to any one of claims 11 to 14, wherein: The N frames of game platform images are N consecutive frames of images in the multiple frames of game platform images.

17. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the image processing method according to any one of claims 1 to 9 is implemented.

Images