Target identification monitoring feedback adjusting system and method

Abstract

The invention relates to the technical field of computer vision and intelligent monitoring, in particular to a target recognition monitoring feedback adjustment system, which is characterized in that a target recognition module performs real-time target detection on an image picture by using a deep learning algorithm to obtain corresponding position and category information of various targets; the environment adaptation module detects the environment condition of the image picture in real time, and dynamically adjusts the parameter or algorithm of the target recognition module; the feedback monitoring module compares the recognition result with a preset standard and judges whether the recognition result has an error or not; and when the recognition result has an error, the self-adjustment module analyzes the cause of the error, and automatically adjusts the parameter or algorithm of the target recognition module. It can be understood that according to the technical scheme shown by the invention, the target recognition accuracy is improved through feedback adjustment, and therefore the user experience is improved; and through self-adjustment and optimization, the system can continuously learn and improve.

Description

Technical Field

[0001] The present invention relates to the technical field of computer vision and intelligent monitoring, and in particular to a target recognition monitoring feedback adjustment system and method. Background Art

[0002] With the development of videos and games, the complexity of the environments and plots in videos and games is increasing, thus placing higher demands on target recognition in videos and games.

[0003] Taking the game scene as an example, accurately and quickly identifying and responding to various targets in the game (including characters, objects, environments, etc.) is crucial to improving the player experience. However, there are many problems in the existing technology. For example, games and videos may have target occlusion, lighting changes, dynamic scene changes, etc., and these scene changes will affect the target recognition accuracy of the existing technology, resulting in insufficient target recognition accuracy and reduced efficiency. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a target recognition monitoring feedback adjustment system and method to solve the problem of insufficient accuracy and reduced efficiency of target recognition due to changes in environmental scenes in the prior art.

[0005] According to a first aspect of an embodiment of the present invention, a target recognition monitoring feedback adjustment system is provided, comprising:

[0006] The target recognition module is used to perform real-time target detection on the image screen using a deep learning algorithm, identify and classify various targets from the image screen, and obtain the corresponding position and category information of various targets;

[0007] An environmental adaptation module, used to detect the environmental conditions of the image in real time and dynamically adjust the parameters or algorithms of the target recognition module according to the environmental conditions;

[0008] A feedback monitoring module, used to compare the recognition result obtained by the target recognition module with a preset standard to determine whether there is an error in the recognition result;

[0009] The self-adjustment module is used to analyze the cause of the error and automatically adjust the parameters or algorithm of the target recognition module when the feedback monitoring module determines that the recognition result is erroneous.

[0010] Preferably, the target recognition monitoring feedback adjustment system further includes:

[0011] A user interaction module, used to obtain user feedback information input by the user, and obtain error data of the recognition result according to the user feedback information;

[0012] The self-adjustment module is used to automatically adjust the parameters or algorithms of the target recognition module according to the error data fed back by the user.

[0013] Preferably, the target recognition monitoring feedback adjustment system further includes:

[0014] The data analysis and reporting module is used to obtain the operating data of the target recognition monitoring feedback adjustment system through the feedback monitoring module, perform statistical analysis on the operating data, and generate a periodic performance report.

[0015] Preferably, the user interaction module is further used to obtain user feedback suggestions input by the user;

[0016] The data analysis report module is also used to send the user feedback suggestions to the developer team according to a preset developer feedback channel.

[0017] Preferably, if the environment adaptation module detects that the environment condition of the image screen is insufficiently illuminated, the image exposure parameters of the target recognition module are dynamically adjusted.

[0018] Preferably, if the environment adaptation module detects that the environment of the image is a complex background, the resolution of the detection algorithm of the target recognition module is dynamically adjusted.

[0019] Preferably, when the self-adjusting module is performing error analysis, if it is detected that the recognition result has recognition errors for a preset number of consecutive times for a preset category of target, the detection algorithm of the target recognition module is automatically adjusted to improve the sensitivity to the preset category of target.

[0020] According to a second aspect of an embodiment of the present invention, a target recognition monitoring feedback adjustment method is provided, comprising:

[0021] Using the deep learning algorithm in the target recognition module to perform real-time target detection on the image screen, identify and classify various targets from the image screen, and obtain the corresponding position and category information of various targets;

[0022] When the target recognition module performs target recognition, the environment of the image screen is detected in real time, and the parameters or algorithms of the target recognition module are dynamically adjusted according to the environment;

[0023] Compare the recognition result obtained by the target recognition module with a preset standard to determine whether an error occurs in the recognition result;

[0024] If it is determined that an error occurs in the recognition result, the cause of the error is analyzed and the parameters or algorithm of the target recognition module are automatically adjusted.

[0025] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0026] It is understandable that the technical solution shown in the present invention uses a deep learning algorithm to perform real-time target detection on the image screen through the target recognition module to obtain the corresponding position and category information of various targets; the environmental adaptation module detects the environmental conditions of the image screen in real time and dynamically adjusts the parameters or algorithms of the target recognition module; the feedback monitoring module compares the recognition result with the preset standard to determine whether the recognition result is wrong; when the recognition result is wrong, the self-adjustment module analyzes the cause of the error and automatically adjusts the parameters or algorithms of the target recognition module. It is understandable that the technical solution shown in the present invention improves the accuracy of target recognition through feedback adjustment, thereby improving the user experience; and through self-adjustment and optimization, the system can continue to learn and improve.

[0027] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0029] Figure 1 is a schematic block diagram of a target recognition monitoring feedback adjustment system according to an exemplary embodiment;

[0030] Figure 2 The present invention is a schematic diagram of the operation flow of a target recognition monitoring feedback adjustment system according to an exemplary embodiment. DETAILED DESCRIPTION

[0031] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0032] In one embodiment, Figure 1 is a schematic block diagram of a target recognition monitoring feedback adjustment system according to an exemplary embodiment. Figure 1 , providing a target recognition monitoring feedback adjustment system, comprising:

[0033] The target recognition module is used to perform real-time target detection on the image screen using a deep learning algorithm, identify and classify various targets from the image screen, and obtain the corresponding position and category information of various targets.

[0034] The environment adaptation module is used to detect the environment conditions of the image picture in real time and dynamically adjust the parameters or algorithms of the target recognition module according to the environment conditions.

[0035] The feedback monitoring module is used to compare the recognition result obtained by the target recognition module with a preset standard to determine whether there is an error in the recognition result.

[0036] The self-adjustment module is used to analyze the cause of the error and automatically adjust the parameters or algorithm of the target recognition module when the feedback monitoring module determines that the recognition result is erroneous.

[0037] In specific practice, Figure 2 is a schematic diagram of an operation flow of a target recognition monitoring feedback adjustment system according to an exemplary embodiment. Figure 2 Taking the game scene as an example, as the games are becoming more and more exquisite, the use of various backgrounds and light and shadow effects has brought challenges to the target detection and recognition technology. The target recognition module of this embodiment can detect the target in the picture in real time using a deep learning algorithm (such as YOLOv8) while the user is playing the game, identify and classify various targets, such as people, objects, etc., and finally derive the category of the target and its location.

[0038] During the real-time recognition process of the target recognition module, if the game scene has changes in light and shadow, complex background, etc., it will affect the accuracy of the deep learning algorithm detection. Therefore, the environmental adaptation module can dynamically adjust the parameters or algorithm of the target recognition module according to the changes in the game scene.

[0039] It should be noted that if the environment adaptation module detects that the environment condition of the image screen is insufficiently illuminated, the image exposure parameters of the target recognition module are dynamically adjusted.

[0040] It should be noted that if the environment adaptation module detects that the background of the image is complex, the resolution of the detection algorithm of the target recognition module is dynamically adjusted.

[0041] The above are just two common examples. How to adjust the parameters and algorithms needs to be configured according to the actual scene of the image.

[0042] It is understandable that the environment adaptation module dynamically adjusts the parameters or algorithms of the target recognition module according to changes in the game environment to ensure that a high recognition rate is maintained under various conditions.

[0043] With the cooperation of the target recognition module and the environment adaptation module, the success rate of target recognition in the image is already very high, but there are still some recognition errors. Therefore, this embodiment introduces a feedback mechanism to continuously adjust the parameters or algorithms of the target recognition module when recognition errors occur.

[0044] The feedback monitoring module can monitor the results of target recognition and evaluate its accuracy. By comparing the recognition results with the preset standard or ground truth data, the feedback monitoring module can detect the recognition errors or omissions and send them to the self-adjustment module.

[0045] The self-tuning module automatically adjusts the parameters or algorithms of the target recognition module based on the data provided by the feedback monitoring module to correct errors and optimize performance. This includes the feedback loop in machine learning, which enables the system to learn from its mistakes and improve.

[0046] It is understandable that, through the cooperation of the feedback monitoring module and the self-adjustment module, when the target is correctly or incorrectly identified, the feedback monitoring module records this event and transmits the incorrect identification result to the self-adjustment module. The self-adjustment module analyzes the cause of the incorrect identification and adjusts the identification strategy to reduce future errors. For example, when the self-adjustment module performs error analysis, if it is detected that the identification result has a preset number of consecutive identification errors for a preset category target, the detection algorithm of the target identification module is automatically adjusted to increase the sensitivity to the preset category target.

[0047] It is understandable that the technical solution shown in this embodiment uses a deep learning algorithm to perform real-time target detection on the image screen through the target recognition module to obtain the corresponding position and category information of various targets; the environmental adaptation module detects the environmental conditions of the image screen in real time and dynamically adjusts the parameters or algorithms of the target recognition module; the feedback monitoring module compares the recognition result with the preset standard to determine whether the recognition result is wrong; when the recognition result is wrong, the self-adjustment module analyzes the cause of the error and automatically adjusts the parameters or algorithms of the target recognition module. It is understandable that the technical solution shown in this embodiment improves the accuracy of target recognition through feedback adjustment, thereby improving the user experience; and through self-adjustment and optimization, the system can continue to learn and improve.

[0048] It should be noted that the target recognition monitoring feedback adjustment system also includes: a user interaction module, which is used to obtain user feedback information input by the user, and derive the error data of the recognition result based on the user feedback information; and a self-adjustment module, which is used to automatically adjust the parameters or algorithms of the target recognition module based on the error data fed back by the user.

[0049] In specific practice, through the cooperation of the target recognition module, the environment adaptation module, the feedback monitoring module and the self-adjustment module, extremely accurate target recognition has been achieved. Even if mistakes occur occasionally, feedback adjustment can be used to reduce the subsequent recognition errors. However, some errors still need to be discovered manually. Taking games as an example, if the user finds that there are continuous recognition errors in a certain place during the game, the user can submit user feedback information through the port provided by the user interaction module. The user interaction module obtains the error data of the recognition result based on the user feedback information, and the self-adjustment module automatically adjusts the parameters or algorithms of the target recognition module based on the error data fed back by the user.

[0050] It should be noted that the user interaction module is also used to obtain user feedback suggestions input by the user; the data analysis report module is also used to send the user feedback suggestions to the developer team according to the preset developer feedback channel.

[0051] In specific practice, users can provide their own feedback suggestions through the user interaction module, and the data analysis report module can send the user feedback suggestions to the developer team according to the preset developer feedback channel.

[0052] It is understood that the user interaction module provides a channel for users to report recognition errors or provide other feedback. This information is used to further train and optimize the object recognition module.

[0053] It should be noted that the data analysis and reporting module is also used to obtain the operating data of the target recognition monitoring feedback adjustment system through the feedback monitoring module, perform statistical analysis on the operating data, and generate periodic performance reports.

[0054] It is understandable that the data analysis report module regularly summarizes all monitoring data and user feedback, and reveals system performance trends and potential problems through data analysis. The development team can use these reports to implement larger-scale system updates and optimizations.

[0055] Through the cooperation of the above modules, the following effects can be achieved:

[0056] Consider a combat game scenario where the player needs to engage multiple enemies. The target recognition module in the system needs to not only recognize the player's character and the different types of enemies, but also distinguish between various weapons and props. The environmental adaptation module may adjust the visual recognition algorithm based on weather effects (such as rain or fog) to maintain accuracy.

[0057] Suppose in a certain level, the player finds that the enemy is often misidentified as part of the environment. The player can report this problem through the user interaction module. The feedback monitoring module records the frequency of this problem, and the self-adjustment module analyzes the cause of the error, which may be insufficient contrast or the color of the enemy's equipment is too similar to the environment. The system then automatically adjusts the recognition parameters to increase sensitivity to these situations.

[0058] Finally, the data analysis reporting module aggregates all this data and generates reports to guide the development team to make corresponding adjustments in future game updates.

[0059] In summary, it can be understood that the target recognition monitoring feedback adjustment system provided by this application can improve the accuracy and response speed of target recognition, thereby enhancing the player's gaming experience. Through self-adjustment and optimization, the system can continuously learn and improve to adapt to the ever-changing gaming environment and player needs. Strengthen the interaction between players and games, allowing players to directly participate in the optimization process of the game. Provide detailed performance reports and analysis to help development teams identify problems and improve games more effectively.

[0060] According to a second aspect of an embodiment of the present invention, a target recognition monitoring feedback adjustment method is provided, comprising:

[0061] The deep learning algorithm in the target recognition module is used to perform real-time target detection on the image screen, identify and classify various targets from the image screen, and obtain the corresponding position and category information of various targets. When the target recognition module performs target recognition, the environmental conditions of the image screen are detected in real time, and the parameters or algorithms of the target recognition module are dynamically adjusted according to the environmental conditions. The recognition result obtained by the target recognition module is compared with the preset standard to determine whether the recognition result is wrong. If it is determined that the recognition result is wrong, the cause of the error is analyzed and the parameters or algorithms of the target recognition module are automatically adjusted.

[0062] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0063] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0064] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0065] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0066] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0067] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0068] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0069] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0070] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A target recognition monitoring feedback adjustment system, characterized in that: include: The target recognition module is used to perform real-time target detection on the image screen using a deep learning algorithm, identify and classify various targets from the image screen, and obtain the corresponding position and category information of various targets; An environmental adaptation module, used to detect the environmental conditions of the image in real time and dynamically adjust the parameters or algorithms of the target recognition module according to the environmental conditions; A feedback monitoring module, used to compare the recognition result obtained by the target recognition module with a preset standard to determine whether there is an error in the recognition result; The self-adjustment module is used to analyze the cause of the error and automatically adjust the parameters or algorithm of the target recognition module when the feedback monitoring module determines that the recognition result is erroneous.

2. The target recognition monitoring feedback adjustment system according to claim 1, characterized in that: Also includes: A user interaction module, used to obtain user feedback information input by the user, and obtain error data of the recognition result according to the user feedback information; The self-adjustment module is used to automatically adjust the parameters or algorithms of the target recognition module according to the error data fed back by the user.

3. The target recognition monitoring feedback adjustment system according to claim 2, characterized in that: Also includes: The data analysis and reporting module is used to obtain the operating data of the target recognition monitoring feedback adjustment system through the feedback monitoring module, perform statistical analysis on the operating data, and generate a periodic performance report.

4. The target recognition monitoring feedback adjustment system according to claim 3, characterized in that: The user interaction module is also used to obtain user feedback suggestions input by the user; The data analysis report module is also used to send the user feedback suggestions to the developer team according to a preset developer feedback channel.

5. The target recognition monitoring feedback adjustment system according to claim 1, characterized in that: If the environment adaptation module detects that the environment condition of the image screen is insufficiently illuminated, the image exposure parameter of the target recognition module is dynamically adjusted.

6. The target recognition monitoring feedback adjustment system according to claim 1, characterized in that: If the environment adaptation module detects that the environment of the image is complex, the resolution of the detection algorithm of the target recognition module is dynamically adjusted.

7. The target recognition monitoring feedback adjustment system according to claim 1, characterized in that: When the self-adjusting module is performing error analysis, if it is detected that the recognition result has recognition errors for a preset number of consecutive times for a preset category of target, the detection algorithm of the target recognition module is automatically adjusted to improve the sensitivity to the preset category of target.

8. A target recognition monitoring feedback adjustment method, characterized in that: include: Using the deep learning algorithm in the target recognition module to perform real-time target detection on the image screen, identify and classify various targets from the image screen, and obtain the corresponding position and category information of various targets; When the target recognition module performs target recognition, the environment of the image screen is detected in real time, and the parameters or algorithms of the target recognition module are dynamically adjusted according to the environment; Compare the recognition result obtained by the target recognition module with a preset standard to determine whether an error occurs in the recognition result; If it is determined that an error occurs in the recognition result, the cause of the error is analyzed and the parameters or algorithm of the target recognition module are automatically adjusted.

Images