Intelligent Control Method, System and Device for Micro-needle Eyebrow Implantation

Abstract

The present invention discloses an intelligent control method, system and device for micro-needle eyebrow implantation, which relates to the technical field of micro-needle eyebrow implantation control. The method includes the following steps: collecting and preprocessing multi-modal original data for micro-needle eyebrow implantation, quantifying and evaluating to determine the value of the comparison database, and optimizing the intelligent control method for micro-needle eyebrow implantation. By collecting and processing the multi-modal original data for micro-needle eyebrow implantation, the present invention respectively quantifies and evaluates the recognition timeliness of the multi-modal data for micro-needle eyebrow implantation and the data fusion performance of the multi-modal data for micro-needle eyebrow implantation through the multi-modal data for micro-needle eyebrow implantation. Then, the quantified value of the data recognition timeliness and the quantified value of the data fusion performance obtained are respectively compared and analyzed with the determination value in the database. Finally, the intelligent control method for micro-needle eyebrow implantation is optimized and adjusted, improving the accuracy of the intelligent control method for micro-needle eyebrow implantation and solving the problem of insufficient accuracy in the prior art due to the differences in data timeliness and data fusion in the intelligent control method for micro-needle eyebrow implantation.

Description

Technical Field

[0001] The present invention relates to the field of microneedle eyebrow transplantation control technology, and in particular to an intelligent control method, system and device for microneedle eyebrow transplantation. Background Art

[0002] Traditional eyebrow transplantation techniques, due to their lack of precision, result in unnatural post-operative results. Microneedle eyebrow transplantation, with its minimally invasive and high-precision features, is gaining market favor, meeting consumers' high demands for natural results. With the rapid development of artificial intelligence (AI) technology, the hair care industry is undergoing a transformation from traditional to intelligent hair transplantation. For example, Damai Microneedle Hair Transplantation automates hair follicle extraction and transplantation through an AI-powered hair transplantation robot, improving surgical efficiency and significantly reducing human error. Through intelligent technology, microneedle eyebrow transplantation significantly shortens post-operative recovery time and enhances the naturalness of the transplanted eyebrow results.

[0003] The existing intelligent control method of microneedle eyebrow transplantation uses an arrayed microneedle system, which can create multiple microholes at the same time to improve efficiency; through computer vision technology, the eyebrow transplantation process can be monitored in real time, and the distribution and density of hair follicles, as well as the implantation position and depth of microneedles can be analyzed through image processing algorithms.

[0004] For example, the medical device control device and medical device disclosed in the invention patent application with publication number CN110941338A, the control device receives external input information and is connected to the controller, the control device is divided into a first control area and a second control area, generates a first control signal according to the external input information received by the first control area; and generates feedback information including position information according to the external input information received by the second control area.

[0005] However, in the process of implementing the technical solutions of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0006] Due to insufficient feature extraction and insufficient fusion of multimodal data, errors in positioning the boundaries of microcirculatory obstruction areas occur, and the axial resolution limitations of OCT imaging cannot be compensated, resulting in the inability to distinguish subtle differences in collagen fiber arrangement density; when the scar thickness suddenly changes, the response delay of the software-controlled robotic arm is greater than the preset threshold, resulting in the actual measurement value of the puncture depth exceeding the allowable error range, and the inability to process sensor data in real time and generate control instructions, resulting in the inability to accurately predict the output behavior of the pump. There is a problem of insufficient accuracy in the intelligent control method of microneedle eyebrow transplantation due to differences in data timeliness and data fusion. Summary of the Invention

[0007] Embodiments of the present application provide an intelligent control method, system and device for micro-needle eyebrow implantation, which solve the problem in the prior art that the accuracy of the intelligent control method for micro-needle eyebrow implantation is insufficient due to differences in data timeliness and data fusion, and improve the accuracy of the intelligent control method for micro-needle eyebrow implantation.

[0008] Embodiments of the present application provide an intelligent control method for micro-needle eyebrow implantation, including the following steps: collecting multi-modal micro-needle eyebrow implantation raw data, preprocessing the multi-modal micro-needle eyebrow implantation raw data to obtain multi-modal micro-needle eyebrow implantation data; quantitatively evaluating the data recognition timeliness and the data fusion performance of the multi-modal micro-needle eyebrow implantation data respectively through the multi-modal micro-needle eyebrow implantation data to obtain a data recognition timeliness quantization value and a data fusion performance quantization value; comparing and analyzing the data recognition timeliness quantization value and the data fusion performance quantization value with the data recognition timeliness determination value and the data fusion performance determination value in the database respectively to obtain an intelligent control optimization method for micro-needle eyebrow implantation.

[0009] Embodiments of the present application provide an intelligent control system for micro-needle eyebrow implantation, including a multi-modal micro-needle eyebrow implantation data acquisition and processing module, a multi-modal micro-needle eyebrow implantation data analysis module and an intelligent control optimization module for micro-needle eyebrow implantation: The multi-modal micro-needle eyebrow implantation data acquisition and processing module: is used to collect multi-modal micro-needle eyebrow implantation raw data and preprocess the multi-modal micro-needle eyebrow implantation raw data to obtain multi-modal micro-needle eyebrow implantation data; The multi-modal micro-needle eyebrow implantation data analysis module: is used to quantitatively evaluate the data recognition timeliness and the data fusion performance of the multi-modal micro-needle eyebrow implantation data respectively through the multi-modal micro-needle eyebrow implantation data to obtain a data recognition timeliness quantization value and a data fusion performance quantization value; The intelligent control optimization module for micro-needle eyebrow implantation: is used to compare and analyze the data recognition timeliness quantization value and the data fusion performance quantization value with the data recognition timeliness determination value and the data fusion performance determination value in the database respectively to obtain an intelligent control optimization method for micro-needle eyebrow implantation.

[0010] Embodiments of the present application provide an intelligent control device for micro-needle eyebrow implantation. The device has one or more programs, and the one or more programs are executed by one or more processors to implement the above method.

[0011] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0012] 1. By collecting and processing the original data of multi-modal micro-needle eyebrow implantation, then quantitatively evaluating the timeliness of multi-modal micro-needle eyebrow implantation data recognition and the fusion performance of multi-modal micro-needle eyebrow implantation data respectively through the multi-modal micro-needle eyebrow implantation data. Next, compare and analyze the quantitatively evaluated values of data recognition timeliness and data fusion performance with the determination values in the database respectively. Finally, optimize and adjust the intelligent control method of micro-needle eyebrow implantation, improving the accuracy of the intelligent control method of micro-needle eyebrow implantation and solving the problem of insufficient accuracy in the existing intelligent control method of micro-needle eyebrow implantation due to differences in data timeliness and data fusion.

[0013] 2. By quantitatively evaluating the timeliness of multi-modal micro-needle eyebrow implantation data recognition and the fusion performance of multi-modal micro-needle eyebrow implantation data respectively through the multi-modal micro-needle eyebrow implantation data, by fusing different modalities of data such as visual and force sense data, etc., more comprehensive and accurate information can be obtained, helping the operator or the automatic control system to make better decisions, improving the overall quality of eyebrow implantation, and thus reducing data conflicts and errors, and enhancing the stability and reliability of the system.

[0014] 3. By comparing and analyzing the quantitatively evaluated values of data recognition timeliness and data fusion performance with the determination values in the database respectively, optimize and adjust the intelligent control method of micro-needle eyebrow implantation, the system's ability to fuse force sense and visual data, and accurately control the micro-needle puncture depth and angle, optimize the data fusion process, reduce redundant operations, improve data processing efficiency, and thus enhance the efficiency of the overall operation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flowchart of the intelligent control method for micro-needle eyebrow implantation provided by the embodiment of the present application;

[0016] Figure 2 It is a mind map of the data recognition timeliness feature extraction optimization method of the intelligent control method for micro-needle eyebrow implantation provided by the embodiment of the present application;

[0017] Figure 3 It is a mind map of the data fusion performance optimization method of the intelligent control method for micro-needle eyebrow implantation provided by the embodiment of the present application;

[0018] Figure 4 It is a schematic structural diagram of the intelligent control system for micro-needle eyebrow implantation provided by the embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] Embodiments of the present application provide an intelligent control method, system, and device for micro-needle eyebrow implantation, which solve the problem in the prior art that the accuracy of the intelligent control method for micro-needle eyebrow implantation is insufficient due to differences in data timeliness and data fusion. By preprocessing the collected multi-modal micro-needle implantation original data, such as denoising and filtering, multi-modal micro-needle implantation data is obtained. The multi-modal micro-needle eyebrow implantation data is respectively used to quantitatively evaluate the recognition timeliness of the multi-modal micro-needle eyebrow implantation data and the data fusion performance of the multi-modal micro-needle eyebrow implantation data, and the quantitative values of data recognition timeliness and data fusion performance are respectively compared and analyzed with the determination values in the database. If there is a deviation between the quantitative value and the determination value, corresponding optimization and adjustment measures are taken to improve the accuracy of the intelligent control method for micro-needle eyebrow implantation.

[0020] The technical solution in the embodiments of the present application is to solve the above problem that the accuracy of the intelligent control method for micro-needle eyebrow implantation is insufficient due to differences in data timeliness and data fusion. The general idea is as follows:

[0021] By respectively quantitatively evaluating the recognition timeliness of the multi-modal micro-needle eyebrow implantation data and the data fusion performance of the multi-modal micro-needle eyebrow implantation data with the multi-modal micro-needle eyebrow implantation data, and comparing and analyzing the quantitative values of data recognition timeliness and data fusion performance with the determination values in the database respectively, the intelligent control method for micro-needle eyebrow implantation is optimized and adjusted, and the accuracy of the intelligent control method for micro-needle eyebrow implantation is improved.

[0022] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings of the specification and specific implementation manners.

[0023] As Figure 1 shown, it is a flowchart of the intelligent control method for micro-needle eyebrow implantation provided by the embodiments of the present application. This method is applied to the intelligent control system for micro-needle eyebrow implantation. The method includes the following steps: collecting multi-modal micro-needle eyebrow implantation original data, preprocessing the multi-modal micro-needle eyebrow implantation original data to obtain multi-modal micro-needle eyebrow implantation data; respectively quantitatively evaluating the recognition timeliness of the multi-modal micro-needle eyebrow implantation data and the data fusion performance of the multi-modal micro-needle eyebrow implantation data with the multi-modal micro-needle eyebrow implantation data to obtain the quantitative value of data recognition timeliness and the quantitative value of data fusion performance; respectively comparing and analyzing the quantitative value of data recognition timeliness and the quantitative value of data fusion performance with the data recognition timeliness determination value and the data fusion performance determination value in the database to obtain the intelligent control optimization method for micro-needle eyebrow implantation.

[0024] Furthermore, the multi-modal micro-needle eyebrow implantation data refers to the data obtained by cleaning and denoising the original data of micro-needle eyebrow implantation collected by an optical coherence tomography scanner; the multi-modal micro-needle eyebrow implantation data includes data recognition timeliness data and data fusion performance data; the data recognition timeliness data includes OCT imaging processing time, robotic arm response delay, multi-modal micro-needle eyebrow implantation data synchronization error coefficient, and collagen fiber arrangement density analysis time; the data fusion performance data includes multi-modal micro-needle eyebrow implantation data synchronization error coefficient, robotic arm response delay, control instruction generation delay, and multi-modal micro-needle eyebrow implantation data processing delay.

[0025] It should be understood that before quantitatively evaluating the data recognition timeliness and data fusion performance of multi-modal micro-needle eyebrow implantation data respectively through the multi-modal micro-needle eyebrow implantation data, it also includes: obtaining the data recognition timeliness quantization value threshold, data fusion performance quantization value threshold, importance score of the data recognition timeliness quantization value, and importance score of the data fusion performance quantization value from the constructed intelligent control database for eyebrow implantation; the data recognition timeliness quantization value threshold includes the optical coherence tomography imaging processing time threshold, robotic arm response delay threshold, collagen fiber arrangement density analysis time threshold, and multi-modal micro-needle eyebrow implantation data timestamp standard value; the importance score of the data recognition timeliness quantization value includes the importance score of OCT imaging processing time, importance score of robotic arm response delay, importance score of multi-modal micro-needle eyebrow implantation data synchronization error coefficient, and importance score of collagen fiber arrangement density analysis time; the data fusion performance quantization value threshold includes the second robotic arm response delay threshold, control instruction generation delay threshold, multi-modal micro-needle eyebrow implantation data processing delay threshold, and multi-modal micro-needle eyebrow implantation data timestamp second standard value; the importance score of the data fusion performance quantization value includes the second importance score of multi-modal micro-needle eyebrow implantation data synchronization error coefficient, second importance score of robotic arm response delay, importance score of control instruction generation delay, importance score of multi-modal micro-needle eyebrow implantation data processing delay, and importance score of the data fusion performance quantization value.

[0026] In this embodiment, the OCT imaging processing time represents the time required from the start of image acquisition to the completion of image processing by integrating a timing function in the OCT imaging system; the robotic arm response delay represents the time delay between the issuance of an instruction by the control system and the start of the corresponding action of the robotic arm by integrating a timestamp recording function in the robotic arm control system; the multimodal microneedle eyebrow implantation data synchronization error represents the time difference between different modal data by comparing the timestamps of different modal data such as OCT images, force feedback data, position data, etc.; the collagen fiber arrangement density analysis time represents the time required from receiving the OCT image to the completion of the collagen fiber arrangement density analysis by integrating a timing function in the image analysis software; the control instruction generation delay represents the time difference between the completion time of data analysis and processing and the control instruction generation time by integrating a timestamp recording function in the microneedle eyebrow implantation control system; the multimodal microneedle eyebrow implantation data processing delay represents the time required from receiving the multimodal microneedle eyebrow implantation data to the completion of data processing by integrating a timing function in the data processing system.

[0027] Furthermore, the specific steps to obtain the data recognition timeliness quantification value are as follows: The result of the ratio analysis of the OCT imaging processing time threshold and the OCT imaging processing time is corrected by the importance score of the OCT imaging processing time, denoted as the first component of the data recognition timeliness quantification value; the result of the ratio analysis of the robotic arm response delay threshold and the robotic arm response delay is corrected by the importance score of the robotic arm response delay, denoted as the second component of the data recognition timeliness quantification value; the multimodal microneedle eyebrow implantation data synchronization error coefficient is averaged, and the result of the averaging process is corrected by the importance score of the multimodal microneedle eyebrow implantation data synchronization error coefficient, denoted as the third component of the data recognition timeliness quantification value. The multimodal microneedle eyebrow implantation data synchronization error coefficient represents the result of the ratio analysis of the multimodal microneedle eyebrow implantation data timestamp standard value and the deviation between the multimodal microneedle eyebrow implantation data timestamp and the multimodal microneedle eyebrow implantation data timestamp standard value; the result of the ratio analysis of the collagen fiber arrangement density analysis time threshold and the collagen fiber arrangement density analysis time is corrected by the importance score of the collagen fiber arrangement density analysis time, denoted as the fourth component of the data recognition timeliness quantification value; the first component, the second component, the third component, and the fourth component of the data recognition timeliness quantification value are coupled to obtain the data recognition timeliness quantification value; the data recognition timeliness quantification value is a quantitative data for measuring the timeliness of the system from data acquisition to the generation of control instructions.

[0028] The specific method for analyzing and obtaining the data recognition timeliness quantification value is as follows:

[0029] ; ;

[0030] Number the preset data recognition timeliness detection points in sequence. Represents the number of the data recognition timeliness detection point. , Represents the total number of the numbers of the data recognition timeliness detection points.

[0031] Divide the preset data recognition timeliness time into data recognition timeliness detection time periods of the same time length. Represents the number of the data recognition timeliness detection time period. , Represents the total number of the numbers of the data recognition timeliness detection time periods.

[0032] Represents the data recognition timeliness quantization value of the

[0033] Represents the th data recognition timeliness detection point under the

[0034] Represents the th

[0035] OCT imaging processing time under the

[0036] th data recognition timeliness detection time period, which refers to the time required from the acquisition of the original data by the Optical Coherence Tomography (OCT) device to the generation of the image available for analysis.

[0037] Represents the mechanical arm response delay threshold, which is the preset mechanical arm response delay threshold obtained from the intelligent control database for eyebrow implantation and can be the average value of the mechanical arm response delays under the historical data recognition timeliness detection time periods in the intelligent control database for eyebrow implantation.

[0038] Represents the The multi-modal microneedle eyebrow implantation data synchronization error under a data recognition timeliness detection point.

[0039] represents the time threshold for collagen fiber arrangement density analysis, which is the preset time threshold for collagen fiber arrangement density analysis obtained from the intelligent control database for eyebrow implantation. It can be the average value of the time for collagen fiber arrangement density analysis during the historical data recognition timeliness detection period in the intelligent control database for eyebrow implantation.

[0040] represents the time for collagen fiber arrangement density analysis under the

[0041] th data recognition timeliness detection period, referring to the time required to analyze the arrangement density of collagen fibers in OCT images or other relevant images. represents the standard value of the multi-modal microneedle eyebrow implantation data timestamp under the

[0042] th data recognition timeliness detection point, which is the preset standard value of the multi-modal microneedle eyebrow implantation data timestamp obtained from the intelligent control database for eyebrow implantation. It can be the average value of the multi-modal microneedle eyebrow implantation data timestamps under the historical data recognition timeliness detection points in the intelligent control database for eyebrow implantation. represents the multi-modal microneedle eyebrow implantation data timestamp under the

[0043] th data recognition timeliness detection point.

[0044] is the preset importance score of the OCT imaging processing time obtained from the intelligent control database for eyebrow implantation.

[0045] is the preset importance score of the robotic arm response delay obtained from the intelligent control database for eyebrow implantation.

[0046] is the preset importance score of the multi-modal microneedle eyebrow implantation data synchronization error coefficient obtained from the intelligent control database for eyebrow implantation.

[0047] is the preset importance score of the time for collagen fiber arrangement density analysis obtained from the intelligent control database for eyebrow implantation.

[0048] In this embodiment, a mapping table of importance scores is obtained through a database. For example, corresponding importance scores are extracted based on the current OCT imaging processing time, robotic arm response delay, multi-modal micro-needle eyebrow implantation data synchronization error coefficient, and collagen fiber arrangement density analysis time, such as the importance score of the OCT imaging processing time, the importance score of the robotic arm response delay, the importance score of the multi-modal micro-needle eyebrow implantation data synchronization error coefficient, and the importance score of the collagen fiber arrangement density analysis time. This mapping table defines a clear set of association rules that convert the specific values of the OCT imaging processing time, robotic arm response delay, multi-modal micro-needle eyebrow implantation data synchronization error coefficient, and collagen fiber arrangement density analysis time into their corresponding importance scores. Under this mechanism, whether it is to achieve an exact one-to-one match or multiple parameters converge into a many-to-one relationship, the dynamic acquisition of importance scores can be effectively achieved.

[0049] The OCT imaging processing time and the collagen fiber arrangement density analysis time jointly affect the efficiency of the entire process from data acquisition to feature extraction. The longer the OCT imaging processing time and the collagen fiber arrangement density analysis time, the more it causes delays in multi-modal micro-needle eyebrow implantation data processing, thereby affecting the response speed and accuracy of the robotic arm. The robotic arm response delay refers to the time difference between the completion of data processing and the execution of an action by the robotic arm. The greater the robotic arm response delay, the more it will cause the robotic arm to be unable to make precise adjustments based on real-time data, and the greater the multi-modal micro-needle eyebrow implantation data synchronization error coefficient.

[0050] There is a negative correlation between the OCT imaging processing time and the quantification value of data recognition timeliness. The longer the OCT imaging processing time, it means the larger the time interval from data acquisition to image generation, and the lower the quantification value of data recognition timeliness. There is a low correlation between the multi-modal micro-needle eyebrow implantation data synchronization error coefficient and the quantification value of data recognition timeliness. The greater the multi-modal micro-needle eyebrow implantation data synchronization error coefficient, the worse the time alignment between different modal data, and the lower the quantification value of data recognition timeliness. There is a negative correlation between the robotic arm response delay and the quantification value of data recognition timeliness. The higher the robotic arm response delay, it indicates the larger the time difference from receiving an instruction to actually executing an action, and the lower the quantification value of data recognition timeliness. There is a negative correlation between the collagen fiber arrangement density analysis time and the quantification value of data recognition timeliness. The longer the collagen fiber arrangement density analysis time, the longer it takes to extract useful features from the image data, and the lower the quantification value of data recognition timeliness.

[0051] In summary, in this embodiment, by quantitatively evaluating the parameters of the OCT imaging processing time, multi-modal micro-needle eyebrow implantation data synchronization error coefficient, robotic arm response delay, and collagen fiber arrangement density analysis time, the timeliness of data recognition can be accurately evaluated, thereby providing an important quantitative basis for optimizing the intelligent control system of micro-needle eyebrow implantation.

[0052] Further, the specific steps for obtaining the data fusion performance quantification value are as follows: perform a proportion analysis on the deviation between the multimodal microneedle eyebrow implantation data timestamp and the second standard value of the multimodal microneedle eyebrow implantation data timestamp and the multimodal microneedle eyebrow implantation data timestamp to obtain the multimodal microneedle eyebrow implantation data synchronization error coefficient, perform an averaging process on the multimodal microneedle eyebrow implantation data synchronization error coefficient, and use the second importance score of the multimodal microneedle eyebrow implantation data synchronization error coefficient to compensate the result of the averaging process to obtain the first component of the data fusion performance quantification value; compensate the result of the proportion analysis of the robotic arm response delay and the second threshold of the robotic arm response delay through the second importance score of the robotic arm response delay to obtain the second component of the data fusion performance quantification value; compensate the result of the proportion analysis of the control instruction generation delay and the control instruction generation delay threshold through the importance score of the control instruction generation delay to obtain the third component of the data fusion performance quantification value; compensate the result of the proportion analysis of the multimodal microneedle eyebrow implantation data processing delay and the multimodal microneedle eyebrow implantation data processing delay threshold through the importance score of the multimodal microneedle eyebrow implantation data processing delay to obtain the fourth component of the data fusion performance quantification value; compensate the result of the averaging process of the data recognition timeliness quantification value through the importance score of the data recognition timeliness quantification value to obtain the fifth component of the data fusion performance quantification value; perform a proportion analysis on the coupling result of the first component, the second component, the third component, and the fourth component of the data fusion performance quantification value and the fifth component of the data fusion performance quantification value to obtain the data fusion performance quantification value; the data fusion performance quantification value is used to reflect the accuracy of the system in integrating multi-source sensor data.

[0053] The specific method for obtaining the data fusion performance quantification value through the above analysis is as follows:

[0054] ;

[0055] ;

[0056] Number the preset data fusion performance detection points in sequence, denotes the number of the data fusion performance detection point, , denotes the total number of the data fusion performance detection point numbers.

[0057] Divide the preset data fusion performance time into data fusion performance detection periods of the same time length, denotes the number of the data fusion performance detection period, , denotes the total number of the data fusion performance detection period numbers.

[0058] denotes the The data fusion performance quantization value for a data fusion performance detection period.

[0059] Indicates the Multimodal micro-needle eyebrow implant data synchronization error coefficient at the

[0060] Indicates the Manipulator response delay during the

[0061] Indicates the second threshold of the manipulator response delay, which is the preset second threshold of the manipulator response delay obtained from the intelligent control database for eyebrow implant, and can be the average value of the manipulator response delay during the historical data fusion performance detection period in the intelligent control database for eyebrow implant.

[0062] Indicates the Control command generation delay during the

[0063] Indicates the control command generation delay threshold, which is the preset control command generation delay threshold obtained from the intelligent control database for eyebrow implant, and can be the average value of the control command generation delay during the historical data fusion performance detection period in the intelligent control database for eyebrow implant.

[0064] Indicates the Multimodal micro-needle eyebrow implant data processing delay during the

[0065] Indicates the multimodal micro-needle eyebrow implant data processing delay threshold, which is the preset multimodal micro-needle eyebrow implant data processing delay threshold obtained from the intelligent control database for eyebrow implant, and can be the average value of the multimodal micro-needle eyebrow implant data processing delay during the historical data fusion performance detection period in the intelligent control database for eyebrow implant.

[0066] Indicates the Multimodal micro-needle eyebrow implant data timestamp at the

[0067] Indicates the The second standard value of the multimodal micro - needle eyebrow - implanting data timestamp under a data fusion performance detection point is the preset second standard value of the multimodal micro - needle eyebrow - implanting data timestamp obtained from the intelligent control database for eyebrow - implanting, which can be the average value of the multimodal micro - needle eyebrow - implanting data timestamp under the historical data fusion performance detection point in the intelligent control database for eyebrow - implanting.

[0068] Indicates the quantification value of data recognition timeliness for the data recognition timeliness detection period.

[0069] Indicates a constant term, which is to avoid meaningless points in the data.

[0070] Is the second importance score of the preset multimodal micro - needle eyebrow - implanting data synchronization error coefficient obtained from the intelligent control database for eyebrow - implanting.

[0071] Is the second importance score of the preset robotic arm response delay obtained from the intelligent control database for eyebrow - implanting.

[0072] Is the importance score of the preset control instruction generation delay obtained from the intelligent control database for eyebrow - implanting.

[0073] Is the importance score of the preset multimodal micro - needle eyebrow - implanting data processing delay obtained from the intelligent control database for eyebrow - implanting.

[0074] Is the importance score of the preset quantification value of data recognition timeliness obtained from the intelligent control database for eyebrow - implanting.

[0075] In this embodiment, a mapping table of importance scores is obtained through the database. For example, according to the current multimodal micro - needle eyebrow - implanting data synchronization error coefficient, robotic arm response delay, control instruction generation delay, multimodal micro - needle eyebrow - implanting data processing delay, and quantification value of data recognition timeliness, the corresponding importance scores are extracted, such as the importance score of the multimodal micro - needle eyebrow - implanting data synchronization error coefficient, the importance score of the robotic arm response delay, the importance score of the control instruction generation delay, the importance score of the multimodal micro - needle eyebrow - implanting data processing delay, and the importance score of the quantification value of data recognition timeliness. This mapping table defines a clear set of association rules, which converts the specific values of the multimodal micro - needle eyebrow - implanting data synchronization error coefficient, robotic arm response delay, control instruction generation delay, multimodal micro - needle eyebrow - implanting data processing delay, and quantification value of data recognition timeliness into their corresponding importance scores. Under this mechanism, whether it is to achieve an exact one - to - one match or a many - to - one relationship formed by multiple parameters, the dynamic acquisition of importance scores can be effectively realized.

[0076] The multi-modal micro-needle eyebrow implantation data synchronization error coefficient reflects the synchronization degree of different modal data such as OCT, near-infrared spectroscopy (NIR), mechanical sensors, etc. in time. The larger the multi-modal micro-needle eyebrow implantation data synchronization error coefficient, the more it affects the fusion quality of multi-modal data, and the longer the control instruction generation delay is, because accurate instructions need to be based on synchronized data; the larger the multi-modal micro-needle eyebrow implantation data synchronization error coefficient, the more likely data is out of sync, and the robotic arm responds based on outdated or incorrect information, resulting in response delays or inaccurate actions; the increase in the multi-modal micro-needle eyebrow implantation data processing delay will lead to a delay in control instruction generation, and further increase the robotic arm response delay, and the robotic arm response delay increases; the larger the multi-modal micro-needle eyebrow implantation data processing delay, the more untimely the multi-modal micro-needle eyebrow implantation data is processed, and the smaller the data recognition timeliness quantization value is.

[0077] There is a positive correlation between the multi-modal micro-needle eyebrow implantation data synchronization error coefficient and the data fusion performance quantization value. The larger the multi-modal micro-needle eyebrow implantation data synchronization error coefficient, the worse the time synchronization between different modal data, and the larger the data fusion performance quantization value; there is a positive correlation between the robotic arm response delay and the data fusion performance quantization value. The longer the robotic arm response delay, the greater the time difference between receiving the control instruction and executing the action, and the larger the data fusion performance quantization value; there is a positive correlation between the control instruction generation delay and the data fusion performance quantization value. The longer the control instruction generation delay, the greater the time difference between the completion of data processing and the generation of control instructions, and the larger the data fusion performance quantization value; there is a positive correlation between the multi-modal micro-needle eyebrow implantation data processing delay and the data fusion performance quantization value. The longer the multi-modal micro-needle eyebrow implantation data processing delay, the greater the time difference between data acquisition and data processing completion, and the larger the data fusion performance quantization value; there is a negative correlation between the data recognition timeliness quantization value and the data fusion performance quantization value. The larger the data recognition timeliness quantization value, the higher the data recognition timeliness of multi-modal micro-needle eyebrow implantation data, and the lower the data fusion performance quantization value.

[0078] In summary, in this embodiment, by quantitatively evaluating the multi-modal micro-needle eyebrow implantation data synchronization error coefficient, robotic arm response delay, control instruction generation delay, and multi-modal micro-needle eyebrow implantation data processing delay parameters, the performance of data fusion can be accurately evaluated, thus providing an important quantitative basis for optimizing the intelligent control system of micro-needle eyebrow implantation.

[0079] Further, the specific steps to obtain the intelligent control optimization method for micro - needle eyebrow implantation are as follows: Obtain the data recognition timeliness determination value and the data fusion performance determination value from the eyebrow implantation intelligent control database; If the data recognition timeliness quantization value is greater than or equal to the data recognition timeliness determination value, do not trigger the data recognition timeliness feature extraction optimization method, otherwise trigger the data recognition timeliness feature extraction optimization method; If the data fusion performance quantization value is lower than or equal to the data fusion performance determination value, do not trigger the data fusion performance optimization method, otherwise trigger the data fusion performance optimization method; The intelligent control optimization method for micro - needle eyebrow implantation includes the data recognition timeliness feature extraction optimization method and the data fusion performance optimization method.

[0080] Among them, the data recognition timeliness feature extraction optimization method is as follows: Compare the obtained data recognition timeliness quantization value with the data recognition timeliness determination value to obtain the data recognition timeliness quantization value deviation, which is used to reflect the timeliness difference degree of the system from data acquisition to generating control instructions; If the obtained data recognition timeliness quantization value deviation is less than or equal to the safety deviation threshold in the eyebrow implantation intelligent control database, no additional processing is performed on the initial configuration; If the obtained data recognition timeliness quantization value deviation is greater than the safety deviation threshold and less than the preset deviation threshold, after local resource adjustment, monitor whether the accuracy loss of optical coherence tomography image processing is greater than the preset accuracy loss threshold. If it is greater, restore the GPU resource allocation, the number of NIR blood flow analysis cores, and the FPGA resources to the state before local resource adjustment, otherwise retain the resource configuration after local resource adjustment. Local resource adjustment means promoting the priority level of a preset key task in the GPU resource allocation by one level, simultaneously restricting the number of processor cores occupied by the NIR blood flow analysis task to the preset core number threshold, and after adjusting the FPGA hardware configuration, transferring the hardware resources for processing NIR data to the GPU; If the obtained data recognition timeliness quantization value deviation is greater than or equal to the preset deviation threshold, perform feature extraction adjustment and task scheduling adjustment. Feature extraction adjustment means pausing the high - frequency detail extraction based on the high - pass filter in optical coherence tomography image processing while retaining the low - frequency contour features. Task scheduling adjustment means integrating the feature data into feature vector data and mapping the feature vector data to the shared memory of the robotic arm controller.

[0081] In this embodiment, as Figure 2As shown, it is a mind map of the optimization method for extracting the timeliness feature of data recognition in the intelligent control method of micro - needle eyebrow implantation provided by the embodiment of the present application. During the multi - modal micro - needle eyebrow implantation process, if the deviation of the data recognition timeliness quantization value is greater than the safety deviation threshold and less than the preset deviation threshold, it indicates a slight performance decline, and local resource adjustment is required to optimize the performance. The steps for local resource adjustment are as follows: By adjusting the allocation of graphics processing unit (GPU) resources, ensure that key tasks such as OCT image processing obtain more computing resources, thereby reducing its processing latency. Increase the priority of the OCT processing task, and the occupancy rate of the OCT processing task will also increase. At the same time, limit the core number of NIR blood flow analysis to be reduced to the preset core number threshold of NIR blood flow analysis. At the same time, transfer some field - programmable gate array (FPGA) resources such as NIR data to data compression, which can reduce the data transmission volume, thereby reducing the latency caused by data transmission. If the deviation is less than the safety deviation threshold and the accuracy is less than the preset threshold, retain the new configuration. If the accuracy loss is greater than the preset threshold, it means that the new configuration causes an unacceptable accuracy loss. At this time, it is necessary to roll back to the previous stable version to ensure the effectiveness of the optimization measures, roll back to the previous stable version, ensure the effectiveness of the optimization measures, and avoid excessive accuracy loss. If the obtained deviation of the data recognition timeliness quantization value is greater than or equal to the preset deviation threshold, perform feature extraction adjustment and task scheduling adjustment. The steps for feature extraction adjustment are as follows: Turn off some high - frequency detail analyses, especially in OCT image processing. For the high - frequency detail extraction using a high - pass filter, these computationally intensive operations are not performed during the execution of real - time tasks, so as to reduce the computational amount while maintaining sufficient image information. At the same time, focus on retaining the low - frequency contour features of the image because these features are usually more critical for real - time tasks and have a small computational amount, thus reducing the processing time. The steps for task scheduling adjustment are as follows: To ensure that real - time tasks such as OCT processing and robotic arm control can obtain sufficient computing resources, suspend some non - real - time tasks. Non - real - time tasks include data recording, postoperative report generation, system log recording, etc., which are temporarily suspended during the execution of real - time tasks. At the same time, preferentially allocate unified computing device architecture (CUDA) cores to real - time tasks, such as OCT image processing and robotic arm control algorithms, which can be achieved through the explicit device management of CUDA, ensuring that these tasks can use most of the computing resources of the GPU, and are achieved by sending signals or using a task scheduler. Use CUDA - related tools and libraries, such as nvidia - smi, cuda - memcheck, to monitor and control the use of GPU resources, ensure that real - time tasks can exclusively occupy GPU resources, and improve the timeliness of data recognition.After the above optimization and adjustment, output the feature data, and splice the extracted OCT features and NIR features into the final feature vector. To reduce the latency during data transmission, directly map the multi-modal micro-needle eyebrow implantation feature data to the shared memory of the robotic arm controller to reduce the latency during data copying.

[0082] It should be further supplemented that the data fusion performance optimization method is as follows: Compare the obtained data fusion performance quantization value with the data fusion performance determination value to obtain the data fusion performance quantization value deviation, which is used to measure the gap between the actual data fusion performance and the expected data fusion performance; if the data fusion performance quantization value deviation is equal to zero, it means that the data fusion performance of the current system exactly reaches the preset benchmark level, indicating that the efficiency and accuracy of the system in integrating multi-modal data meet the expectations; if the data fusion performance quantization value deviation is less than zero, it means that the data fusion performance of the current system is better than the preset benchmark level, and the data fusion efficiency is greater than the expected data fusion efficiency; if the data fusion performance quantization value deviation is greater than zero, perform data fusion adjustment and resource allocation adjustment. Data fusion adjustment means triggering a window reduction instruction when the adjustment effect after window adjustment based on the obtained data fusion performance quantization value deviation meets the preset adjustment effect condition, otherwise triggering a window increase instruction. The preset adjustment effect condition means that the latency deviation is greater than the preset threshold, the accuracy deviation is less than the preset threshold, and the newly obtained data fusion performance quantization value deviation is less than zero, then trigger the window reduction, otherwise trigger the window increase instruction. Resource allocation adjustment includes explicitly managing through unified computing devices and architecture explicit devices, and preferentially allocating graphics processing unit resources to optical coherence tomography processing and robotic arm control tasks.

[0083] In this embodiment, as Figure 3As shown, it is a mind map of the data fusion performance optimization method for the intelligent control method of micro - needle eyebrow implantation provided by the embodiments of this application. In a multi - modal micro - needle eyebrow implantation system, in order to ensure that the quantization value of data fusion performance remains within the positive range, thereby improving system performance and eyebrow implantation effect, if the deviation of the data fusion performance quantization value is greater than zero, data fusion adjustment and resource allocation adjustment are carried out. The specific steps of data fusion adjustment are as follows: learn the spatial offset between OCT and NIR through a deformable convolutional network to align the feature maps, and sample the feature maps according to the feature map offset to better correspond the micro - circulation information with the structural boundary information. According to the alignment effect, if the deviation of the data fusion performance quantization value is equal to zero, it reaches the expectation and no adjustment is needed; if the deviation of the data fusion performance quantization value is less than zero, it may be over - fitting and the alignment constraint of the feature map needs to be appropriately relaxed; if the deviation of the data fusion performance quantization value is greater than zero, it does not reach the expectation, calculate the attention weight matrix to calculate the correlation between OCT and NIR features. First, represent the features of OCT and NIR images in vector form, which is achieved through the convolutional neural network (CNN) feature extraction method. Extract the tomographic features of OCT as the query matrix, and use the optical flow network to extract the blood flow velocity field information of NIR as the key - value matrix. For each OCT feature vector (as Query), calculate its dot product with the NIR feature vector (as Key). This dot product reflects the correlation between OCT features and NIR features, and through the attention mechanism, dynamic weighted fusion of structural flow and functional flow features is realized, effectively suppressing the influence of noise regions such as blood vessel rupture areas. First, measure the actual delay time when the system processes specific tasks such as OCT imaging processing and robotic arm response, which can be achieved through a timing function. Compare the actual delay time with the preset delay threshold, and calculate the difference between the two, which is the delay deviation. The delay deviation represents the gap between the actual delay time and the expected delay time; compare the actual accuracy with the preset accuracy threshold, and calculate the difference between the two, which is the accuracy deviation. The accuracy deviation represents the gap between the actual accuracy and the expected accuracy; and follow the adjustment principle that delay takes precedence over accuracy. If the delay is greater than the preset delay threshold, the accuracy is less than the preset accuracy threshold, and the deviation of the re - obtained data fusion performance quantization value is less than zero, the trigger window is reduced to the minimum window limit threshold; if the delay is less than the preset delay threshold, the accuracy is greater than the preset accuracy threshold, and the deviation of the re - obtained data fusion performance quantization value is greater than zero, the trigger window is increased to the maximum window limit threshold.The adjustment direction is based on the deviation value of the data fusion performance quantization value, ensuring that the adjusted window size is between the minimum window limit threshold and the maximum window limit threshold. The maximum window limit threshold is to prevent memory overflow, and the minimum window limit threshold is to ensure basic context information. If it exceeds, it is forced to be set to the boundary value. A matching adjustment value is determined based on the deviation value of the data fusion performance quantization value. The adjustment value is determined according to the amount of deviation of the data fusion performance quantization value. This adjustment value is added to the original window size to increase or decrease the window. The matching process is based on the mapping relationship established between the deviation value of the data fusion performance quantization value and the adjustment value. For example, the calculated deviation value of the data fusion performance quantization value is used as the input, and the corresponding adjustment value is found according to the mapping relationship. The resource allocation adjustment is as follows: The tasks in the system are divided into OCT processing, robotic arm control, and other task categories. High priorities are set for OCT processing and robotic arm control tasks to ensure that OCT processing and robotic arm control tasks have priority in resource allocation. A preset proportion of GPU resources is reserved for OCT processing and robotic arm control tasks to ensure that the tasks can obtain resources immediately when needed. According to the execution situation of system tasks, the resource allocation strategy is dynamically adjusted. If the GPU utilization rate is lower than the preset GPU utilization rate, the remaining GPU resources are calculated and released to other tasks for use to improve resource utilization. If the GPU utilization rate is greater than the preset GPU utilization rate, the resources used by other tasks in the GPU are released to OCT processing and robotic arm control tasks for use, realizing the effective management and optimal allocation of GPU resources, ensuring that OCT processing and robotic arm control tasks obtain sufficient resource support during execution, thereby improving the overall performance and real-time performance of the system, and ensuring that the quantization value of the data fusion performance always remains within the positive range, thereby enhancing the overall performance of the system and the effect of eyebrow implantation.

[0084] As Figure 4 shown, it is the structural diagram of the intelligent control system for micro-needle eyebrow implantation provided by the embodiment of the present application. The intelligent control system for micro-needle eyebrow implantation provided by the embodiment of the present application includes: a multi-modal micro-needle eyebrow implantation data acquisition and processing module, a multi-modal micro-needle eyebrow implantation data analysis module, and an intelligent control and optimization module for micro-needle eyebrow implantation: The multi-modal micro-needle eyebrow implantation data acquisition and processing module: used to acquire multi-modal micro-needle eyebrow implantation raw data, preprocess the multi-modal micro-needle eyebrow implantation raw data, and obtain multi-modal micro-needle eyebrow implantation data; The multi-modal micro-needle eyebrow implantation data analysis module: used to quantitatively evaluate the timeliness of multi-modal micro-needle eyebrow implantation data recognition and the data fusion performance of multi-modal micro-needle eyebrow implantation data through the multi-modal micro-needle eyebrow implantation data, and obtain the data recognition timeliness quantization value and the data fusion performance quantization value; The intelligent control and optimization module for micro-needle eyebrow implantation: used to compare and analyze the data recognition timeliness quantization value and the data fusion performance quantization value with the data recognition timeliness determination value and the data fusion performance determination value in the database respectively, and obtain the intelligent control and optimization method for micro-needle eyebrow implantation.

[0085] The intelligent control device for micro-needle eyebrow implantation provided by the embodiments of the present application includes: The device has one or more programs, and the one or more programs are executed by one or more processors to implement the above-mentioned method.

[0086] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0087] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0088] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0089] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0090] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made to these embodiments by those skilled in the art once they learn of the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0091] Obviously, those skilled in the art can make various changes and deformations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and deformations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and deformations.

Claims

1. The intelligent control method for micro-needle eyebrow implantation, characterized in that, It includes the following steps: Collect the original data of multi-modal micro-needle eyebrow implantation, preprocess the original data of multi-modal micro-needle eyebrow implantation to obtain the data of multi-modal micro-needle eyebrow implantation; Quantitatively evaluate the recognition timeliness of the multi-modal micro-needle eyebrow implantation data and the data fusion performance of the multi-modal micro-needle eyebrow implantation data respectively through the multi-modal micro-needle eyebrow implantation data, and obtain the quantitative value of data recognition timeliness and the quantitative value of data fusion performance; Compare and analyze the quantitative value of data recognition timeliness and the quantitative value of data fusion performance with the data recognition timeliness determination value and the data fusion performance determination value in the database respectively to obtain the intelligent control optimization method for micro-needle eyebrow implantation; The multi-modal micro-needle eyebrow implantation data represents the data obtained by cleaning and denoising the original data of micro-needle eyebrow implantation collected by an optical coherence tomography scanner; The multi-modal micro-needle eyebrow implantation data includes data recognition timeliness data and data fusion performance data; The data recognition timeliness data includes OCT imaging processing time, robotic arm response delay, multi-modal micro-needle eyebrow implantation data synchronization error coefficient, and collagen fiber arrangement density analysis time; The data fusion performance data includes multi-modal micro-needle eyebrow implantation data synchronization error coefficient, robotic arm response delay, control instruction generation delay, and multi-modal micro-needle eyebrow implantation data processing delay; Before quantitatively evaluating the recognition timeliness of the multi-modal micro-needle eyebrow implantation data and the data fusion performance of the multi-modal micro-needle eyebrow implantation data respectively through the multi-modal micro-needle eyebrow implantation data, it also includes: Obtain the threshold of the quantitative value of data recognition timeliness, the threshold of the quantitative value of data fusion performance, the importance score of the quantitative value of data recognition timeliness, and the importance score of the quantitative value of data fusion performance from the constructed intelligent control database for eyebrow implantation; The threshold of the quantitative value of data recognition timeliness includes the threshold of OCT imaging processing time, the threshold of robotic arm response delay, the threshold of collagen fiber arrangement density analysis time, and the standard value of the multi-modal micro-needle eyebrow implantation data timestamp; The importance score of the quantitative value of data recognition timeliness includes the importance score of OCT imaging processing time, the importance score of robotic arm response delay, the importance score of the multi-modal micro-needle eyebrow implantation data synchronization error coefficient, and the importance score of collagen fiber arrangement density analysis time; The threshold of the quantitative value of data fusion performance includes the second threshold of robotic arm response delay, the threshold of control instruction generation delay, the threshold of multi-modal micro-needle eyebrow implantation data processing delay, and the second standard value of the multi-modal micro-needle eyebrow implantation data timestamp; The importance score of the quantitative value of data fusion performance includes the second importance score of the multi-modal micro-needle eyebrow implantation data synchronization error coefficient, the second importance score of robotic arm response delay, the importance score of control instruction generation delay, the importance score of multi-modal micro-needle eyebrow implantation data processing delay, and the importance score of the quantitative value of data fusion performance; The specific steps to obtain the quantitative value of data recognition timeliness are: Correct the result of the ratio analysis of the OCT imaging processing time threshold and the OCT imaging processing time through the importance score of the OCT imaging processing time, and record it as the first component of the quantitative value of data recognition timeliness; The result of the ratio analysis of the robotic arm response delay threshold and the robotic arm response delay by the importance score of the robotic arm response delay is corrected, denoted as the second component of the data recognition timeliness quantification value; The multi-modal microneedle eyebrow implantation data synchronization error coefficient is averaged, and the result of the averaging process is corrected using the importance score of the multi-modal microneedle eyebrow implantation data synchronization error coefficient, denoted as the third component of the data recognition timeliness quantification value. The multi-modal microneedle eyebrow implantation data synchronization error coefficient represents the result of the ratio analysis of the deviation between the multi-modal microneedle eyebrow implantation data timestamp standard value, the multi-modal microneedle eyebrow implantation data timestamp, and the multi-modal microneedle eyebrow implantation data timestamp standard value; The result of the ratio analysis of the collagen fiber arrangement density analysis time threshold and the collagen fiber arrangement density analysis time is corrected by the importance score of the collagen fiber arrangement density analysis time, denoted as the fourth component of the data recognition timeliness quantification value; The first component, the second component, the third component, and the fourth component of the data recognition timeliness quantification value are coupled to obtain the data recognition timeliness quantification value; The data recognition timeliness quantification value is quantitative data for measuring the timeliness of the system from data acquisition to generating control instructions; The specific steps for obtaining the data fusion performance quantification value are as follows: The ratio analysis of the deviation between the multi-modal microneedle eyebrow implantation data timestamp, the multi-modal microneedle eyebrow implantation data timestamp, and the second standard value of the multi-modal microneedle eyebrow implantation data timestamp is performed to obtain the multi-modal microneedle eyebrow implantation data synchronization error coefficient. The multi-modal microneedle eyebrow implantation data synchronization error coefficient is averaged, and the result of the averaging process is compensated using the second importance score of the multi-modal microneedle eyebrow implantation data synchronization error coefficient to obtain the first component of the data fusion performance quantification value; The result of the ratio analysis of the robotic arm response delay and the second threshold of the robotic arm response delay is compensated by the second importance score of the robotic arm response delay to obtain the second component of the data fusion performance quantification value; The result of the ratio analysis of the control instruction generation delay and the control instruction generation delay threshold is compensated by the importance score of the control instruction generation delay to obtain the third component of the data fusion performance quantification value; The result of the ratio analysis of the multi-modal microneedle eyebrow implantation data processing delay and the multi-modal microneedle eyebrow implantation data processing delay threshold is compensated by the importance score of the multi-modal microneedle eyebrow implantation data processing delay to obtain the fourth component of the data fusion performance quantification value; The result of the averaging process of the data recognition timeliness quantification value is compensated by the importance score of the data recognition timeliness quantification value to obtain the fifth component of the data fusion performance quantification value; The ratio analysis is performed on the coupling result of the first component, the second component, the third component, and the fourth component of the data fusion performance quantification value and the fifth component of the data fusion performance quantification value to obtain the data fusion performance quantification value; The data fusion performance quantification value is used to reflect the accuracy of the system when integrating multi-source sensor data.

2. The intelligent control method for micro-needle eyebrow implantation according to claim 1, characterized in that, The specific steps of the intelligent control optimization method for obtaining micro-needle eyebrow implantation are as follows: Obtain the data recognition timeliness determination value and the data fusion performance determination value from the eyebrow implantation intelligent control database; If the data recognition timeliness quantization value is greater than or equal to the data recognition timeliness determination value, do not trigger the data recognition timeliness feature extraction optimization method, otherwise trigger the data recognition timeliness feature extraction optimization method; If the data fusion performance quantization value is lower than or equal to the data fusion performance determination value, do not trigger the data fusion performance optimization method, otherwise trigger the data fusion performance optimization method; The intelligent control optimization method for micro-needle eyebrow implantation includes the data recognition timeliness feature extraction optimization method and the data fusion performance optimization method.

3. The intelligent control method of micro-needle eyebrow implantation according to claim 2, characterized in that, The specific content of the data recognition timeliness feature extraction optimization method is as follows: Compare the obtained data recognition timeliness quantization value with the data recognition timeliness determination value to obtain the data recognition timeliness quantization value deviation, which is used to reflect the timeliness difference degree of the system from data acquisition to generating control instructions; If the obtained data recognition timeliness quantization value deviation is less than or equal to the safety deviation threshold in the eyebrow implantation intelligent control database, no additional processing is performed on the initial configuration; If the obtained data recognition timeliness quantization value deviation is greater than the safety deviation threshold and less than the preset deviation threshold, after performing local resource adjustment, monitor whether the accuracy loss of optical coherence tomography image processing is greater than the preset accuracy loss threshold. If it is greater, restore the GPU resource allocation, the number of cores for NIR blood flow analysis, and the FPGA resources to the state before local resource adjustment. Otherwise, retain the resource configuration after local resource adjustment. The local resource adjustment means promoting the priority of a preset key task in the GPU resource allocation by one level, restricting the number of processor cores occupied by the NIR blood flow analysis task to the preset core number threshold, and transferring the hardware resources for processing NIR data to the GPU after adjusting the FPGA hardware configuration; If the obtained data recognition timeliness quantization value deviation is greater than or equal to the preset deviation threshold, perform feature extraction adjustment and task scheduling adjustment. The feature extraction adjustment means pausing the high-frequency detail extraction based on the high-pass filter in the optical coherence tomography image processing while retaining the low-frequency contour features. The task scheduling adjustment means integrating the feature data into feature vector data and mapping the feature vector data to the shared memory of the robotic arm controller.

4. The intelligent control method of the micro-needle eyebrow implantation according to claim 2, characterized in that The specific content of the data fusion performance optimization method is as follows: Compare the obtained data fusion performance quantization value with the data fusion performance determination value to obtain the data fusion performance quantization value deviation, which is used to measure the gap degree between the actual data fusion performance and the expected data fusion performance; If the data fusion performance quantization value deviation is equal to zero, it indicates that the efficiency and accuracy of the system in integrating multi-modal data meet the expectations; If the data fusion performance quantization value deviation is less than zero, it indicates that the efficiency of data fusion is greater than the expected data fusion efficiency; If the deviation of the data fusion performance quantization value is greater than zero, data fusion adjustment and resource allocation adjustment are performed. The data fusion adjustment means that when the adjustment effect after window adjustment based on the obtained deviation of the data fusion performance quantization value meets the preset adjustment effect condition, a window reduction instruction is triggered; otherwise, a window increase instruction is triggered. The preset adjustment effect condition means that the delay deviation is greater than a preset threshold, the precision deviation is less than the preset threshold, and the deviation of the newly obtained data fusion performance quantization value is less than zero. The resource allocation adjustment includes explicitly managing through a unified computing device and architecture, and preferentially allocating graphics processing unit resources to optical coherence tomography processing and robotic arm control tasks.

5. The intelligent control system for micro-needle eyebrow implantation is used to implement the intelligent control method described in any one of claims 1-4, and is characterized in that, It includes a multi-modal micro-needle eyebrow implantation data acquisition and processing module, a multi-modal micro-needle eyebrow implantation data analysis module, and an intelligent control optimization module for micro-needle eyebrow implantation: The multi-modal micro-needle eyebrow implantation data acquisition and processing module: is used to acquire multi-modal micro-needle eyebrow implantation raw data, and preprocess the multi-modal micro-needle eyebrow implantation raw data to obtain multi-modal micro-needle eyebrow implantation data; The multi-modal micro-needle eyebrow implantation data analysis module: is used to quantitatively evaluate the timeliness of multi-modal micro-needle eyebrow implantation data recognition and the data fusion performance of multi-modal micro-needle eyebrow implantation data respectively through the multi-modal micro-needle eyebrow implantation data, and obtain a data recognition timeliness quantization value and a data fusion performance quantization value; The intelligent control optimization module for micro-needle eyebrow implantation: is used to compare and analyze the data recognition timeliness quantization value and the data fusion performance quantization value with the data recognition timeliness determination value and the data fusion performance determination value in the database respectively, and obtain an intelligent control optimization method for micro-needle eyebrow implantation.

6. An apparatus for applying the intelligent control method of micro-needle eyebrow implantation according to any one of claims 1-4, characterized in that: The device has one or more programs, and the one or more programs are executed by one or more processors to implement the above method.

Images