Mirror assembly effect evaluation system and method based on Internet of Things

Abstract

The invention discloses a mirror assembly effect evaluation system and method based on the Internet of Things, and particularly relates to the technical field of intelligent assembly evaluation.The mirror use information after assembly is collected through an Internet of Things sensor, a use information set is generated, mirror reflection is analyzed, and whether an early sign that the assembly effect is not matched with the environment exists or not is judged; when an unfit sign is detected, extracting a light adaptability feature and a visual aesthetic feature from the information set, and respectively generating a light adaptability index and a visual aesthetic index; then inputting the indexes into a pre-trained machine learning model, calculating an assembly adaptation degree, and dividing the lenses into a height adaptation type and a low adaptation type; for the low-degree adaptive lenses, the system automatically carries out reassembly and feature extraction until all the lenses are highly adaptive; according to the invention, the reflection effect and the visual performance of each lens in various illumination modes can be ensured to be in the optimal state, and the overall use experience of a user is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent assembly evaluation, and more specifically, to a mirror assembly effect evaluation system and method based on the Internet of Things. Background Art

[0002] In many application scenarios, mirrors are not only used as imaging tools, but also play a role in enhancing the beauty of space and functionality. However, in actual use, traditional mirror assembly often shows the problem of insufficient adaptability, especially under different lighting conditions. These problems are particularly prominent in mirrors with multiple lenses spliced ​​together. Uneven light, reduced clarity and brightness differences at the splicing point often affect the overall effect and bring poor visual experience to users. In order to achieve high-quality assembly effects, experienced technicians are usually required to debug and optimize, but this method is difficult to accurately quantify the assembly effect, resulting in uneven quality and difficulty in standardization.

[0003] With the rapid development of Internet of Things technology, real-time monitoring and data analysis through intelligent sensors have become possible. Internet of Things technology can remotely collect and analyze the usage data of mirrors after assembly, and realize intelligent detection and evaluation of assembly effects. For example, through Internet of Things sensors, the reflected light intensity, brightness uniformity, clarity and other parameters of the mirror can be collected under different ambient lighting conditions, and a data set can be generated for systematic analysis. Based on the data support of the Internet of Things, the adaptability effect of the mirror can be quantified from the assembly process, and a feedback mechanism can be formed to optimize the assembly process. This method not only improves the accuracy and consistency of the assembly effect, but also can discover potential adaptability risks at an early stage, so as to make timely adjustments and improve the overall assembly quality and user experience. Therefore, a mirror assembly effect evaluation system and method based on the Internet of Things is proposed. Summary of the invention

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] The mirror assembly effect evaluation method based on the Internet of Things includes the following steps:

[0006] Collect usage information of the assembled mirror through IoT sensors to obtain a usage information set;

[0007] Performing a specular reflection analysis based on the usage information set, and then judging whether there is an early sign that the mirror assembly effect is not compatible with the usage environment based on the specular reflection analysis results;

[0008] When there is an early sign that the mirror assembly effect is not compatible with the use environment, light adaptability features and visual aesthetic features are extracted from the use information set, and a light adaptability index and a visual aesthetic index are generated respectively;

[0009] The light adaptability index and the visual aesthetics index are input into the pre-trained machine learning model to output the assembly fitness. According to the assembly fitness, the lenses for assembly are divided into two types: high-fitting assembly and low-fitting assembly.

[0010] The lenses for low-fitting assembly are reassembled and feature extraction is performed again until all the lenses for assembly are classified as the assembly type of high-fitting assembly.

[0011] The mirror is assembled from a plurality of lenses, and the plurality of lenses are numbered respectively, and each lens corresponds to a unique number.

[0012] In a preferred embodiment, specular reflection analysis refers to:

[0013] When the assembled mirror is used in an application environment, the application environment is preset with multiple lighting modes. In each brightness mode, sampling data obtained by sampling the mirror reflected light intensity at the connection is found from the usage information set, so as to obtain comprehensive connection data in each brightness mode, and obtain multiple groups of sampling data. Each group of sampling data corresponds to a unique lighting mode, and then the average value and standard deviation of each group of sampling data are calculated respectively to obtain a stability combination [PJi, BZi] for reflecting the reflection intensity; i represents the type number of the lighting mode, PJi represents the average value of the sampling data of the i-th lighting mode, and BZi represents the standard deviation of the sampling data of the i-th lighting mode.

[0014] In a preferred embodiment, judging whether there is an early sign that the mirror assembly effect is not compatible with the use environment according to the mirror reflection analysis result refers to:

[0015] Obtain the mirror reflection analysis result, i.e., the stability combination [PJi, BZi]; the i-th lighting mode is preset with mirror reflection standard range 1 and mirror reflection standard range 2. If the stability combination PJi intersects with the mirror reflection standard range 1 and the stability combination BZi intersects with the mirror reflection standard range 2, a normal signal is generated. If the stability combination PJi intersects with the mirror reflection standard range 1 and the stability combination BZi intersects with the mirror reflection standard range 2, an abnormal signal is generated. If an abnormal signal is generated in any lighting mode, it indicates that there are early signs that the mirror assembly effect is not compatible with the use environment.

[0016] In a preferred embodiment, the logic for obtaining the light adaptability index is:

[0017] Extract the light reflection intensity data of each individual lens using the light sensor, and perform multiple samplings under different lighting modes. The extracted light adaptability characteristic data includes the average reflected light intensity and the standard deviation of the reflected light brightness fluctuation under each lighting mode;

[0018] For each lens j, a light adaptability coefficient LS I jm is generated according to the light adaptability characteristics. This coefficient represents the degree of light adaptation of the lens j under the current lighting conditions. The calculation formula is:

[0019] L represents the brightness of light under the current lighting conditions, i.e., the current lighting mode of type m; Ravg represents the average reflected light intensity under the current lighting conditions; σR represents the standard deviation of the reflected light brightness fluctuation under the current lighting conditions; k represents the preset fluctuation adjustment coefficient; LSIjm represents the degree of adaptation of lens j to light under the current lighting mode of type m;

[0020] The following processing is performed on all light modes to obtain the light adaptability index of each lens under the full brightness range: U represents the total number of light modes, Q is the preset sensitivity coefficient, Q is greater than 1, LSIj total Indicates the light adaptability index of lens j in the full brightness range.

[0021] In a preferred embodiment, the logic for obtaining the visual aesthetics index is:

[0022] The visual aesthetic characteristic indexes of each individual lens are extracted, including the image clarity index, the light distribution uniformity index, and the visual contrast index, and then the visual aesthetic coefficient of lens j under the current lighting conditions, that is, the current lighting mode of type m, is calculated. The calculation formula is:

[0023] QXm represents the corresponding value of the image clarity index of lens j in the current lighting mode of type m, GXm represents the corresponding value of the light distribution uniformity index of lens j in the current lighting mode of type m, SDm represents the corresponding value of the visual contrast index of lens j in the current lighting mode of type m, VSIjm represents the visual aesthetics coefficient of lens j in the current lighting mode of type m, w1, w2, w3 are all preset non-zero proportional coefficients;

[0024] The formula for calculating the visual aesthetics index is:

[0025] VSI total It represents the visual aesthetic index of lens j in the full brightness range.

[0026] In a preferred embodiment, the logic for obtaining the image clarity index is as follows: analyzing the specular reflection image by an image processing algorithm, using edge detection to quantify the clarity, and the corresponding value of the image clarity index is expressed as the average value of the edge intensity;

[0027] The logic for obtaining the light distribution uniformity index is as follows: the light sensor collects the light intensity at multiple positions on the mirror surface and calculates its mean square error. The result obtained is the corresponding value of the light distribution uniformity index.

[0028] The logic for obtaining the visual contrast index is: in different brightness areas, the contrast is calculated by the brightness difference, specifically the average brightness difference between bright and dark areas:

[0029] Lm max Indicates the average brightness of the bright area, Lm min Indicates the average brightness of dark areas.

[0030] In a preferred embodiment, the pre-trained machine learning model is a polynomial regression model.

[0031] In a preferred embodiment, the lenses for assembly are divided into two types of assembly: high-fitting assembly and low-fitting assembly according to the degree of assembly adaptation:

[0032] The assembly fitness of lens j is compared with the preset assembly fitness threshold. If the assembly fitness of lens j is greater than or equal to the preset assembly fitness threshold, lens j is classified as the assembly type of highly fit assembly. If the assembly fitness of lens j is less than the preset assembly fitness threshold, lens j is classified as the assembly type of low fit assembly.

[0033] In a preferred embodiment, the mirror assembly effect evaluation system based on the Internet of Things includes:

[0034] An information collection module collects usage information of the assembled mirror through an IoT sensor to obtain a usage information set;

[0035] A mirror reflection analysis module performs mirror reflection analysis based on the usage information set, and then determines whether there is an early sign that the mirror assembly effect is not compatible with the usage environment based on the mirror reflection analysis result;

[0036] A feature extraction module, when there is an early sign that the mirror assembly effect is not compatible with the use environment, extracts light adaptability features and visual aesthetic features from the use information set, and generates a light adaptability index and a visual aesthetic index respectively;

[0037] The single lens type classification module inputs the light adaptability index and visual aesthetics index into the pre-trained machine learning model, outputs the assembly fitness, and classifies the lenses for assembly into two types of assembly: high-fit assembly and low-fit assembly according to the assembly fitness;

[0038] The assembly optimization module reassembles the lenses with low fit and re-extracts features until all the lenses for assembly are classified as highly fit.

[0039] Technical effects and advantages of the present invention:

[0040] By collecting the light adaptability and visual aesthetics data of the mirror through the Internet of Things sensor, the present invention can accurately quantify the assembly effect of each lens. By using a pre-trained machine learning model to analyze the assembly adaptability, the assembly quality can be evaluated in a systematic and standardized manner, avoiding reliance on manual experience and improving the accuracy and consistency of assembly. The present invention ensures that the reflection effect and visual performance of each lens under a variety of lighting modes are in the best state through the calculation of the light adaptability index and the visual aesthetics index. For lenses with poor adaptability, the system can automatically prompt for readjustment to ensure that the assembled mirror can present a consistent visual effect under a variety of lighting conditions in the application environment, thereby improving the user's overall experience.

[0041] The present invention can quickly detect signs of mismatch through mirror reflection analysis in the early stage of assembly, thereby timely discovering and solving potential assembly risks and reducing later maintenance costs. Through this real-time monitoring and adjustment mechanism, quality problems caused by assembly defects are avoided, and the stability and reliability of the product are improved. The present invention introduces Internet of Things technology, so that the assembled mirror data can be uploaded and analyzed in real time without frequent manual inspection. For poorly adapted lenses, the system can efficiently perform optimization operations through automatic classification and feedback of machine learning models, reducing the time cost of manual debugging and significantly improving assembly efficiency and automation level. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0043] Figure 1 This is a schematic diagram of the mirror assembly effect evaluation method based on the Internet of Things in the present invention.

[0044] Figure 2 This is a schematic diagram of the mirror assembly effect evaluation system based on the Internet of Things in the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Reference Figure 1 - Figure 2 The following embodiments are obtained:

[0047] Embodiment 1: A mirror assembly effect evaluation method based on the Internet of Things comprises the following steps:

[0048] The use information of the assembled mirror is collected through the IoT sensor to obtain the use information set; the purpose of this step is to collect data on the mirror in the actual use environment. Through the IoT sensor, the system can collect information such as the mirror reflection light intensity, ambient brightness, clarity, etc. in real time to form a complete use information set. This information provides data support for subsequent analysis and judgment, so that the evaluation is based on the actual use environment.

[0049] Perform specular reflection analysis based on the usage information set, and then determine whether there are early signs that the mirror assembly effect is not compatible with the usage environment based on the specular reflection analysis results; this step aims to determine whether the mirror assembly effect meets expectations in the current usage environment through specular reflection data analysis. By analyzing the reflection intensity uniformity and brightness fluctuations at the connection, potential signs of mismatch can be detected early. If there are signs of mismatch, adjustments can be made in time before the problem expands.

[0050] When there are early signs that the mirror assembly effect is not compatible with the use environment, light adaptability characteristics and visual aesthetic characteristics are extracted from the use information set, and light adaptability index and visual aesthetic index are generated respectively; when early signs of mismatch are detected, light adaptability and visual aesthetic characteristics are further extracted from the use information set. The light adaptability index reflects the reflective adaptation effect of the lens under different brightness conditions, while the visual aesthetic index reflects the clarity and visual consistency of the lens. These two indices provide quantitative indicators for the accuracy and aesthetics of the assembly effect, and provide a basis for subsequent optimization of assembly.

[0051] The light adaptability index and visual aesthetics index are input into the pre-trained machine learning model, and the assembly fit is output. According to the assembly fit, the lenses for assembly are divided into two types: highly fit and poorly fit. The goal of this step is to use the machine learning model to make an overall evaluation of the lens assembly effect. The pre-trained polynomial regression model calculates the assembly fit based on the light adaptability and visual aesthetics index, and divides the assembled lenses into highly fit and poorly fit. The classification results can help the system determine which lenses meet the requirements and which lenses need further adjustment.

[0052] The lenses with low-fitting assembly are reassembled and feature extraction is performed again until all the lenses for assembly are classified as highly-fitting assembly types. When some lenses are judged to be low-fitting, the system will trigger the reassembly process, adjust the position and angle of these lenses, and collect feature data again. This step is repeated through iterations, so that all lenses eventually meet the highly-fitting assembly requirements, ensuring that the assembly effect of the overall mirror in the use environment meets the design standards and has the best visual experience and light adaptation effect.

[0053] The mirror is assembled from multiple lenses, which are numbered separately, and each lens corresponds to a unique number. In this mirror assembly system, each lens is numbered and corresponds to a unique number, which helps to track and analyze each lens individually during the multi-lens assembly process. This numbering system ensures that when collecting light adaptability and visual aesthetics data, the specific characteristics and assembly effects of each lens can be accurately located and identified, which facilitates individual adjustment and optimization, and improves the management accuracy of the entire assembly process.

[0054] Specular reflection analysis refers to:

[0055] When the assembled mirror is used in an application environment, the application environment is preset with multiple lighting modes. In each brightness mode, sampling data obtained by sampling the mirror reflected light intensity at the connection is found from the usage information set, so as to obtain comprehensive connection data in each brightness mode, and obtain multiple groups of sampling data. Each group of sampling data corresponds to a unique lighting mode, and then the average value and standard deviation of each group of sampling data are calculated respectively to obtain a stability combination [PJi, BZi] for reflecting the reflection intensity; i represents the type number of the lighting mode, PJi represents the average value of the sampling data of the i-th lighting mode, and BZi represents the standard deviation of the sampling data of the i-th lighting mode.

[0056] A variety of lighting modes are preset in the application environment. By sampling the intensity of mirror reflection light under different brightness modes, the adaptability of the mirror assembly effect under various lighting conditions can be more comprehensively evaluated. This method can ensure that the mirror has a stable visual effect under different ambient brightness. Sampling the intensity of reflected light at the joints of the mirrors ensures that the splicing effect between the lenses is good, without obvious brightness differences or uneven reflections. Since the joints are the areas where light adaptation and visual consistency are most likely to have problems, special testing of the data at the joints can effectively judge the assembly quality.

[0057] Judging from the results of the mirror reflection analysis, early signs of whether the mirror assembly effect is not suitable for the use environment refer to:

[0058] Obtain the mirror reflection analysis result, i.e., the stability combination [PJi, BZi]; the i-th lighting mode is preset with mirror reflection standard range 1 and mirror reflection standard range 2. If the stability combination PJi intersects with the mirror reflection standard range 1 and the stability combination BZi intersects with the mirror reflection standard range 2, a normal signal is generated. If the stability combination PJi intersects with the mirror reflection standard range 1 and the stability combination BZi intersects with the mirror reflection standard range 2, an abnormal signal is generated. If an abnormal signal is generated in any lighting mode, it indicates that there are early signs that the mirror assembly effect is not compatible with the use environment.

[0059] The "stability combination" of the reflected light intensity in each lighting mode reflects the light stability of the mirror in different lighting modes. If the stability combination meets the standard range in all preset lighting modes, it indicates that the assembly effect of the mirror is stable and reliable; otherwise, there may be early signs of mismatch. If the stability combination does not meet the standard range in any lighting mode, an "abnormal signal" can be generated, indicating that there may be early signs of mismatch in the assembly effect. This early warning mechanism enables detection and adjustment before the problem expands, improving the fit and quality of the overall assembly. Through multi-mode sampling and calculation of stability combinations, mirror reflection analysis can systematically evaluate the fit of the assembled mirror in various lighting environments, ensure that there are no obvious defects at the joints, and achieve high-quality assembly effects.

[0060] The light adaptability index is designed to quantify the light adaptation effect of a single lens under various ambient lighting conditions. This index reflects the reflection stability and adaptability of the lens to light under different lighting modes, which helps to determine whether the assembly effect of the lens in the use environment meets expectations. The larger the light adaptability index, the higher the adaptability of the lens to light, and the better it can reflect and adapt to the current ambient lighting conditions. This usually means that under various lighting modes, the reflection effect and light stability of the lens are better, and the ability to adapt to the environment is stronger. The smaller the light adaptability index, the worse the light adaptability of the lens under certain lighting modes, the reflection effect may be uneven or unstable, and there is a risk of mismatch. The logic for obtaining the light adaptability index is:

[0061] Extract the light reflection intensity data of each individual lens using the light sensor, and perform multiple samplings under different lighting modes. The extracted light adaptability characteristic data includes the average reflected light intensity and the standard deviation of the reflected light brightness fluctuation under each lighting mode;

[0062] For each lens j, a light adaptability coefficient LS I jm is generated according to the light adaptability characteristics. This coefficient represents the degree of light adaptation of the lens j under the current lighting conditions. The calculation formula is:

[0063] L represents the brightness of light under the current lighting conditions, that is, the current lighting mode of type m, which can also be understood as the intensity of ambient light. This value is used to standardize the reflected light intensity data so that the results under different brightness modes can be reasonably compared. Ravg represents the average reflected light intensity under the current lighting conditions, which represents the reflection effect of lens j under the current brightness conditions. The higher this value is, the stronger the reflection of the lens is. σR represents the standard deviation of the reflected light brightness fluctuation under the current lighting conditions, which is used to evaluate the reflection stability of the lens. A lower standard deviation indicates a more stable reflection effect. k represents the preset fluctuation adjustment coefficient. LSIjm represents the degree of adaptation of lens j to light under the current lighting mode of type m.

[0064] The following processing is performed on all light modes to obtain the light adaptability index of each lens under the full brightness range: U represents the total number of light modes, indicating how many different brightness conditions need to be considered. Q is the preset sensitivity coefficient. The Q value is greater than 1, which is used to highlight the influence of the adaptability differences between different modes, so that the final result can more sensitively reflect the adaptability of different brightness modes. total It represents the light adaptability index of lens j in the full brightness range, and the overall light adaptability effect of lens j under various brightness conditions.

[0065] The visual aesthetics index is used to quantify the visual aesthetics of a single lens in different brightness modes. The index reflects the comprehensive effects of the lens in terms of image clarity, light distribution uniformity, and visual contrast, helping to determine whether the lens presents a good visual experience in the assembly environment. The larger the visual aesthetics index, the better the aesthetics of the lens in the current assembly environment and the better the visual performance. A higher index means excellent performance in image clarity, light distribution uniformity, and visual contrast, and can provide more comfortable visual effects in various brightness modes. A smaller visual aesthetics index may indicate that the lens has poor aesthetics in certain lighting modes, and clarity, uniformity, or contrast may be insufficient, affecting the overall visual effect. The logic for obtaining the visual aesthetics index is:

[0066] The visual aesthetic characteristic indexes of each individual lens are extracted, including the image clarity index, the light distribution uniformity index, and the visual contrast index, and then the visual aesthetic coefficient of lens j under the current lighting conditions, that is, the current lighting mode of type m, is calculated. The calculation formula is:

[0067] QXm represents the corresponding value of the image clarity index of lens j in the current lighting mode of type m, which represents the clarity of lens j in the current lighting mode m. The corresponding value is obtained by image processing algorithm analysis. A higher value means higher image clarity and better visual performance of the lens. GXm represents the corresponding value of the light distribution uniformity index of lens j in the current lighting mode of type m, which represents the light distribution uniformity of lens j in the current lighting mode m. The corresponding value is calculated by the light intensity distribution on the mirror surface. A lower value indicates a more uniform light distribution. SDm represents the corresponding value of the visual contrast index of lens j in the current lighting mode of type m, which reflects the contrast effect of lens j in the current lighting mode m, and is usually calculated by the brightness difference between bright and dark areas. A higher value means a stronger contrast and a more vivid visual performance. VSIjm represents the visual aesthetics coefficient of lens j in the current lighting mode of type m. w1, w2, and w3 are all preset non-zero proportional coefficients used to adjust the influence of clarity, uniformity, and contrast on visual aesthetics.

[0068] The formula for calculating the visual aesthetics index is:

[0069] VSI total It represents the visual aesthetic index of lens j in the full brightness range, which is obtained by the nonlinear combination of the visual aesthetic coefficients under all lighting modes, and represents the overall visual aesthetic performance of lens j in the full brightness range.

[0070] The logic for obtaining the image clarity index is: analyze the mirror reflection image through the image processing algorithm, use edge detection to quantify the clarity, and the corresponding value of the image clarity index is expressed as the average value of the edge intensity; the image clarity index reflects the visual clarity of the lens under the current lighting conditions. The higher the edge intensity, the clearer the image and the better the visual performance of the lens. The clarity index is used to quantify the visual quality of the lens to ensure that it can provide good visual effects under different lighting conditions.

[0071] The logic for obtaining the light distribution uniformity index is: the light sensor collects the light intensity at multiple locations on the mirror surface and calculates its mean square error. The result is the corresponding value of the light distribution uniformity index; the light distribution uniformity index reflects whether the light distribution of the lens is uniform under the current lighting conditions. The smaller the mean square error, the more uniform the light distribution, indicating that there is no obvious brightness difference on the lens surface. Uniform light distribution helps improve the overall visual experience and reduce visual discomfort and discontinuity.

[0072] The logic for obtaining the visual contrast index is: in different brightness areas, the contrast is calculated by the brightness difference, specifically the average brightness difference between bright and dark areas:

[0073] Lm max Indicates the average brightness of the bright area, Lm min Indicates the average brightness of dark areas. The visual contrast index is used to measure the contrast performance of the lens under the current lighting conditions. A higher contrast index means that the lens can provide a clearer boundary between light and dark areas, making the visual effect more vivid and lively. Contrast is essential to the overall visual aesthetics and can enhance the details of the image.

[0074] The pre-trained machine learning model is a polynomial regression model, and the corresponding function expression is: SPj = F(LSIj total , VSIj total ); SPj represents the fitting degree of lens j.

[0075] According to the degree of fitting, the lenses used for fitting are divided into two types: high fitting and low fitting.

[0076] The assembly fitness of lens j is compared with the preset assembly fitness threshold. If the assembly fitness of lens j is greater than or equal to the preset assembly fitness threshold, lens j is classified as the assembly type of highly fit assembly. If the assembly fitness of lens j is less than the preset assembly fitness threshold, lens j is classified as the assembly type of low fit assembly.

[0077] By evaluating and classifying the lens's fit, it is possible to ensure that each lens meets the preset fit requirements after assembly. A lens with a high fit indicates that it has achieved an ideal fit in terms of light adaptability and visual aesthetics, and can meet the needs of the use environment. If a lens is classified as a low fit, it means that its fit is not ideal, which may affect the overall visual experience and light fit, and requires further adjustment or reassembly. Lenses classified as low fit can be marked, and the system can further optimize these lenses without additional processing of the already highly fit lenses. This method saves time and resources and makes the assembly process more efficient. Through iterative optimization, all lenses eventually meet the high fit standard, which can ensure the consistency and stability of the entire assembly system. The highly fit assembly type ensures the visual quality and light fit of the lens in the actual use environment, thereby improving the user experience. Users can get a seamless visual effect, and the lenses can maintain consistent light fit and visual aesthetics under various lighting conditions. Through fit classification, it is possible to identify and solve the misfit problem in assembly early, reducing the potential quality risks caused by low-fit lenses. This classification method helps to ensure that each lens meets the quality standards before the product is put into use, thereby reducing subsequent maintenance and replacement costs.

[0078] Embodiment 2: A mirror assembly effect evaluation system based on the Internet of Things, comprising:

[0079] An information collection module collects usage information of the assembled mirror through an IoT sensor to obtain a usage information set;

[0080] A mirror reflection analysis module performs mirror reflection analysis based on the usage information set, and then determines whether there is an early sign that the mirror assembly effect is not compatible with the usage environment based on the mirror reflection analysis result;

[0081] A feature extraction module, when there is an early sign that the mirror assembly effect is not compatible with the use environment, extracts light adaptability features and visual aesthetic features from the use information set, and generates a light adaptability index and a visual aesthetic index respectively;

[0082] The single lens type classification module inputs the light adaptability index and visual aesthetics index into the pre-trained machine learning model, outputs the assembly fitness, and classifies the lenses for assembly into two types of assembly: high-fit assembly and low-fit assembly according to the assembly fitness;

[0083] The assembly optimization module reassembles the lenses with low fit and re-extracts features until all the lenses for assembly are classified as highly fit.

[0084] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0085] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0086] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0087] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0088] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A mirror assembly effect evaluation method based on the Internet of Things, characterized in that: The following steps are involved: Collect usage information of the assembled mirror through IoT sensors to obtain a usage information set; Performing a specular reflection analysis based on the usage information set, and then judging whether there is an early sign that the mirror assembly effect is not compatible with the usage environment based on the specular reflection analysis results; When there is an early sign that the mirror assembly effect is not compatible with the use environment, light adaptability features and visual aesthetic features are extracted from the use information set, and a light adaptability index and a visual aesthetic index are generated respectively; The light adaptability index and the visual aesthetics index are input into the pre-trained machine learning model to output the assembly fitness. According to the assembly fitness, the lenses for assembly are divided into two types: high-fitting assembly and low-fitting assembly. The lenses for low-fitting assembly are reassembled and feature extraction is performed again until all the lenses for assembly are classified as the assembly type of high-fitting assembly.

2. The mirror assembly effect evaluation method based on the Internet of Things according to claim 1, characterized in that: The mirror is assembled from a plurality of lenses, and the plurality of lenses are numbered respectively, and each lens corresponds to a unique number.

3. The mirror assembly effect evaluation method based on the Internet of Things according to claim 2 is characterized in that: Specular reflection analysis refers to: When the assembled mirror is used in an application environment, the application environment is preset with multiple lighting modes. In each brightness mode, sampling data obtained by sampling the mirror reflected light intensity at the connection is found from the usage information set, so as to obtain comprehensive connection data in each brightness mode, and obtain multiple groups of sampling data. Each group of sampling data corresponds to a unique lighting mode, and then the average value and standard deviation of each group of sampling data are calculated respectively to obtain a stability combination [PJi, BZi] for reflecting the reflection intensity; i represents the type number of the lighting mode, PJi represents the average value of the sampling data of the i-th lighting mode, and BZi represents the standard deviation of the sampling data of the i-th lighting mode.

4. The mirror assembly effect evaluation method based on the Internet of Things according to claim 3 is characterized in that: Judging from the results of the mirror reflection analysis, early signs of whether the mirror assembly effect is not suitable for the use environment refer to: Obtain the mirror reflection analysis result, i.e., the stability combination [PJi, BZi]; the i-th lighting mode is preset with mirror reflection standard range 1 and mirror reflection standard range 2. If the stability combination PJi intersects with the mirror reflection standard range 1 and the stability combination BZi intersects with the mirror reflection standard range 2, a normal signal is generated. If the stability combination PJi intersects with the mirror reflection standard range 1 and the stability combination BZi intersects with the mirror reflection standard range 2, an abnormal signal is generated. If an abnormal signal is generated in any lighting mode, it indicates that there are early signs that the mirror assembly effect is not compatible with the use environment.

5. The mirror assembly effect evaluation method based on the Internet of Things according to claim 4 is characterized in that: The logic for obtaining the light adaptability index is: Extract the light reflection intensity data of each individual lens using the light sensor, and perform multiple samplings under different lighting modes. The extracted light adaptability characteristic data includes the average reflected light intensity and the standard deviation of the reflected light brightness fluctuation under each lighting mode; For each lens j, a light adaptability coefficient LSIjm is generated according to the light adaptability characteristics. This coefficient represents the degree of light adaptation of lens j under the current lighting conditions. The calculation formula is: L represents the brightness of light under the current lighting conditions, i.e., the current lighting mode of type m; Ravg represents the average reflected light intensity under the current lighting conditions; σR represents the standard deviation of the reflected light brightness fluctuation under the current lighting conditions; k represents the preset fluctuation adjustment coefficient; LSIjm represents the degree of adaptation of lens j to light under the current lighting mode of type m; The following processing is performed on all light modes to obtain the light adaptability index of each lens under the full brightness range: U represents the total number of light modes, Q is the preset sensitivity coefficient, Q is greater than 1, LSIj total Indicates the light adaptability index of lens j in the full brightness range.

6. The mirror assembly effect evaluation method based on the Internet of Things according to claim 5 is characterized in that: The logic for obtaining the visual aesthetics index is: The visual aesthetic characteristic indexes of each individual lens are extracted, including the image clarity index, the light distribution uniformity index, and the visual contrast index, and then the visual aesthetic coefficient of lens j under the current lighting conditions, that is, the current lighting mode of type m, is calculated. The calculation formula is: QXm represents the corresponding value of the image clarity index of lens j in the current lighting mode of type m, GXm represents the corresponding value of the light distribution uniformity index of lens j in the current lighting mode of type m, SDm represents the corresponding value of the visual contrast index of lens j in the current lighting mode of type m, VSIjm represents the visual aesthetics coefficient of lens j in the current lighting mode of type m, w1, w2, w3 are all preset non-zero proportional coefficients; The formula for calculating the visual aesthetics index is: VSI total It represents the visual aesthetic index of lens j in the full brightness range.

7. The method for evaluating mirror assembly effect based on the Internet of Things according to claim 6, characterized in that: The logic of obtaining the image clarity index is as follows: the specular reflection image is analyzed by an image processing algorithm, the clarity is quantified using edge detection, and the corresponding value of the image clarity index is expressed as the average value of the edge intensity; The logic for obtaining the light distribution uniformity index is as follows: the light sensor collects the light intensity at multiple positions on the mirror surface and calculates its mean square error. The result obtained is the corresponding value of the light distribution uniformity index. The logic for obtaining the visual contrast index is: in different brightness areas, the contrast is calculated by the brightness difference, specifically the average brightness difference between bright and dark areas: Lm max Indicates the average brightness of the bright area, Lm min Indicates the average brightness of dark areas.

8. The mirror assembly effect evaluation method based on the Internet of Things according to claim 7 is characterized in that: The pre-trained machine learning model is a polynomial regression model.

9. The mirror assembly effect evaluation method based on the Internet of Things according to claim 8, characterized in that: According to the degree of fitting, the lenses used for fitting are divided into two types: high fitting and low fitting. The assembly fitness of lens j is compared with the preset assembly fitness threshold. If the assembly fitness of lens j is greater than or equal to the preset assembly fitness threshold, lens j is classified as the assembly type of highly fit assembly. If the assembly fitness of lens j is less than the preset assembly fitness threshold, lens j is classified as the assembly type of low fit assembly.

10. A mirror assembly effect evaluation system based on the Internet of Things, implemented based on the mirror assembly effect evaluation method based on the Internet of Things according to any one of claim 19, characterized in that: include: An information collection module collects usage information of the assembled mirror through an IoT sensor to obtain a usage information set; A mirror reflection analysis module performs mirror reflection analysis based on the usage information set, and then determines whether there is an early sign that the mirror assembly effect is not compatible with the usage environment based on the mirror reflection analysis result; A feature extraction module, when there is an early sign that the mirror assembly effect is not compatible with the use environment, extracts light adaptability features and visual aesthetic features from the use information set, and generates a light adaptability index and a visual aesthetic index respectively; The single lens type classification module inputs the light adaptability index and visual aesthetics index into the pre-trained machine learning model, outputs the assembly fitness, and classifies the lenses for assembly into two types of assembly: high-fit assembly and low-fit assembly according to the assembly fitness; The assembly optimization module reassembles the lenses with low fit and re-extracts features until all the lenses for assembly are classified as highly fit.

Images