A comprehensive quality evaluation method based on the mechanical properties of wigs
By acquiring mechanical testing data of wig samples, a comprehensive evaluation model based on machine learning was constructed, which solved the problems of subjectivity and insufficient comprehensive judgment in wig quality evaluation, and realized objective and comprehensive scoring and ranking of wig quality.
Patent Information
- Application Number
- CN202610367052.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-03-24
- Publication Date
- 2026-06-30
AI Technical Summary
Existing methods for evaluating wig quality rely on human experience, resulting in highly subjective evaluation results. Single-indicator testing is insufficient to comprehensively reflect wig quality, leading to inadequate overall judgment capabilities.
By acquiring the original mechanical testing data of wig samples, multiple mechanical evaluation indicators are extracted, a comprehensive evaluation model based on machine learning algorithms is constructed, and the quality score results are output.
This approach achieves objectivity, quantification, and comprehensiveness in wig quality evaluation, improves the accuracy and stability of evaluation results, identifies key influencing factors, and provides a basis for quality control and product selection.
Smart Images
Figure CN122309969A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of quality inspection and intelligent evaluation technology, specifically to a comprehensive quality evaluation method based on the mechanical properties of wigs, which can be applied to scenarios such as comprehensive quality evaluation, quality grading, quality control, and product screening of wig products. Background Technology
[0002] As wigs are increasingly used in daily consumer goods, beauty styling, medical applications, and cross-border e-commerce, the market is demanding higher standards for the objectivity, scientific rigor, and standardization of wig quality evaluation. However, existing wig quality evaluation methods still have significant limitations.
[0003] First, current wig quality assessments rely heavily on human experience. Evaluators typically make subjective judgments based on factors such as feel, smoothness, pulling sensation, and appearance. This approach is easily influenced by individual experience, evaluation standards, and differences in usage scenarios, resulting in inconsistent evaluation results and making it difficult to establish unified, objective, and reproducible quality evaluation standards.
[0004] Secondly, current technologies for testing wig quality often focus on a single performance indicator, such as testing only a tensile strength, breaking performance, or a single mechanical parameter. Since wig quality is essentially the result of the combined effects of multiple mechanical properties, a single indicator cannot comprehensively reflect the overall quality level of a wig sample.
[0005] Furthermore, there is usually a certain correlation and coupling between different mechanical properties of wigs. For example, indicators such as toughness, elastic recovery ability, elongation at break and stability often jointly affect the final user experience. Traditional evaluation methods are difficult to systematically integrate the above-mentioned multi-indicator information, resulting in insufficient comprehensive judgment ability.
[0006] Therefore, there is an urgent need to provide an evaluation method that can comprehensively analyze multiple mechanical properties of wigs and output quality scores, thereby achieving objectivity, quantification, and intelligence in wig quality evaluation. Summary of the Invention
[0007] Purpose of the invention
[0008] The purpose of this invention is to address the problems of strong subjectivity, single evaluation dimensions, and insufficient comprehensive judgment ability in existing wig quality evaluation methods, and to provide a comprehensive quality evaluation method based on the mechanical properties of wigs.
[0009] This invention obtains the original mechanical testing data of wig samples, extracts multiple mechanical evaluation indicators to characterize wig quality, and uses these evaluation indicators to construct a comprehensive evaluation model, thereby achieving comprehensive scoring, quality grading, or ranking evaluation of wig quality.
[0010] Technical solution
[0011] To achieve the above objectives, the present invention adopts the following technical solution:
[0012] A comprehensive quality evaluation method based on the mechanical properties of wigs includes the following steps:
[0013] Step S1: Obtain the original mechanical test data of the wig sample to be tested;
[0014] Step S2: Extract multiple mechanical evaluation indicators to characterize the quality of the wig based on the original mechanical test data to form sample feature data;
[0015] Step S3: Preprocess the sample feature data, including one or more of outlier identification, missing value handling, and standardization.
[0016] Step S4: Construct a comprehensive evaluation model for wig quality based on the sample feature data;
[0017] Step S5: Output the comprehensive quality score of the wig sample to be tested based on the comprehensive wig quality evaluation model.
[0018] Furthermore, the original mechanical test data includes one or more of the following: force data, displacement data, time series data, stress change data, and strain change data generated during the tensile testing of the wig sample.
[0019] Furthermore, the mechanical evaluation indicators include one or more of the following: tensile strength, elongation at break, elastic recovery, toughness, compliance, stability, curve slope, and energy integral.
[0020] Furthermore, the preprocessing also includes one or more of the following: data smoothing, normalization, noise reduction, and feature selection.
[0021] Furthermore, the comprehensive evaluation model is an evaluation model constructed based on machine learning algorithms.
[0022] Furthermore, the machine learning algorithm includes one or more of the following: gradient boosting tree algorithm, support vector machine algorithm, random forest algorithm, neural network algorithm, and ensemble learning algorithm.
[0023] Furthermore, the comprehensive quality score result is one or more of the following: continuous numerical score, quality grade result, or quality ranking result.
[0024] Furthermore, while outputting the comprehensive quality score, it can also output information on the contribution of each mechanical evaluation index to the score and / or the identification results of key influencing factors.
[0025] Beneficial effects
[0026] Compared with the prior art, the present invention has the following beneficial effects:
[0027] First, this invention is based on multiple mechanical properties of wig samples, breaking through the traditional evaluation methods that rely on experience-based judgment or single-index testing, and improving the objectivity of wig quality evaluation.
[0028] Secondly, by extracting multiple mechanical evaluation indicators and constructing a comprehensive evaluation model, this invention achieves multi-dimensional analysis of wig quality and improves the comprehensiveness of the evaluation results.
[0029] Third, this invention can uniformly map multiple mechanical test information into a comprehensive quality score, grade result or ranking result, which improves the accuracy, stability and practicality of wig quality evaluation.
[0030] Fourth, this invention can also identify key mechanical indicators that affect the quality score of wigs, providing a basis for enterprises to carry out quality control, process optimization and product selection. Attached Figure Description
[0031] Figure 1 This is a flowchart illustrating a comprehensive quality evaluation method for wigs based on their mechanical properties, as described in this invention. Figure 2 This is a schematic diagram illustrating the conversion of raw mechanical test data into mechanical evaluation indicators in this invention. Figure 3 This is a schematic diagram illustrating the construction and output of the comprehensive evaluation model for wig quality in this invention. Detailed Implementation
[0032] The present invention will be further described below with reference to specific embodiments, but the scope of protection of the present invention is not limited to the following embodiments.
[0033] Implementation Method 1: Overall Process
[0034] This embodiment provides a comprehensive quality evaluation method based on the mechanical properties of wigs.
[0035] First, the raw mechanical testing data of the wig sample to be tested is obtained. The raw mechanical testing data can be derived from the data generated during the tensile testing of the wig sample, and may specifically include force data, displacement data, time series data, stress change data, and strain change data.
[0036] Secondly, multiple mechanical evaluation indicators for characterizing wig quality are extracted from the original mechanical testing data. These mechanical evaluation indicators include, but are not limited to, tensile strength, elongation at break, elastic recovery, toughness, flexibility, stability, curve slope characteristics, and energy integral characteristics. Through the above indicator extraction process, the original test curves and sequence data can be transformed into sample characteristic data suitable for subsequent modeling and analysis.
[0037] Then, the sample feature data is preprocessed. The preprocessing includes outlier identification, missing value handling, and standardization, and may further include data smoothing, normalization, noise reduction, and feature selection to improve the stability and generalization ability of the subsequent evaluation model.
[0038] Furthermore, a comprehensive evaluation model for wig quality is constructed based on the preprocessed sample feature data. This evaluation model can be a model built using machine learning algorithms, such as a gradient boosting tree model, a support vector machine model, a random forest model, a neural network model, or other ensemble learning models. During the model training phase, historical sample data with quality labels can be used. These quality labels can originate from expert ratings, quality inspection levels, standard sample levels, or other available evaluation results.
[0039] Finally, the feature data of the wig samples to be tested are input into the trained comprehensive wig quality evaluation model, and the comprehensive quality score is output. The result can be a continuous numerical score, or it can be further mapped to quality levels such as excellent, good, medium, and poor, or it can be used to generate sample ranking results.
[0040] Implementation Method 2: Extraction of Mechanical Evaluation Indicators
[0041] In this embodiment, the mechanical evaluation index is extracted from the original mechanical test data.
[0042] Suppose that a total of [number] wig samples were obtained during the testing process. There are 1 sampling points, and the force value corresponding to each sampling point is denoted as . ,in Then the maximum tensile force can be expressed as:
[0043]
[0044] in, This indicates the maximum stress value of the wig sample during the testing process, and can be used as an evaluation index related to tensile strength.
[0045] Furthermore, let the initial length of the wig sample be... The length at break is Then the elongation at break can be expressed as:
[0046]
[0047] in, It indicates the relative elongation of a sample at fracture and can be used to characterize the sample's ductility.
[0048] In some implementations, if the sample length recovers to its original value after loading is completed... The elastic recovery rate can then be expressed as:
[0049]
[0050] in, It indicates the sample's ability to recover after the force is released and can be used to measure its elastic recovery performance.
[0051] Furthermore, if the integral area of the wig sample under the force-displacement curve is denoted as... Therefore, this area can be used as an energy integral feature to characterize the sample's overall energy absorption capacity during the stress process, that is:
[0052]
[0053] in, Indicates the displacement at the time of fracture. Displacement is The force function at time.
[0054] Using the above method, a set of characteristic variables that can comprehensively characterize the quality of wigs can be extracted from the original mechanical test data.
[0055] Implementation Method 3: Construction of a Comprehensive Evaluation Model
[0056] In this embodiment, a comprehensive evaluation model is constructed based on machine learning methods.
[0057] Let the sample feature vector obtained after preprocessing be:
[0058]
[0059] in, express One mechanical evaluation index.
[0060] The overall quality score can be expressed as:
[0061]
[0062] in, This represents the overall quality score of the wig sample to be tested. This represents the comprehensive evaluation model for wig quality obtained through training.
[0063] Specifically, multiple wig samples with known quality levels are used as training samples. Their corresponding mechanical evaluation indicators are extracted and combined with relevant quality labels to construct a training set. Then, machine learning algorithms are used to train the model, resulting in a comprehensive wig quality evaluation model. After the model training is complete, the feature data of the wig sample to be tested is input into the model to obtain the corresponding comprehensive quality score.
[0064] Preferably, if the output result is a continuous score, the scoring interval can be divided into multiple quality level intervals. Let the quality level thresholds be respectively... , and ,and The quality grade It can be represented as:
[0065]
[0066] The above method enables the conversion from continuous quality scoring to graded results, thus meeting the grading requirements in actual quality management scenarios.
[0067] Implementation Method 4: Interpretation of Evaluation Results
[0068] In this embodiment, in addition to outputting the overall quality score, key indicator information that affects the evaluation results can also be output.
[0069] For example, let the first The contribution of each mechanical evaluation index to the overall score is: Then the contribution of each indicator can be normalized and expressed as:
[0070]
[0071] in, Indicates the first The relative contribution of each indicator to the overall quality score. This indicates the total number of indicators.
[0072] Specifically, by using feature importance analysis, contribution analysis, or other model interpretation methods, mechanical evaluation indicators that have a significant impact on the overall quality score can be identified. This helps users further understand the main mechanical factors that contribute to the quality of wig samples and provides a basis for product optimization and process adjustment.
Claims
1. A wig mechanical property-based quality comprehensive evaluation method, characterized by comprising the following steps: Specifically, the steps include the following: S1: Obtain the original mechanical test data of the wig sample to be tested; S2: Extract features from the original mechanical test data to obtain multiple mechanical evaluation indicators for characterizing the quality of the wig, forming sample feature data; S3: Perform preprocessing operations on the sample feature data to obtain preprocessed sample feature data. The preprocessing operations include one or more of outlier identification, missing value handling, and standardization. S4: Based on the preprocessed sample feature data, construct a comprehensive evaluation model for wig quality; S5: Input the preprocessed sample feature data corresponding to the wig sample to be tested into the comprehensive evaluation model for wig quality to obtain the comprehensive quality score result of the wig sample to be tested.
2. The method according to claim 1, characterized in that: In step S1, the original mechanical test data specifically includes one or more of the following: force data, displacement data, time series data, stress change data, and strain change data generated by the wig sample during the mechanical test.
3. The method according to claim 1, characterized in that: In step S2, the mechanical evaluation index includes one or more of the following: tensile strength index, elongation at break index, elastic recovery index, toughness characterization index, compliance characterization index, stability index, curve slope characteristic, and energy integral characteristic.
4. The comprehensive quality evaluation method based on the mechanical properties of wigs according to claim 1, characterized in that: In step S2, the calculation formulas used for the maximum tensile force index and the elongation at break index are as follows: in, The maximum tensile force index. For the first The force value corresponding to each sampling point The total number of sampling points. Elongation at break is an indicator. This is the initial length of the wig sample. This is the length of the wig sample when it breaks.
5. The comprehensive quality evaluation method based on the mechanical properties of wigs according to claim 1, characterized in that: In step S2, the calculation formulas used for the elastic recovery index and energy integral characteristic are as follows: ,in, As an elasticity response indicator, To determine the sample recovery length after unloading. It is an energy integral characteristic. This represents the displacement at the time of fracture. For displacement The force function corresponding to the time.
6. The comprehensive quality evaluation method based on the mechanical properties of wigs according to claim 1, characterized in that: In step S3, the calculation formula used for standardizing the sample feature data is as follows: ,in, For the first The original values of each mechanical evaluation index, The standardized index value, For the first The mean of each mechanical evaluation index in the sample set. For the first The standard deviation of each mechanical evaluation index in the sample set To prevent zero items.
7. The comprehensive quality evaluation method based on the mechanical properties of wigs according to claim 1, characterized in that: In step S4, the comprehensive evaluation model for wig quality is an evaluation model built based on machine learning algorithms. The machine learning algorithms used include one or more of the following: gradient boosting tree algorithm, support vector machine algorithm, random forest algorithm, neural network algorithm, and ensemble learning algorithm.
8. The comprehensive quality evaluation method based on the mechanical properties of wigs according to claim 1, characterized in that: In steps S4 and S5, the expression used for the overall quality score result is as follows: ,in, This represents the overall quality score of the wig sample to be tested. These are multiple mechanical evaluation indicators after preprocessing. This is the comprehensive evaluation model for wig quality obtained through training.
9. The comprehensive quality evaluation method based on the mechanical properties of wigs according to claim 1, characterized in that: In step S5, the overall quality score result can be further converted into a quality grade result. The calculation expression used for the quality grade result is as follows: ,in, For quality grade results, To provide a comprehensive quality score, , , The preset quality level threshold is met. .
10. The comprehensive quality evaluation method based on the mechanical properties of wigs according to claim 1, characterized in that: Along with the overall quality score, the contribution of each mechanical evaluation index to the overall quality score is also output. The calculation formula used is as follows: ,in, For the first The relative contribution ratio of each mechanical evaluation index to the overall quality score result. For the first The contribution of each mechanical evaluation index This represents the total number of mechanical evaluation indicators.