CN-121049252-B - Intelligent detection method for surface defects of protective film based on machine vision

CN121049252BCN 121049252 BCN121049252 BCN 121049252BCN-121049252-B

Abstract

The application provides an intelligent detection method for surface defects of a protective film based on machine vision, which comprises the steps of extracting stress concentration areas and touch signal attenuation characteristics from potential risks of touch response delay, carrying out grouping analysis on the stress concentration areas and the touch signal attenuation characteristics to obtain a combined attenuation trend of optical transmittance and touch signal attenuation, identifying deformation degree of the stress concentration areas, determining touch signal attenuation degree of a folding area according to the combined attenuation trend of the optical transmittance and touch signal attenuation, evaluating and obtaining durability of a folding screen, extracting early warning indexes about screen curvature radius and light scattering intensity defects from attenuation trend prediction results, and generating a complete report containing the optical transmittance and touch signal attenuation defects.

Inventors

FENG BIN
XU JINJIA
YANG YONGHUA
LI YIWEI

Assignees

佛山市佳世达薄膜科技有限公司

Dates

Publication Date: 20260505
Application Date: 20250813

Claims (10)

1. The intelligent detection method for the surface defects of the protective film based on machine vision is characterized by comprising the following steps of: The method comprises the steps of constructing a reflectivity and stress distribution data set by collecting polarized light reflectivity and stress distribution data of a folding screen from a flat state to a limit folding angle, and carrying out feature extraction on the reflectivity and stress distribution data set to obtain an initial optical performance attenuation index, wherein the initial optical performance attenuation index comprises a reflectivity attenuation coefficient and a stress distribution non-average value; calculating a display definition value according to the polarized light reflectivity and stress distribution data, determining a definition decline trend by analyzing the corresponding relation between the initial attenuation index of the optical performance and the display definition value, extracting a touch response delay characteristic based on the definition decline trend, extracting a stress concentration region and a touch signal attenuation characteristic from the touch response delay characteristic, wherein the touch signal attenuation characteristic comprises a signal delay time and an attenuation amplitude, carrying out grouping analysis on the stress concentration region and the touch signal attenuation characteristic to obtain a joint attenuation trend of optical transmittance and touch signal attenuation, calculating the attenuation degree of the touch signal of a folding region according to the joint attenuation trend of the optical transmittance and the touch signal attenuation, combining the deformation degree of the stress concentration region to obtain a durability index, calculating a comprehensive performance attenuation index comprising optical transmittance attenuation, touch response delay and material fatigue degree, determining a folding angle change rate and a material accumulation angle change and refraction index according to the comprehensive performance attenuation index, predicting the relation between the folding angle change and the optical refraction index, obtaining a scattering angle change rate and the material fatigue strength from the prediction result, predicting the radius of the folding angle change, obtaining a scattering result, and obtaining early warning index data.
2. The intelligent detection method for surface defects of a protective film based on machine vision according to claim 1, wherein the steps of constructing a reflectivity and stress distribution data set by collecting polarized light reflectivity and stress distribution data of a folding screen from a flat state to a limit folding angle, and performing feature extraction on the reflectivity and stress distribution data set to obtain an initial attenuation index of optical performance include: Collecting polarized light reflection signals according to preset angle intervals, simultaneously obtaining pressure values of all measuring points under corresponding angles, constructing a reflectivity and stress distribution data set, calculating the difference value between the reflectivity under each folding angle and the reflectivity in an initial flat state to obtain an attenuation proportion by dividing the initial reflectivity, performing linear fitting on the folding angles and the attenuation proportion by a least square method, taking the slope of a fitting straight line as a reflectivity attenuation coefficient, calculating the sum of squares of the difference values of stress values and average stress of all the measuring points under each folding angle to obtain a standard deviation by dividing the number of the measuring points, and taking the ratio of the standard deviation to the average stress as an uneven stress distribution to obtain an initial attenuation index of optical performance containing the reflectivity attenuation coefficient and the uneven stress distribution.
3. The intelligent machine vision-based protective film surface defect detection method according to claim 1, wherein calculating a display sharpness value according to the polarized light reflectivity and stress distribution data, determining a sharpness decrease trend by analyzing a correspondence between the initial attenuation index of the optical performance and the display sharpness value, comprises: Generating a reflectivity distribution matrix according to the polarized light reflectivity, constructing a stress field distribution diagram according to the stress distribution data, calculating the spatial frequency response of a display area to obtain the display definition value, and generating a definition decline trend curve by processing the display definition value and the initial optical performance attenuation index through regression analysis.
4. The intelligent detection method for surface defects of a protective film based on machine vision according to claim 1, wherein the extracting stress concentration areas and touch signal attenuation characteristics from the touch response delay characteristics, wherein the touch signal attenuation characteristics include signal delay time and attenuation amplitude, and performing grouping analysis on the stress concentration areas and touch signal attenuation characteristics to obtain a combined attenuation trend of optical transmittance and touch signal attenuation comprises: The method comprises the steps of receiving a touch response delay characteristic, reading space coordinates in the touch response delay characteristic, determining the boundary of the stress concentration area, collecting touch signal waveforms in the boundary, calculating the signal delay time and the attenuation amplitude, constructing a characteristic combination matrix according to the signal delay time group, measuring the optical transmittance of the position in the matrix, generating a regression equation of the transmittance and the attenuation amplitude, and fitting to obtain a combined attenuation trend curve.
5. The intelligent detection method for surface defects of a protective film based on machine vision according to claim 1, wherein the calculating the touch signal attenuation degree of a folded region according to the combined attenuation trend of the optical transmittance and the touch signal attenuation, and combining the deformation degree of the stress concentration region, to obtain the durability index comprises: scanning the surface profile of the stress concentration area, calculating the local curvature radius, generating a deformation degree distribution map, acquiring the touch signal attenuation degree according to the joint attenuation trend, calculating the damage rate, and fitting a degradation curve to obtain the durability index.
6. The intelligent machine vision-based protective film surface defect detection method according to claim 1, wherein the calculating of the comprehensive performance degradation index including optical transmittance degradation, touch response delay and material fatigue by the durability index and the deformation degree includes: Obtaining the maximum value of the durability index and the deformation degree, calculating the comprehensive performance attenuation index through weighted summation, measuring the polarization state change of a folding area, calculating the optical refractive index in an inversion mode, extracting the refractive index change rate according to the relation between the optical refractive index and the folding angle, and if the refractive index change rate exceeds a threshold value, combining components in the comprehensive performance attenuation index to generate an early warning signal.
7. The intelligent machine vision-based protective film surface defect detection method according to claim 1, wherein the calculating of the comprehensive performance degradation index including optical transmittance degradation, touch response delay and material fatigue by the durability index and the deformation degree includes: Collecting folding times and environmental temperature data, measuring the surface hardness and tensile strength of a protective film to generate a material mechanical property data set, scanning the distribution of deformation degree to determine the boundary of a deformation area, measuring the optical transmittance, touch response time and material elastic modulus in the boundary, calculating attenuation rate and fatigue life index, and determining the contribution duty ratio of material aging and stress concentration.
8. The intelligent detection method for surface defects of a protective film based on machine vision according to claim 1, wherein the analyzing the relationship between the change of the folding angle and the optical refractive index according to the comprehensive performance attenuation index, determining the change rate of the folding angle and the material fatigue accumulation value, and obtaining the attenuation trend prediction result comprises: measuring the deformation depth of the stress concentration area, calculating the relative damage amount, accumulating to obtain the material fatigue accumulation value, recording a folding angle sequence, calculating the folding angle change rate, and adjusting the folding angle change rate and the material fatigue accumulation value according to the early warning level of the comprehensive performance attenuation index to generate a predicted attenuation value sequence.
9. The intelligent detection method for surface defects of a protective film based on machine vision according to claim 1, wherein the analyzing the relationship between the change of the folding angle and the optical refractive index according to the comprehensive performance attenuation index, determining the change rate of the folding angle comprises: And extracting stress distribution according to the evolution track, calculating the angle response of the rigidity of the material and the external load, and determining the change rate of the folding angle.
10. The intelligent detection method for surface defects of a protective film based on machine vision according to claim 1, wherein the step of extracting the radius of curvature of a screen and the light scattering intensity from the attenuation trend prediction result to obtain early warning index data comprises the following steps: And (3) calculating the radius of curvature of the screen and the light scattering intensity according to the attenuation trend prediction result, marking an early warning state, extracting the optical transmittance and the touch signal response time of the corresponding position, classifying the defect types, and generating a data summary table containing risk levels.

Description

Intelligent detection method for surface defects of protective film based on machine vision Technical Field The invention relates to the technical field of information, in particular to an intelligent detection method for surface defects of a protective film based on machine vision. Background The folding screen technology is a core breakthrough in the field of intelligent equipment, the folding characteristic of the folding screen technology greatly expands the functionality and portability of mobile equipment, and the folding screen technology has a key meaning for improving user experience and promoting industrial upgrading. However, the existing performance analysis method of the folding screen protective film has significant defects in the dynamic use scene. The methods rely on static tests, so that the real-time attenuation trend of the optical performance and the touch performance of the folding screen in the repeated folding process is difficult to capture, and the durability problem of the protective film cannot be predicted accurately. In addition, the existing scheme lacks comprehensive consideration of multiple physical field coupling effects when analyzing the correlation between complex stress distribution and polarized light reflectivity change of a folding area, and particularly the polarized light reflectivity of a screen display area is gradually enhanced in the dynamic process of the folding screen from flattening to limiting folding angles, which directly leads to display definition degradation. The reduced definition area is often accompanied by stress concentration, and the non-uniformity of stress distribution further weakens the touch sensitivity of the protective film. The interactive attenuation of the optical performance and the touch performance constitutes a core challenge of the performance evaluation of the folding screen protective film. The enhancement of the reflectivity of polarized light results from microstructural changes in the material of the folded region under repeated deformations, which in turn induce non-uniformity in the stress distribution. For example, after the folding screen is repeatedly folded 1000 times, the protective film near the folding axis region may generate local stress concentration, so that the touch signal response is delayed. Therefore, how to analyze the attenuation trend of the optical performance and the touch performance of the folding area by inputting the continuous variation signals of the polarized light reflectivity and the stress distribution and generate the early warning report of the attenuation of the performance of the folding screen protective film becomes a key problem for improving the durability and the user experience of the folding screen. Disclosure of Invention The invention provides a machine vision-based intelligent detection method for surface defects of a protective film, which mainly comprises the following steps: The method comprises the steps of constructing a reflectivity and stress distribution data set by collecting polarized light reflectivity and stress distribution data of a folding screen from a flat state to a limit folding angle, and carrying out feature extraction on the reflectivity and stress distribution data set to obtain an initial optical performance attenuation index, wherein the initial optical performance attenuation index comprises a reflectivity attenuation coefficient and a stress distribution non-average value; calculating a display definition value according to the polarized light reflectivity and stress distribution data, determining a definition decline trend by analyzing the corresponding relation between the initial attenuation index of the optical performance and the display definition value, extracting a touch response delay characteristic based on the definition decline trend, extracting a stress concentration region and a touch signal attenuation characteristic from the touch response delay characteristic, wherein the touch signal attenuation characteristic comprises a signal delay time and an attenuation amplitude, carrying out grouping analysis on the stress concentration region and the touch signal attenuation characteristic to obtain a joint attenuation trend of optical transmittance and touch signal attenuation, calculating the attenuation degree of the touch signal of a folding region according to the joint attenuation trend of the optical transmittance and the touch signal attenuation, combining the deformation degree of the stress concentration region to obtain a durability index, calculating a comprehensive performance attenuation index comprising optical transmittance attenuation, touch response delay and material fatigue degree, determining a folding angle change rate and a material accumulation angle change and refraction index according to the comprehensive performance attenuation index, predicting the relation between the folding angle change and the optical refraction