Visual detection system and detection method based on gold-plated cover plate production and processing

Abstract

The invention relates to the technical field of visual inspection, in particular to a visual inspection system and method based on gold-plated cover plate production and processing, a surface plasmon excitation state is constructed by exciting an illumination module to project coherent light beams, and a multi-source sensing module is matched to synchronously acquire a phase pattern, a multi-directional polarization image and a micropore photon sequence; an environment artifact is stripped based on polarization consistency by using an environment rejection module, space coordinates are dynamically aligned based on thermal parameters through a feature correction module, and the compactness of a microporous coating is quantified in combination with photon statistical characteristics, so that accurate rejection of false mirror images is realized on the physical optical level, and nonlinear measurement drift caused by temperature fluctuation is eliminated; meanwhile, the quality of the inner wall of the micropore is inversed through the photon sequence, a deep hole detection blind area is successfully overcome, and the detection precision and the yield control capability of the gold-plated cover plate in a complex production environment are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of visual inspection technology, and in particular to a visual inspection system and method based on the production and processing of gold-plated cover plates. Background Technology

[0002] As a key component in precision electronic packaging and signal transmission, the surface quality and plating integrity of the functional holes of gold-plated covers directly affect the conductivity and long-term reliability of the devices. In the production and processing stage, visual inspection technology has become the core means to ensure product yield. However, due to the extremely high reflectivity and mirror properties of gold-plated materials, existing visual inspection systems exhibit obvious limitations in complex industrial environments.

[0003] Currently, the most prominent technical challenge in the detection process is the serious contamination of image quality by environmental information. Since the surface of the cover plate is similar to a mirror, mechanical structures, lighting components, and even external ambient light in the production line can easily be reflected on the surface of the cover plate through the mirror effect, forming complex parasitic artifacts in the image. These artifacts are highly similar to real surface defects in terms of grayscale distribution and geometric shape, making it difficult for the system to accurately peel off the real defect features.

[0004] Interference from such environmental reflection artifacts often prevents conventional image processing logic from suppressing false alarms while ensuring detection rate. This technical defect not only causes automated inspection equipment to frequently trigger error warnings, forcing companies to invest a lot of manpower in manual review, but also limits the deep application of high-precision gold-plated parts in unmanned intelligent production lines, becoming a key bottleneck restricting the improvement of energy efficiency in precision manufacturing in the industry. Summary of the Invention

[0005] The main objective of this invention is to provide a visual inspection system based on the production and processing of gold-plated cover plates, which aims to alleviate the problem of high false alarm rate caused by parasitic artifacts in existing systems when performing visual inspection.

[0006] To achieve the above objectives, the present invention provides a visual inspection system based on the production and processing of gold-plated cover plates, the system comprising: An excitation illumination module is used to project a coherent light beam onto the surface of the gold-plated cover plate to be tested. The multi-source sensing module is used to acquire the phase pattern, multi-directional polarization characteristic image and photon sequence of the inner wall of the micropore of the gold-plated cover plate under test in the surface plasmon excitation state, and to acquire the thermal compensation reference parameters of the corresponding station. An environmental rejection module is used to obtain the polarization degree distribution of the surface of the gold-plated cover plate to be tested based on the multi-directional polarization feature image, and to identify the polarization consistency of the total reflection area based on the polarization degree distribution, so as to reject environmental artifacts generated by the environmental background and retain the diffuse reflection features that characterize the real surface defects. The feature correction module is used to obtain the deformation component of the gold-plated cover plate to be tested according to the thermal compensation reference parameters, and to dynamically align the spatial coordinates of the diffuse reflection feature and the phase pattern to obtain the corrected morphological feature map. The index association module is used to obtain the statistical characteristics of the photon sequence on the inner wall of the micropore, obtain the density index characterizing the coating quality based on the statistical characteristics, and associate the density index with the corresponding micropore location region in the morphology feature map. The detection processing module is used to extract the extreme values ​​of phase abrupt changes represented by the phase pattern in the corrected morphology feature map, determine whether the extreme values ​​of phase abrupt changes are located in the micropore location region, and if the determination result is yes, then the extreme values ​​of phase abrupt changes are corrected by attribute compensation according to the density index, and the micropore defect attribute is determined; if the determination result is no, then the surface defect attribute is determined according to the diffuse reflection feature, and the final detection conclusion is output.

[0007] Optionally, the illumination module includes a collimated laser source and a waveguide coupling unit. The waveguide coupling unit is used to adjust the incident light angle to match the excitation conditions of surface plasmon waves, so as to enhance the modulation sensitivity of the phase pattern to surface micro-damage.

[0008] Optionally, the multi-source sensing module includes a polarization-sensitive unit, a photon counting array, and an infrared sensing unit; the photon counting array is used to capture the photon sequence and acquire its impact frequency fluctuation data.

[0009] Optionally, the feature correction module is further configured to correct the air refractive index deviation in the imaging optical path according to the thermal compensation reference parameters, so as to compensate for the visual pixel shift caused by the temperature gradient.

[0010] To achieve the above objectives, the present invention also provides a visual inspection method based on the production and processing of gold-plated cover plates, the method comprising the following steps: The phase pattern, multi-directional polarization characteristic image, and photon sequence of the inner wall of the micropore of the gold-plated cover plate under test in the surface plasmon excitation state are obtained, and the thermal compensation reference parameters of the corresponding station are obtained. The polarization degree distribution of the gold-plated cover plate surface to be tested is obtained based on the multi-directional polarization feature image, and the polarization consistency of the total reflection area is identified based on the polarization degree distribution to eliminate environmental artifacts caused by the environmental background and retain diffuse reflection features that characterize real surface defects. The deformation component of the gold-plated cover plate to be tested is obtained according to the thermal compensation reference parameters, and the spatial coordinates of the diffuse reflection feature and the phase pattern are dynamically aligned to obtain the corrected morphological feature map. The statistical characteristics of the photon sequence on the inner wall of the micropore are obtained, and the density index characterizing the coating quality is obtained based on the statistical characteristics. The density index is then associated with the corresponding micropore location region in the morphology feature map. Extract the extreme values ​​of phase abrupt changes represented by the phase pattern from the corrected morphology feature image, determine whether the extreme values ​​of phase abrupt changes are located within the micropore location region, if the determination result is yes, then perform attribute compensation correction on the extreme values ​​of phase abrupt changes according to the compactness index, and determine the micropore defect attributes; if the determination result is no, then determine the surface defect attributes according to the diffuse reflection characteristics, and output the final detection conclusion.

[0011] Optionally, the process of removing environmental artifacts includes: A polarization feature template for environmental reflection sources is established. The multi-directional polarization feature image is compared with the polarization feature template, and artifact removal is achieved by suppressing pixels that conform to the template features.

[0012] Optionally, the process of determining the properties of micropore defects includes: The optical path phase error caused by the micropore depth is reverse-convolved according to the density index, and the presence of plating defects on the inner wall of the micropore is determined according to the corrected phase step height.

[0013] Optionally, the attribute compensation correction process also includes: A scattering convolution kernel characterizing the reflection pattern of the inner wall of the micropore is constructed based on the density index; The inverse compensation logic is performed on the phase abrupt extrema according to the scattering convolution kernel to eliminate phase noise generated by secondary reflections from the inner wall of the micropore.

[0014] Optionally, when determining the properties of surface defects, if the extreme value of the phase abrupt change is positively correlated with the gray-scale gradient of the diffuse reflection feature, it is determined to be a physical scratch.

[0015] Optionally, when determining the surface defect properties, if the extreme value of the phase abrupt change is negatively correlated with the gray-scale gradient of the diffuse reflection feature, it is determined to be surface oxidation.

[0016] The beneficial effects that this invention can achieve are as follows: This invention constructs a surface plasmon excitation state by projecting a coherent beam through an excitation illumination module, and simultaneously acquires physical information such as phase patterns, multi-directional polarization characteristic images, photon sequences of the micropore inner wall, and thermal compensation reference parameters through a multi-source sensing module. This effectively solves problems encountered during the production and processing of gold-plated covers, including environmental artifacts caused by high specular reflection, geometric deformation errors due to processing residual heat, and internal detection blind spots caused by the large aspect ratio of the micropores. It achieves high-precision automated inspection in complex industrial scenarios. Specifically, the system utilizes polarization degree distribution to identify the polarization consistency of the total reflection region, accurately removing false images generated by the environmental background from a physical optics perspective, ensuring that only diffuse reflection characteristics representing real surface defects are retained. This significantly reduces the false alarm rate of highly reflective workpieces. Simultaneously, by calculating deformation components using thermal compensation reference parameters and dynamically aligning spatial coordinates, the nonlinear drift caused by temperature fluctuations in precision measurements is eliminated, ensuring strict registration of multidimensional data at the micrometer level. More importantly, for micro-hole regions, the system innovatively utilizes the statistical characteristics of photon sequences to quantify coating density indicators and uses this to correct extreme values ​​of phase abrupt changes. This mechanism successfully overcomes phase noise caused by weak signals and multiple reflections inside deep holes, enabling the system to accurately distinguish between real defects and roughness interference on the inner wall of micro-holes. Combined with the diffuse reflection characteristic discrimination logic for non-micro-hole regions, the system ultimately outputs reliable detection conclusions containing comprehensive quality information for both the surface and inner wall. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0018] Figure 1 This is a structural block diagram of the system in Embodiment 1 of the present invention; Figure 2 This is a flowchart illustrating the method in Embodiment 2 of the present invention.

[0019] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0022] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral part; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0023] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the meaning of "and / or" throughout the text includes three parallel solutions; for example, "A and / or B" includes solution A, solution B, or a solution where both A and B are satisfied simultaneously. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0024] Example 1 As attached Figure 1 As shown, this embodiment provides a visual inspection system based on the production and processing of gold-plated cover plates. The system includes: An excitation illumination module is used to project a coherent light beam onto the surface of the gold-plated cover plate to be tested. The multi-source sensing module is used to acquire the phase pattern, multi-directional polarization characteristic image and photon sequence of the inner wall of the micropore of the gold-plated cover plate under test in the surface plasmon excitation state, and to acquire the thermal compensation reference parameters of the corresponding station. An environmental rejection module is used to obtain the polarization degree distribution of the surface of the gold-plated cover plate to be tested based on the multi-directional polarization feature image, and to identify the polarization consistency of the total reflection area based on the polarization degree distribution, so as to reject environmental artifacts generated by the environmental background and retain the diffuse reflection features that characterize the real surface defects. The feature correction module is used to obtain the deformation component of the gold-plated cover plate to be tested according to the thermal compensation reference parameters, and to dynamically align the spatial coordinates of the diffuse reflection feature and the phase pattern to obtain the corrected morphological feature map. The index association module is used to obtain the statistical characteristics of the photon sequence on the inner wall of the micropore, obtain the density index characterizing the coating quality based on the statistical characteristics, and associate the density index with the corresponding micropore location region in the morphology feature map. The detection processing module is used to extract the extreme values ​​of phase abrupt changes represented by the phase pattern in the corrected morphology feature map, determine whether the extreme values ​​of phase abrupt changes are located in the micropore location region, and if the determination result is yes, then the extreme values ​​of phase abrupt changes are corrected by attribute compensation according to the density index, and the micropore defect attribute is determined; if the determination result is no, then the surface defect attribute is determined according to the diffuse reflection feature, and the final detection conclusion is output.

[0025] It should be noted that in actual industrial production lines, gold-plated cover plates have always been a challenge for automated inspection due to their extremely high surface reflectivity and complex microporous structure. Conventional vision technology often cannot distinguish between benign processing textures and malignant defects on the metal surface, and it is also difficult to detect the plating state inside deep holes. This system constructs a precision inspection solution that can adapt to dynamic production environments by multidimensional excitation of physical light fields and data decoupling.

[0026] First, a coherent beam emitted from a collimated laser source is projected onto the surface of the gold-plated cover plate under test at a specific incident angle using a waveguide coupling unit. The incident angle is set to match the electronic oscillation frequency of the gold film surface, thereby exciting conductive surface plasmon waves at the gold film-air interface. This surface wave is extremely sensitive to subwavelength structural changes on the gold film surface; even micron-level stress cracks or nanometer-level thickness abrupt changes will cause a significant phase shift in the reflected light field, thus providing a physical basis for subsequent phase pattern acquisition. Simultaneously, the multi-source sensing module operates synchronously. Its integrated polarization-sensitive unit, single-photon counting array, and infrared sensing unit simultaneously acquire data from four dimensions: phase interference, polarization state, weak photon energy, and thermal radiation. This allows for the acquisition of the phase pattern, multi-directional polarization characteristic image, and photon sequence of the micropore inner wall of the gold-plated cover plate under test in the surface plasmon excitation state, and the real-time reading of the corresponding thermal compensation reference parameters.

[0027] Understandably, in the inspection of gold-plated covers, the biggest source of interference comes from the bright reflections formed on the mirror gold film by the external environment (such as robotic arms and light source supports). These reflections are extremely difficult to distinguish from oxidation defects in terms of grayscale. The environmental removal module is based on the physical optics principle that the specular reflection (total internal reflection) of a smooth surface will maintain the polarization state of the incident light, while the diffuse reflection caused by rough defects will change the polarization state. Thus, the artifacts are removed by calculating the polarization degree distribution.

[0028] During this process, the environment rejection module calculates the linear polarization distribution of the surface under test using the following expression: ; In the formula, This indicates the coordinate system of the gold-plated cover plate to be tested. The linear polarization degree value at that point is used to quantify the degree of polarization consistency of the reflected light at that point; This represents the coordinates of the polarization-sensitive element in the image acquired under the 0-degree polarization analysis direction. The gray intensity value at that location; This represents the coordinates of the polarization-sensitive unit in the image acquired under a 90-degree polarization analysis direction. The gray intensity value at that location; This represents the coordinates of the polarization-sensitive unit in the image acquired at a 45-degree polarization analysis direction. The gray intensity value at that location; This represents the coordinates of the polarization-sensitive unit in the image acquired at a 135-degree polarization analysis direction. The gray intensity value at that location; The total light intensity at that coordinate point is usually obtained by averaging or summing the four components mentioned above, and is derived from the synthesis of multi-directional polarization feature images.

[0029] It is understandable that the mirror artifacts produced by the environmental background undergo specular reflection, and their... The polarization value is usually extremely high (close to 1), while real surface scratches or oxidation points cause depolarization due to surface roughness. With a low value, the above formula can accurately construct a polarization confidence map through the difference operation of orthogonal polarization components. By setting a threshold to filter out the high polarization degree total reflection area, environmental reflections can be mathematically removed from the image, leaving only the diffuse reflection features that characterize the real surface defects, thereby greatly reducing the false alarm rate.

[0030] Subsequently, the feature correction module aligns the data in spatial dimensions. Since the gold plating process usually involves high-temperature treatment, the surface of the cover plate after it comes off the production line has residual heat, and the cooling time of different batches is different, which causes the cover plate to undergo micron-level thermal expansion and contraction deformation. This makes it impossible for the phase pattern (reflecting the micro morphology) and the diffuse reflection feature (reflecting the macro texture) to be precisely aligned in pixel coordinates. Based on the thermal compensation reference parameters, the feature correction module uses a dynamic affine transformation model to correct the coordinates.

[0031] In this step, the feature correction module uses the following expression to calculate the coordinate offset after thermal deformation: ; In the formula, This represents the target pixel coordinate vector in the corrected topographic feature map, which is used for subsequent multi-source data fusion. This represents the initial pixel coordinate vector in the original acquired image, which originates from the original imaging data of the multi-source sensing module; The linear thermal expansion coefficient of the gold-plated cover plate substrate is derived from a material property database or preset process parameters. The real-time temperature of the gold-plated cover plate under test is derived from the thermal field data collected by the infrared sensing unit. The reference ambient temperature used during the calibration of the vision system is derived from the system calibration records; This represents the optical path offset compensation term caused by changes in the air refractive index, which is related to the ambient air temperature at the testing station. Related.

[0032] It should be noted that in precision inspection, simple size scaling is not enough. The air density gradient around the high-temperature workpiece (thermal lensing effect) will cause nonlinear drift of visual pixels. This expression not only compensates for the deformation of the material itself, but also introduces an optical path deviation compensation term to ensure the dynamic alignment of diffuse reflection features and phase patterns in spatial coordinates. This processing logic enables the system to obtain corrected morphological feature maps with consistent geometric accuracy under different temperature conditions, avoiding misjudgment of defects caused by misalignment.

[0033] To address the detection blind zone inside micropores, where external light struggles to illuminate the bottom due to the large aspect ratio of the micropores, conventional imaging only yields a black spot. This system utilizes a (single) photon counting array to capture sparse photon sequences that escape after multiple reflections within the pore wall, and then uses statistical analysis to infer the pore wall quality.

[0034] Here, the index correlation module uses the following expression to calculate the density index, which characterizes the coating quality: ; In the formula, This index indicates the density of the coating on the inner wall of the micropores. The higher the value, the denser the coating and the smoother the surface. Conversely, a lower value indicates that there is a lack of coating or roughness. The statistical average of the number of photon sequences captured per unit time is derived from the integral data of a single-photon counting array; The statistical variance representing the time interval between photon arrivals is used to reflect the temporal coherence fluctuations of the reflected light. This represents the system's dark count noise constant, used to prevent the denominator from being zero and to correct for background noise interference; The effective attenuation coefficient for photon propagation within the aperture is derived from the convolution estimation of the microaperture geometry and the reflectivity of the ideal gold film. The design depth of the micropores is derived from the product's CAD design parameters.

[0035] It is understandable that if the coating on the hole wall is dense and uniform (of good quality), the reflection path of photons within the hole is relatively simple, and the escaped photon stream has a more concentrated temporal distribution (small variance, high mean), resulting in a higher calculated value. The quality is relatively high; if there are plating defects or severe pits (poor quality), photons will undergo diffuse scattering and absorption within the holes, resulting in a sharp decrease in the number of photons and a large dispersion in arrival time (large variance). Significantly reduced. Once calculated, this indicator is directly correlated with the corresponding micropore location region in the morphology feature map, serving as a hidden attribute of the micropore in subsequent determinations.

[0036] Finally, the system extracts the extreme values ​​of phase abrupt changes (i.e., points where the phase changes drastically) from the corrected morphology feature map. If this abrupt change point is located within the micropore region, the geometric depth of the micropore itself will also cause phase entanglement, and a simple phase abrupt change cannot be directly characterized as a defect. In this case, the system must call the aforementioned compactness index. We will construct a reverse compensation model based on physical optics.

[0037] In this specific scenario, the detection and processing module uses the following attribute compensation correction expression: ; In the formula, This represents the corrected phase step height, used to characterize the actual physical depth change of the micropore inner wall, i.e., the net value after removing scattering noise. This indicates the detected extreme value of the original phase abrupt change, which originates from direct measurement data of the phase pattern; and These represent the Fourier transform and inverse Fourier transform operators, respectively, used for inverse convolution processing in the frequency domain; This represents a scattering convolution kernel constructed based on a compactness index. This kernel function describes the blurring or trailing effect on the phase signal caused by secondary reflection of light within the aperture under the current coating quality.

[0038] The logic of the above expression lies in determining whether the phase jump is caused by the depth of the micro-aperture itself or by spurious noise signals generated by the roughness of the aperture wall. In the specific scenario of micro-aperture detection, the rougher the aperture wall (the lower the density index), the greater the phase noise of the reflected light field. The logic of this expression is actually an adaptive denoising and restoration process: the system uses the density index to generate a convolution kernel describing the degree of disorder in the light field within the aperture, and through inverse convolution operation, restores the true physical depth signal from the disordered measured phase. If the corrected... If an abnormal step height is still observed (e.g., a sudden change in depth exceeding the tolerance), it is determined that there is an exposed bottom or structural defect in the inner wall of the micropore.

[0039] Conversely, if the extreme value of the phase change is determined to be located in a non-microporous region (i.e., a planar region), the system then determines the surface defect attributes based on diffuse reflection characteristics. In this case, the system no longer needs complex convolution correction, but directly compares the correlation between the phase change and the diffuse reflection gray level: if the phase change is accompanied by an abnormal increase in diffuse reflection intensity (positive correlation), it indicates that there is strong scattering caused by physical protrusions or deep scratches at that location; if it is accompanied by a decrease in intensity (negative correlation), it is determined to be surface oxidation or contamination light absorption. Through this logical closed loop of regional and strategic division, the system achieves comprehensive and accurate detection of the gold-plated cover plate.

[0040] In this embodiment, the illumination module includes a collimated laser source and a waveguide coupling unit. The waveguide coupling unit is used to adjust the incident light angle to match the excitation conditions of surface plasmon waves, so as to enhance the modulation sensitivity of the phase pattern to surface micro-damage.

[0041] In this embodiment, the multi-source sensing module includes a polarization-sensitive unit, a photon counting array, and an infrared sensing unit; the photon counting array is used to capture the photon sequence and obtain its impact frequency fluctuation data.

[0042] In this embodiment, the feature correction module is also used to correct the air refractive index deviation in the imaging optical path according to the thermal compensation reference parameter, so as to compensate for the visual pixel shift caused by the temperature gradient.

[0043] Understandably, the excitation illumination module integrates a collimated laser source and a waveguide coupling unit. The collimated laser source outputs a monochromatic beam with high coherence and a specific polarization state, while the waveguide coupling unit (typically a Kretschmann prism structure or a grating coupler) handles precise beam shaping and angle control. The core function of the waveguide coupling unit is to dynamically adjust the angle of the incident light, ensuring that the projection of its wave vector onto the gold film surface satisfies the momentum matching condition with the surface plasmon wave vector at the gold-air interface. Only at this specific resonant angle can the incident light energy be coupled into the surface plasmon wave to the maximum extent, forming an evanescent wave propagating along the metal surface. This evanescent wave field is confined to a very small area on the gold film surface, thus exhibiting extremely high sensitivity to changes in the surface microstructure. When there are minute scratches or lattice damage on the gold-plated surface, the local resonant condition is disrupted, leading to a drastic abrupt change in phase in the reflected light field.

[0044] To quantify the modulation effect of this microscopic damage on the phase pattern, the following phase modulation sensitivity expression is used in this system: ; In the formula, It represents the modulation sensitivity of the phase pattern to surface micro-damage and is used to measure the phase shift caused by a unit change in defect depth. The differential change in the phase of the reflected light field is derived from the output of the phase retrieval algorithm in the multi-source sensing module. The effective refractive index representing surface plasmon waves is derived from the physical dispersion relationship between the dielectric constant of the gold film and the dielectric constant of the surrounding medium. The physical depth or equivalent geometric size of surface micro-damage is derived from calibration data of standard defective samples; This represents the effective propagation distance of surface plasmon waves on the gold film surface, which is derived from the coupling efficiency of the waveguide coupling unit and the gold film loss coefficient. This indicates the center wavelength emitted by the collimated laser source, which is derived from the laser's factory specifications. The normalized wave vector representing the surface plasmon wave is determined by the incident light angle and the refractive index of the waveguide coupling unit. The energy coupling efficiency of the waveguide coupling unit is derived from the calibration parameters of the optical system.

[0045] Based on the above expression, it can be understood that the phase sensitivity is proportional to the effective propagation distance and related to the reciprocal of the wavelength. By adjusting the incident angle through the waveguide coupling unit to strictly match the resonance condition, the rate of change of the effective refractive index can be maximized. This allows even nanoscale surface micro-damage to induce macroscopically measurable phase shifts in the phase pattern. This design is significantly superior to traditional intensity imaging, enabling the system to detect latent defects such as stress cracks in the early stages of coating.

[0046] For the multi-source sensing module, the system integrates a polarization-sensitive unit, a photon counting array, and an infrared sensing unit to achieve synchronous capture of multi-dimensional information. In particular, to address the challenge of detecting the inner wall of micro-holes, the system utilizes the extremely high temporal resolution and single-photon-level sensitivity of the photon counting array (usually a single-photon avalanche diode array, SPAD) to capture the weak photon sequence escaping from deep within the micro-hole. In the micro-hole detection scenario, light undergoes multiple reflections within the hole. If the inner wall coating is smooth and dense, the reflected photon stream has a stable statistical distribution in the time domain. If the inner wall is rough or poorly coated, the photon flight path becomes random and chaotic, causing drastic fluctuations in the photon frequency reaching the detector.

[0047] To quantify this fluctuation and infer the inner wall mass, the system uses the following expression to calculate the impact frequency fluctuation data: ; In the formula, This represents the impact frequency fluctuation data of the photon sequence, and this value is directly related to the surface roughness of the inner wall of the micropore; This represents the number of photons captured by the photon counting array in the i-th time slice, derived from the real-time counting output of the SPAD array; It represents the average number of photons per unit time slice within the entire sampling period, derived from the statistical average of the photon sequence; This represents the total number of time slices within the sampling period, determined by the detector's sampling frequency. The time coherence decay function representing photon propagation within a micropore is derived from a geometric model preset of the ratio of micropore depth to aperture size.

[0048] The processing logic of the above expression lies in characterizing the optical roughness of the micropore by statistically analyzing the dispersion of photon arrival time (the ratio of standard deviation to mean, i.e., a variant of the coefficient of variation). The temporal coherence attenuation function is introduced in the expression to eliminate the natural optical path attenuation effect caused by the micropore depth itself. In practical applications, if there are granular protrusions on the inner wall of the micropore or substrate exposure due to incomplete plating, diffuse scattering of photons will occur, resulting in… The numerical value is significantly increased; this design enables the system to accurately determine the coating quality using only a sparse photon flow in a deep hole environment with almost no light.

[0049] Finally, regarding the feature correction module, in addition to handling the thermal expansion and contraction of the material itself, the system also specifically considers the impact of thermo-optical effects on the imaging optical path. When the gold-plated cover plate has just finished processing or cleaning and drying, its surface temperature is significantly higher than the ambient temperature. This will form an air layer with a temperature gradient near the surface of the cover plate. Since the refractive index of air changes with temperature, light will be deflected when passing through this non-uniform air medium (similar to the mirage effect), causing the position of the image pixels captured by the camera to shift. If no correction is made, this shift will be mistakenly identified as the cover plate being out of tolerance.

[0050] Therefore, the feature correction module corrects the air refractive index deviation in the imaging optical path based on the thermal compensation reference parameters. The compensation expression used is as follows: ; In the formula, This represents the visual pixel offset caused by the temperature gradient, used for inverse compensation in the image coordinate system; The magnification factor, representing the optical imaging system, is derived from the lens calibration parameters; The distance from the surface of the cover plate to the imaging lens is derived from the mechanical mounting dimensions of the system. It represents the functional relationship between the refractive index of air and temperature, following the simplified form of the Edlen equation; The distribution of air temperature gradient along the optical axis Z is derived from the thermal field model established by the infrared sensing unit. Z represents the distance variable along the optical axis, and the integration interval is from the cover plate surface to the lens optical center.

[0051] By integrating the continuous refraction path of light in non-uniform hot air, the lateral displacement of the final imaging point on the image plane is calculated. In high-precision visual inspection (micrometer level), a temperature difference of tens of degrees can cause changes in the air's refractive index, resulting in errors of several pixels. By acquiring thermal compensation reference parameters (surface temperature and ambient temperature) in real time, the system can construct a temperature gradient model along the optical path. Then the pixel offset is calculated. This is then subtracted from the measurement results; this adaptive design ensures that the system maintains consistent measurement accuracy during the cover plate cooling process (dynamic temperature changes), avoiding false size alarms caused by the hot air lens effect, and greatly improving the reliability of the detection data.

[0052] Example 2: As attached Figure 2 As shown, this embodiment provides a visual inspection method based on the production and processing of gold-plated cover plates. The method includes the following steps: The phase pattern, multi-directional polarization characteristic image, and photon sequence of the inner wall of the micropore of the gold-plated cover plate under test in the surface plasmon excitation state are obtained, and the thermal compensation reference parameters of the corresponding station are obtained. The polarization degree distribution of the gold-plated cover plate surface to be tested is obtained based on the multi-directional polarization feature image, and the polarization consistency of the total reflection area is identified based on the polarization degree distribution to eliminate environmental artifacts caused by the environmental background and retain diffuse reflection features that characterize real surface defects. The deformation component of the gold-plated cover plate to be tested is obtained according to the thermal compensation reference parameters, and the spatial coordinates of the diffuse reflection feature and the phase pattern are dynamically aligned to obtain the corrected morphological feature map. The statistical characteristics of the photon sequence on the inner wall of the micropore are obtained, and the density index characterizing the coating quality is obtained based on the statistical characteristics. The density index is then associated with the corresponding micropore location region in the morphology feature map. Extract the extreme values ​​of phase abrupt changes represented by the phase pattern from the corrected morphology feature image, determine whether the extreme values ​​of phase abrupt changes are located within the micropore location region, if the determination result is yes, then perform attribute compensation correction on the extreme values ​​of phase abrupt changes according to the compactness index, and determine the micropore defect attributes; if the determination result is no, then determine the surface defect attributes according to the diffuse reflection characteristics, and output the final detection conclusion.

[0053] In this embodiment, the process of removing environmental artifacts includes: A polarization feature template for environmental reflection sources is established. The multi-directional polarization feature image is compared with the polarization feature template, and artifact removal is achieved by suppressing pixels that conform to the template features.

[0054] In this embodiment, the process of determining the properties of micropore defects includes: The optical path phase error caused by the micropore depth is reverse-convolved according to the density index, and the presence of plating defects on the inner wall of the micropore is determined according to the corrected phase step height.

[0055] In this embodiment, the attribute compensation and correction process further includes: A scattering convolution kernel characterizing the reflection pattern of the inner wall of the micropore is constructed based on the density index; The inverse compensation logic is performed on the phase abrupt extrema according to the scattering convolution kernel to eliminate phase noise generated by secondary reflections from the inner wall of the micropore.

[0056] In this embodiment, when determining the surface defect properties, if the extreme value of the phase abrupt change is positively correlated with the gray-scale gradient of the diffuse reflection feature, it is determined to be a physical scratch.

[0057] In this embodiment, when determining the surface defect properties, if the extreme value of the phase abrupt change is negatively correlated with the gray-scale gradient of the diffuse reflection feature, it is determined to be surface oxidation.

[0058] It should be noted that, based on the hardware architecture of the aforementioned system, this embodiment further elaborates on the specific execution logic and data processing flow of a visual inspection method for the production and processing of gold-plated cover plates. Specifically, when executing this method, a multi-source data synchronous acquisition program is first initiated, and the illumination module is excited to project a coherent beam, causing the free electrons on the surface of the gold-plated cover plate to undergo collective oscillation and enter a surface plasmon excitation state. In this state, the multi-source sensing module synchronously captures the phase pattern, multi-directional polarization feature image, and sparse photon sequence of the micropore inner wall. At the same time, in order to cope with the influence of residual heat from processing, the system reads the thermal compensation reference parameters (including surface temperature gradient and ambient temperature) of the corresponding station in real time, providing a physical reference for subsequent spatial coordinate alignment.

[0059] This embodiment introduces a recognition logic for the polarization consistency of the total internal reflection region. The reflection of the environmental background (such as the light source support or robotic arm) on the gold film surface is a specular reflection, and its polarization state is highly consistent with the incident light. However, real surface defects will cause depolarization effects. In order to further improve the removal accuracy, this embodiment specifically establishes a polarization feature template of the environmental reflection source. This process compares the real-time acquired multi-directional polarization feature image with the preset template in a high dimension, and achieves accurate removal of artifacts by suppressing pixels that match the template features.

[0060] In this process, the removal of environmental artifacts uses the following polarization template matching and suppression expression: ; In the formula, This represents the grayscale value of a clean image that retains diffuse reflection features after artifact removal, which serves as input data for subsequent defect identification. This represents the original polarization-synthesized intensity image, which contains real defect information and environmental artifact noise; This represents a nonlinear adjustment factor used to control the suppression of edge smoothness, derived from empirical parameters during the system debugging phase; The similarity score between the polarization state of the current pixel region and the polarization feature template of the environmental reflection source is derived from the dot product operation of the local polarization vectors; The similarity threshold for determining whether an image is an environmental artifact is derived from calibration data of the production line environment background.

[0061] The above expression constructs a soft threshold mask of the Sigmoid form, which is applied to the polarization feature at a certain point. Highly compatible with the environment template (i.e., similarity higher than) When the weights within the parentheses approach 0, the original image... Artifacts are strongly suppressed; conversely, when the point exhibits significant depolarization characteristics (real defects), the weight approaches 1, and the diffuse reflection features are fully preserved. Compared with hard threshold clipping, this nonlinear suppression logic can effectively avoid jagged cutting marks at the edges of artifacts, ensuring the continuity of diffuse reflection features.

[0062] Subsequently, the deformation components of the gold-plated cover plate under test are obtained using thermal compensation reference parameters, and the spatial coordinates of the diffuse reflection features and phase pattern are dynamically aligned. This step ensures that features of different physical dimensions correspond precisely at the same pixel coordinates, laying the foundation for subsequent correlation analysis.

[0063] Entering the crucial stage of micro-hole detection, the statistical characteristics of photon sequences are used to quantify the hole wall quality. The roughness of the inner wall of the micro-hole directly determines the complexity of the photon reflection path within the hole. In this embodiment, a density index characterizing the coating quality is obtained through statistical analysis and correlated with the corresponding micro-hole location region in the morphology feature map. When the system extracts the phase abrupt change extreme value characterized by the phase pattern and confirms that the extreme value is located within the micro-hole region, it is necessary to solve the problems of phase entanglement caused by hole depth and phase noise aliasing caused by hole wall roughness. To this end, this embodiment introduces anti-convolution logic to construct a scattering convolution kernel based on the density index and perform attribute compensation correction on the phase data.

[0064] Here, the attribute compensation correction process adopts the following phase inversion compensation expression based on the scattering convolution kernel: ; In the formula, This indicates the corrected phase step height, used to accurately reflect whether there is any exposed substrate or structural defects on the inner wall of the micropore; This represents the distribution of extreme values ​​of phase abrupt changes obtained from the original measurement, including the true depth signal and secondary reflection noise; This represents the regularization coefficient for reverse compensation, used to control the intensity of noise removal and prevent excessive smoothing from causing the loss of true depth information. This represents the two-dimensional convolution operator. Indicated based on density index The constructed scattering convolution kernel function characterizes the point diffusion effect of the light field under a specific roughness.

[0065] Among them, scattering convolution kernel The construction logic is as follows: ; In the formula, The dispersion scale parameter of the scattering nucleus is negatively correlated with the compactness index η (the worse the compactness, the stronger the scattering dispersion). , This represents the local coordinate variables of the convolution kernel in the spatial domain.

[0066] In micropore scenarios, the rougher the inner wall (lower density index), the more non-mirror reflections occur, leading to spatial haloing and tailing of the phase signal, manifested as high-frequency noise superposition. This embodiment simulates this physical haloing process using a Gaussian scattering convolution kernel, which transforms the original phase... The operation with the convolution kernel actually estimates the phase noise component caused by roughness, and then subtracts it from the original measurement (i.e., reverse compensation). The corrected phase step height can remove the interference of surface quality and purely reflect the geometric depth of the micropore, so that the system can accurately determine whether there is substrate exposure due to missed plating (i.e., substrate exposure defect).

[0067] Finally, when the phase abrupt change extreme value is determined to be located in the non-microporous region (i.e. planar region), the system enters the classification logic of surface defect attributes. At this time, the gray-scale gradient of diffuse reflection characteristics becomes the key criterion for distinguishing physical scratches from surface oxidation. Since physical scratches usually appear as grooves or protrusions on the metal surface, their edges will produce strong edge scattering, resulting in an increase in diffuse reflection light intensity. On the other hand, the surface oxide layer usually appears as black light-absorbing spots, resulting in a decrease in diffuse reflection light intensity.

[0068] To quantify this determination process, this embodiment employs the following surface defect attribute association discriminant expression: ; In the formula, The discrimination coefficient represents the properties of surface defects, and its positive or negative sign directly determines the defect category; The sign function is used to extract the directionality of phase abrupt extrema. The local gradient vector representing the extreme value of a phase abrupt change; This represents the covariance operation, used to measure the correlation between the changing trends of the phase gradient and the diffuse gradient; This represents the local gradient distribution of the phase pattern in the defect region; This represents the gray-level gradient distribution of the retained diffuse reflection feature image in the same region; This represents variance operations, used for normalization.

[0069] The above expression is essentially a locally weighted Pearson correlation coefficient calculation, when When the correlation is >0 (positive correlation), it means that the intensity of diffuse reflection light changes drastically (usually becomes brighter) where the phase changes abruptly. This is consistent with the optical characteristics of physical scratches, that is, the destruction of the shape is accompanied by strong scattering.

[0070] when When the correlation is <0 (negative correlation), it means that at the location of a phase abrupt change (caused by changes in oxide layer thickness or refractive index), the diffuse reflection intensity decreases (darkens), which is consistent with the optical characteristics of surface oxidation or oil stains, i.e., material absorption leads to a decrease in reflectivity. This judgment logic based on multi-dimensional feature correlation greatly improves the accuracy of identifying complex defects on gold-plated surfaces compared to a single grayscale threshold classification, effectively preventing benign machining marks from being misjudged as oxidation, or shallow oxidation from being misjudged as deep scratches.

[0071] In summary, in the aforementioned embodiments, a specially modulated laser beam is used to irradiate the surface of the cover plate. This beam is not for illumination, but to excite a special physical wave (surface plasmon wave) on the gold film surface. This wave is extremely sensitive to very small, invisible lattice damage or microcracks on the surface. Once the surface is damaged, the phase of the reflected light changes. Simultaneously, the system collects three sets of core data: first, phase data reflecting the microstructure; second, multi-angle polarization data reflecting the surface reflection characteristics; and third, weak photon signals reflected from the tiny holes in the cover plate. Furthermore, considering the residual heat of the cover plate after processing on the production line, the system also monitors the temperature in real time for subsequent dimensional error correction.

[0072] After data acquisition, the system does not directly identify defects but first performs two steps of data cleaning. The first step is to remove environmental reflections. Because the gold-plated surface acts like a mirror, it easily reflects the surrounding equipment, which appears as defects to ordinary cameras. The system uses polarization data for analysis. Environmental reflections usually maintain the polarization direction of light (total internal reflection), while real scratches or oxidation points scatter the light (diffuse reflection). By calculating the consistency of polarization, the system can treat all reflective areas as background and remove them, retaining only the real defect features that scatter the light. The second step is to correct for thermal deformation. Based on the acquired temperature data, the system calculates the dimensional changes of the cover plate due to thermal expansion and contraction, as well as the refraction deviation of light by hot air. The acquired image coordinates are stretched or compressed to ensure that all image feature points are precisely aligned with the standard coordinate system.

[0073] After data cleaning is completed, the system enters the core defect determination stage, which is divided into two logical paths: micropores and surfaces.

[0074] For micro-hole regions, conventional cameras often struggle to clearly capture the inner walls of the holes. The system analyzes previously acquired photon sequences. If the inner wall coating is dense and smooth, the statistical regularity of photon reflection is stable. If the inner wall is rough or has incomplete coating, photons will scatter randomly. Based on this, the system calculates a density index. When the system detects abrupt changes in the phase data at the micro-hole location, it uses this density index to construct a mathematical model (scattering convolution kernel) to filter out spurious phase noise caused by the roughness of the hole wall, restoring the true physical depth of the inner wall. If the corrected depth still shows an anomaly, it is determined that there are defects such as incomplete coating or exposed substrate on the inner wall of the micro-hole.

[0075] For surface areas, the judgment logic is relatively intuitive. The system compares the relationship between the phase change point and the diffuse reflection characteristics. If the diffuse reflection intensity also increases abnormally (becomes brighter) at the location where the phase changes, it indicates that there is a physical protrusion or deep scratch that causes strong scattering, and the system judges it as a physical scratch. Conversely, if the diffuse reflection intensity decreases (becomes darker) at the location where the phase changes, it indicates that the material has changed and absorbed light, and the system judges it as surface oxidation.

[0076] In other words, this invention uses physical means to separate real defects from false reflections, makes up for errors caused by thermal deformation, uses photon signals to compensate for the shortcomings of not being able to see deep holes, and finally achieves comprehensive detection of gold-plated cover plates from surface scratches to internal plating defects through cross-verification of multi-dimensional data.

[0077] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0078] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0079] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk) and includes several instructions to cause a multimedia terminal device (which may be a mobile phone, computer, television receiver, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0080] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A visual inspection system based on the production and processing of gold-plated cover plates, characterized in that, The system includes: An excitation illumination module is used to project a coherent light beam onto the surface of the gold-plated cover plate to be tested. The multi-source sensing module is used to acquire the phase pattern, multi-directional polarization characteristic image and photon sequence of the inner wall of the micropore of the gold-plated cover plate under test in the surface plasmon excitation state, and to acquire the thermal compensation reference parameters of the corresponding station. An environmental rejection module is used to obtain the polarization degree distribution of the surface of the gold-plated cover plate to be tested based on the multi-directional polarization feature image, and to identify the polarization consistency of the total reflection area based on the polarization degree distribution, so as to reject environmental artifacts generated by the environmental background and retain the diffuse reflection features that characterize the real surface defects. The feature correction module is used to obtain the deformation component of the gold-plated cover plate to be tested according to the thermal compensation reference parameters, and to dynamically align the spatial coordinates of the diffuse reflection feature and the phase pattern to obtain the corrected morphological feature map. The index association module is used to obtain the statistical characteristics of the photon sequence on the inner wall of the micropore, obtain the density index characterizing the coating quality based on the statistical characteristics, and associate the density index with the corresponding micropore location region in the morphology feature map. The detection processing module is used to extract the extreme values ​​of phase abrupt changes represented by the phase pattern in the corrected morphology feature map, determine whether the extreme values ​​of phase abrupt changes are located in the micropore location region, and if the determination result is yes, then the extreme values ​​of phase abrupt changes are corrected by attribute compensation according to the density index, and the micropore defect attribute is determined; if the determination result is no, then the surface defect attribute is determined according to the diffuse reflection feature, and the final detection conclusion is output.

2. The visual inspection system based on the production and processing of gold-plated cover plates as described in claim 1, characterized in that, The illumination module includes a collimated laser source and a waveguide coupling unit. The waveguide coupling unit is used to adjust the incident light angle to match the excitation conditions of surface plasmon waves, so as to enhance the modulation sensitivity of the phase pattern to surface micro-damage.

3. The visual inspection system based on the production and processing of gold-plated cover plates as described in claim 1, characterized in that, The multi-source sensing module includes a polarization-sensitive unit, a photon counting array, and an infrared sensing unit; the photon counting array is used to capture the photon sequence and obtain its impact frequency fluctuation data.

4. The visual inspection system based on the production and processing of gold-plated cover plates as described in claim 1, characterized in that, The feature correction module is also used to correct the air refractive index deviation in the imaging optical path according to the thermal compensation reference parameters, so as to compensate for the visual pixel shift caused by the temperature gradient.

5. A visual inspection method based on the production and processing of gold-plated cover plates, characterized in that, The method includes the following steps: The phase pattern, multi-directional polarization characteristic image, and photon sequence of the inner wall of the micropore of the gold-plated cover plate under test in the surface plasmon excitation state are obtained, and the thermal compensation reference parameters of the corresponding station are obtained. The polarization degree distribution of the gold-plated cover plate surface to be tested is obtained based on the multi-directional polarization feature image, and the polarization consistency of the total reflection area is identified based on the polarization degree distribution to eliminate environmental artifacts caused by the environmental background and retain diffuse reflection features that characterize real surface defects. The deformation component of the gold-plated cover plate to be tested is obtained according to the thermal compensation reference parameters, and the spatial coordinates of the diffuse reflection feature and the phase pattern are dynamically aligned to obtain the corrected morphological feature map. The statistical characteristics of the photon sequence on the inner wall of the micropore are obtained, and the density index characterizing the coating quality is obtained based on the statistical characteristics. The density index is then associated with the corresponding micropore location region in the morphology feature map. Extract the extreme values ​​of phase abrupt changes represented by the phase pattern from the corrected morphology feature image, determine whether the extreme values ​​of phase abrupt changes are located within the micropore location region, if the determination result is yes, then perform attribute compensation correction on the extreme values ​​of phase abrupt changes according to the compactness index, and determine the micropore defect attributes; if the determination result is no, then determine the surface defect attributes according to the diffuse reflection characteristics, and output the final detection conclusion.

6. The visual inspection method based on the production and processing of gold-plated cover plates as described in claim 5, characterized in that, The process of removing environmental artifacts includes: A polarization feature template for environmental reflection sources is established. The multi-directional polarization feature image is compared with the polarization feature template, and artifact removal is achieved by suppressing pixels that conform to the template features.

7. The visual inspection method based on the production and processing of gold-plated cover plates as described in claim 5, characterized in that, The process of determining the properties of micropore defects includes: The optical path phase error caused by the micropore depth is reverse-convolved according to the density index, and the presence of plating defects on the inner wall of the micropore is determined according to the corrected phase step height.

8. The visual inspection method based on the production and processing of gold-plated cover plates as described in claim 7, characterized in that, The attribute compensation and correction process also includes: A scattering convolution kernel characterizing the reflection pattern of the inner wall of the micropore is constructed based on the density index; The inverse compensation logic is performed on the phase abrupt extrema based on the scattering convolution kernel to eliminate phase noise generated by secondary reflections from the inner wall of the micropore.

9. The visual inspection method based on the production and processing of gold-plated cover plates as described in claim 8, characterized in that, When determining the properties of surface defects, if the extreme value of the phase abrupt change is positively correlated with the gray-scale gradient of the diffuse reflection feature, it is determined to be a physical scratch.

10. A visual inspection method based on the production and processing of gold-plated cover plates as described in claim 8 or 9, characterized in that, When determining the properties of surface defects, if the extreme value of the phase abrupt change is negatively correlated with the gray-scale gradient of the diffuse reflection feature, it is determined to be surface oxidation.

Images