Sand blasting precision detection method for wearable watch bottom shell

Abstract

The invention discloses a sand blasting precision detection method for a wearable watch bottom shell, and relates to the field of sand blasting precision detection for bottom shells, and the method comprises the following steps: carrying out the judgment of a window mutual information value, a pixel point gradient vector, a pixel convex state, a pixel concave state and a sand blasting texture of a visual image, and obtaining a bottom shell sand blasting visual evaluation value; analyzing the visual evaluation value of the bottom shell sand blasting to obtain a visual evaluation set H of the bottom shell sand blasting, performing depth analysis on the bottom shell sand blasting according to the surface physical characteristics, the particle distribution characteristics, the mechanical state, the material performance parameters and the surface optical performance of the bottom shell sand blasting to obtain a deep residual corrosion value of the bottom shell sand blasting, and analyzing the deep residual corrosion value of the bottom shell sand blasting to obtain a deep residual corrosion value of the bottom shell sand blasting. And a bottom shell sand blasting depth evaluation set W is obtained, union processing is carried out on the set H and the set W, warehousing operation and label display corresponding to all levels of comprehensive evaluation signals are obtained through analysis, sand blasting precision judgment can be carried out on the watch bottom shell from multiple dimensions, the sand blasting process can be adjusted in time, and the production cost of an enterprise is reduced.

Description

Technical Field

[0001] The invention relates to the field of sandblasting accuracy detection for bottom shells, and in particular to a sandblasting accuracy detection method for bottom shells of wearable watches. Background Art

[0002] As consumers' requirements for the refinement and quality of watch appearance continue to increase, the bottom case of a watch is an important part of the watch, so its surface treatment quality is becoming more and more critical. The sandblasting process can give the bottom case of the watch a high-quality appearance and a unique texture on the bottom of the watch, so the precision detection of sandblasting is very important. Traditional detection methods mainly rely on visual judgment and conventional measuring tools, so the sandblasting precision detection method for the bottom case of wearable watches came into being.

[0003] When the traditional sandblasting accuracy detection method for wearable watch bottom cases is used, it is impossible to judge the sandblasting accuracy of the watch bottom cases in multiple dimensions, cannot classify and store watch bottom cases with different sandblasting accuracy qualities, and cannot promptly remind process parameters with unqualified sandblasting accuracy. As a result, the sandblasting quality of the watch bottom cases produced by the company is uneven, affecting the company's reputation, and the sandblasting process cannot be adjusted in time, increasing the company's production costs, resulting in a large amount of rework and scrapping due to quality problems.

[0004] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0005] In order to solve the technical problems raised by the above background technology, the present invention is proposed. An embodiment of the present invention provides a sandblasting accuracy detection method for a bottom shell of a wearable watch.

[0006] The purpose of the present invention can be achieved by the following technical solution: A sandblasting accuracy detection method for a bottom shell of a wearable watch comprises the following steps:

[0007] Step 1: Use a camera to collect an image of the sandblasted bottom shell, and determine the window mutual information value, pixel gradient vector, pixel convex state, pixel concave state, and sandblasted texture of the visual image to obtain the visual evaluation value of the sandblasted bottom shell;

[0008] Step 2: Analyze and determine the bottom shell sandblasting visual evaluation value and the reference interval to obtain the bottom shell sandblasting visual evaluation set H;

[0009] Step 3: Conduct in-depth analysis of the bottom shell sandblasting based on the surface physical characteristics, particle distribution characteristics, mechanical state, material performance parameters and surface optical properties of the bottom shell sandblasting to obtain the deep residual corrosion value of the bottom shell sandblasting;

[0010] Step 4: Analyze and determine the bottom shell sandblasting depth residual corrosion value and the reference threshold value to obtain the bottom shell sandblasting depth evaluation set W;

[0011] Step 5: By performing a union process on the sets H and W, and analyzing them, the storage operations and label displays corresponding to the comprehensive evaluation signals at each level are obtained.

[0012] Further, the bottom shell sandblasting visual evaluation analysis steps are as follows:

[0013] Step 106: The normalized image is divided into several regions, with the pixel in each region as the center. The gray value of the center pixel is marked as zh, and the gray value of the area is marked as L. q , where q = 1, 2, ..., 8, represents the sequence number of the field pixel, according to the formula Get the grayscale value of each neighborhood pixel relative to the central pixel X q , where a1, a2 and a3 are natural constants, according to the formula The pair value DZZ of the central pixel is obtained, where e is a natural constant. The pair values ​​of all central pixels in the image are counted, and divided into G intervals according to the pair value distribution of the central pixel of the image. The pair value histogram of the central pixel is established according to the formula The symmetric distance DJ between the A and B intervals is obtained, where g is the interval number, which is a positive integer and has a maximum value of G. The symmetric distances between each two regions in the image are counted and summed to obtain the uniform deviation of the sandblasting texture of the bottom shell;

[0014] Step 107: Calculate the bottom shell sandblasting information consistency value, the bottom shell sandblasting gradient value, the bottom shell sandblasting convex value, the bottom shell sandblasting concave value and the bottom shell sandblasting texture deviation value to obtain the bottom shell sandblasting visual evaluation value Spz.

[0015] Furthermore, the analysis steps of the bottom shell sandblasting convex stroke value and the bottom shell sandblasting concave stroke value are as follows:

[0016] Step 104: for the normalized image obtained, a grayscale threshold of 1 is set. When a pixel in the image is greater than the grayscale threshold of 1, the pixel is marked as a convex region candidate point 1. The reflected light grayscale dispersion value of each pixel is calculated. The reflected light grayscale dispersion value is the standard deviation of the reflected light grayscale in the area centered on the pixel in the convex region candidate point 1. When the reflected light grayscale dispersion value is greater than the set threshold PT1, the loudness point is marked as a convex region candidate point 2. The convex region candidate points 1 and 2 are merged to obtain a convex region. The difference between the highest grayscale value of the convex region and the average grayscale value of the non-convex region is obtained, and a weighted calculation is performed with the number of convex regions, and the result is multiplied by the corresponding weight factor coefficient to obtain the bottom shell sandblasting convexity value.

[0017] Step 105: For the normalized image obtained, set grayscale threshold 2. When a pixel in the image is less than grayscale threshold 2, mark the pixel as candidate point 1 for the concave area. Calculate the grayscale discreteness value of the reflected light of each pixel. When the grayscale discreteness value of the reflected light is greater than the set threshold PT2, mark the pixel as candidate point 2 for the concave area. Merge candidate points 1 and 2 for the concave area to obtain the concave area. Obtain the difference between the average grayscale value of the non-concave area and the minimum grayscale value of the concave area. Perform weighted calculation on the difference with the number of concave areas and multiply by the corresponding weight factor coefficient to obtain the bottom shell sandblasting concave value.

[0018] Furthermore, the steps for analyzing the consistent value of the bottom shell sandblasting information and the gradient noise value of the bottom shell sandblasting are as follows:

[0019] Step 101: The light source is irradiated onto the surface of the bottom case of the watch, and image information is obtained by using a camera. The image is normalized and gray-scaled. The normalized image is gradually expanded from a 1×1 pixel window to an n×n pixel window in sequence, and the gray value of each point in each window is calculated as the reflected light intensity of each point in the window;

[0020] Step 102: Divide the reflected light intensity range of two different window scales into m intervals, calculate the probability of the emitted light intensity of each window scale in each interval, marked as edge probability, and the probability that the intensity of the two window sizes falls into their respective specific intervals at the same time, marked as joint probability, for each pair of interval combinations that appear, there are m×m, divide the joint probability by the product of the two edge probabilities, marked as division value one, take the logarithm of the division value one and multiply it by the joint probability to obtain the product value, add all the product values ​​to obtain the mutual information value of the two different window scales, repeat the above steps to obtain all the mutual information values ​​of the two different window scales, and add them to obtain the consistent value of the bottom shell sandblasting information;

[0021] Step 103: For the normalized image obtained, an image gradient calculation algorithm is used to obtain the gradient vector of each pixel in the image, and the standard deviation of the gradient vector direction in the image and the extreme value of the gradient vector modulus in the image are summed to obtain the bottom shell sandblasting gradient noise value.

[0022] Furthermore, the storage operation and label display analysis steps corresponding to the comprehensive evaluation signals at each level are as follows:

[0023] Step 501: Perform a union process on the bottom shell sandblasting visual evaluation set H and the bottom shell sandblasting depth evaluation set W. When H∪W={h1, w1}, a first-level comprehensive evaluation signal for bottom shell sandblasting accuracy determination is generated; when H∪W={h1, w2} or {h1, w3} or {h2, w1} or {h2, w2} or {h3, w1}, a second-level comprehensive evaluation signal for bottom shell sandblasting accuracy determination is generated; when H∪W={h2, w3} or {h3, w2}, a third-level comprehensive evaluation signal for bottom shell sandblasting accuracy determination is generated; when H∪W={h3, w3}, a fourth-level comprehensive evaluation signal for bottom shell sandblasting accuracy determination is generated;

[0024] Step 502: If the sandblasting of the bottom shell generates a first-level comprehensive evaluation signal for determining the accuracy of the bottom shell sandblasting, the sandblasting of the bottom shell will be reclassified into set Y1, put into storage, and the label will display label one; if the sandblasting of the bottom shell generates a second-level comprehensive evaluation signal for determining the accuracy of the bottom shell sandblasting, the sandblasting of the bottom shell will be reclassified into set Y2, put into storage, and the label will display label one; if the sandblasting of the bottom shell generates a third-level comprehensive evaluation signal for determining the accuracy of the bottom shell sandblasting, the sandblasting of the bottom shell will be reclassified into set Y3, and storage into storage is not allowed, and the label will display label one; if the sandblasting of the bottom shell generates a fourth-level comprehensive evaluation signal for determining the accuracy of the bottom shell sandblasting, the sandblasting of the bottom shell will be reclassified into set Y4, and storage into storage is not allowed, and the label will display label one.

[0025] Further, the bottom shell sandblasting depth evaluation set W analysis steps are as follows:

[0026] Step 402: when the deep residual corrosion value of the bottom shell sandblasting is greater than or equal to the reference threshold value TG1, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 3 depth assessment judgment signal; when the deep residual corrosion value of the bottom shell sandblasting is less than the reference threshold value TG1 and greater than the reference threshold value TG2, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 2 depth assessment judgment signal; when the deep residual corrosion value of the bottom shell sandblasting is less than or equal to the reference threshold value TG2, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 1 depth assessment judgment signal;

[0027] Step 403: A set W is established according to the bottom shell sandblasting depth assessment and determination signal, and the bottom shell sandblasting first, second and third level depth assessment and determination signals are respectively determined as elements w1, w2 and w3.

[0028] Further, the bottom shell sandblasting depth evaluation set W analysis steps are as follows:

[0029] Step 302: The surface material characteristic value of the sandblasting of the bottom shell, the particle distribution value of the sandblasting of the bottom shell, the force bias residual value of the sandblasting of the bottom shell, the material corrosion capacity value of the sandblasting of the bottom shell, and the surface light variation value of the sandblasting of the bottom shell are normalized, and the force bias residual value of the sandblasting of the bottom shell is used as the radius of the circle on the upper surface of the frustum, the material corrosion capacity value of the sandblasting of the bottom shell is used as the radius of the circle on the lower surface of the frustum, and the surface light variation value of the sandblasting of the bottom shell is used as the height of the frustum to construct a frustum, and the center point of the frustum is used as the center of the sphere, and a sphere is constructed inside the frustum, and the sum of the surface material characteristic value of the sandblasting of the bottom shell and the particle distribution value of the sandblasting of the bottom shell is used as the radius of the sphere, and the non-overlapping volume formed by the frustum and the sphere is identified, and marked as the deep residual corrosion value of the sandblasting of the bottom shell.

[0030] Furthermore, the analysis steps of the material corrosion capacity value and the surface light variation value of the bottom shell sandblasting are as follows:

[0031] The material performance parameters of the bottom shell sandblasting refer to the weighted calculation of the sandblasting chemical activity, the galvanic corrosion sensitivity of the bottom shell, and the biocompatibility, and multiply them by the corresponding influencing factor coefficients to obtain the material corrosion capacity value of the bottom shell sandblasting. The galvanic corrosion sensitivity of the bottom shell refers to the polarization curve measurement of the bottom shell after sandblasting to obtain the value of the corrosion current density. Biocompatibility refers to the number of times itching, erythema, rash, swelling, redness and fever occur after wearing the watch for a certain period of time. The surface optical performance of the bottom shell sandblasting refers to the sum of the surface light scattering characteristic value, the surface color stability value, and the optical reflectivity anisotropy value to obtain the surface light variation value of the bottom shell sandblasting. The surface light scattering characteristic value refers to the light scattering spectrum obtained by the light scattering instrument, the range of scattered light intensity and scattering angle in the spectrum, and the sum value. The surface color stability value refers to the chromaticity change value of the bottom shell sandblasting after a certain period of time in the salt spray test. The optical reflectivity anisotropy value refers to the optical reflectivity of each angle of the bottom shell sandblasting measured by a multi-angle spectrophotometer to obtain the standard deviation of the optical reflectivity.

[0032] Furthermore, the surface material characteristic value, particle distribution value and force residual value of the bottom shell sandblasting are analyzed as follows: Step 301: The surface physical characteristics of the bottom shell sandblasting refer to the surface friction distribution value, surface flatness, surface gloss and surface wettability value. The sum of the surface flatness, surface gloss and surface wettability values ​​is divided by the surface friction distribution value to obtain the surface material characteristic value of the bottom shell sandblasting. The surface friction distribution value refers to the standard deviation of the friction force in different areas of the bottom shell sandblasting surface. The surface wettability value refers to the value of the contact angle of the bottom shell sandblasting. The particle distribution characteristics refer to the roundness of sandblasting particles, the electrostatic distribution value of sandblasting particles, and the crushing rate of sandblasting particles. The roundness of sandblasting particles is divided by the sum of the electrostatic distribution value of sandblasting particles and the crushing rate of sandblasting particles to obtain the particle distribution value of bottom shell sandblasting. The electrostatic distribution value of sandblasting particles refers to the static electricity of sandblasting particles. The mechanical state of bottom shell sandblasting refers to the addition of particle attachment force and residual stress to obtain the force residual value of bottom shell sandblasting, among which particle attachment force refers to the value of particles on the tape when a standard tape is pasted on the sandblasting surface of the bottom shell and the tape is torn off from the bottom shell surface at a certain speed.

[0033] Further, the bottom shell sandblasting visual evaluation set H analysis steps are as follows:

[0034] Step 201: when the visual evaluation value of the bottom shell sandblasting is in the gradient reference interval NM1, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 3 visual evaluation judgment signal; when the visual evaluation value of the bottom shell sandblasting is in the gradient reference interval NM2, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 2 visual evaluation judgment signal; when the visual evaluation value of the bottom shell sandblasting is in the gradient reference interval NM3, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 1 visual evaluation judgment signal;

[0035] Step 202: A set H is established based on the bottom shell sandblasting visual evaluation judgment signal, and the first, second and third level visual evaluation judgment signals of the bottom shell sandblasting are respectively determined as elements h1, h2 and h3.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The present invention determines the window mutual information value, pixel gradient vector, pixel convex state, pixel concave state, and sandblasting texture of the visual image to obtain a visual evaluation value of the sandblasting of the bottom shell, analyzes and determines the visual evaluation value of the sandblasting of the bottom shell and a reference interval to obtain a visual evaluation set H for sandblasting of the bottom shell, conducts an in-depth analysis of the sandblasting of the bottom shell based on the surface physical characteristics, particle distribution characteristics, mechanical state, material performance parameters, and surface optical properties of the sandblasting of the bottom shell to obtain a deep residual corrosion value of the sandblasting of the bottom shell, analyzes and determines the deep residual corrosion value of the sandblasting of the bottom shell and a reference threshold to obtain a bottom shell sandblasting depth evaluation set W, which can determine the sandblasting accuracy of the watch bottom shell from multiple dimensions and increase the accuracy of sandblasting accuracy recognition.

[0038] 2. The present invention performs union processing on the sets H and W, and analyzes the storage operations and label displays corresponding to the comprehensive evaluation signals at each level. It can classify and store watch bottom shells with different sandblasting precision qualities, and can also promptly remind process parameters with unqualified sandblasting precision, thereby improving the company's reputation, and can timely adjust the sandblasting process, reduce the company's production costs, and reduce rework and scrapping due to quality problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present invention. Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0040] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present invention.

[0041] like Figure 1 As shown, a sandblasting accuracy detection method for the bottom shell of a wearable watch, the specific steps are as follows:

[0042] Step 1: Use a camera to collect an image of the sandblasted bottom shell, and determine the window mutual information value, pixel gradient vector, pixel convex state, pixel concave state, and sandblasted texture of the visual image to obtain the visual evaluation value of the sandblasted bottom shell;

[0043] Step 2: Analyze and determine the bottom shell sandblasting visual evaluation value and the reference interval to obtain the bottom shell sandblasting visual evaluation set H;

[0044] Step 3: Conduct in-depth analysis of the bottom shell sandblasting based on the surface physical characteristics, particle distribution characteristics, mechanical state, material performance parameters and surface optical properties of the bottom shell sandblasting to obtain the deep residual corrosion value of the bottom shell sandblasting;

[0045] Step 4: Analyze and determine the bottom shell sandblasting depth residual corrosion value and the reference threshold to obtain the bottom shell sandblasting depth evaluation set W;

[0046] Step 5: By performing a union process on the sets H and W, and analyzing and obtaining the storage operations and label displays corresponding to the comprehensive evaluation signals at each level;

[0047] Among them, the specific solution steps for the bottom shell sandblasting visual evaluation are as follows:

[0048] Step 101: The light source is irradiated onto the surface of the bottom case of the watch, and image information is obtained by using a camera. The image is normalized and gray-scaled. The normalized image is gradually expanded from a 1×1 pixel window to an n×n pixel window in sequence, and the gray value of each point in each window is calculated as the reflected light intensity of each point in the window;

[0049] Step 102: Divide the reflected light intensity range of two different window scales into m intervals, calculate the probability of the emitted light intensity of each window scale in each interval, marked as edge probability, and the probability that the intensity of the two window sizes falls into their respective specific intervals at the same time, marked as joint probability, for each pair of interval combinations that appear, there are m×m, divide the joint probability by the product of the two edge probabilities, marked as division value one, take the logarithm of the division value one and multiply it by the joint probability to obtain the product value, add all the product values ​​to obtain the mutual information value of the two different window scales, repeat the above steps to obtain all the mutual information values ​​of the two different window scales, and add them to obtain the consistent value of the bottom shell sandblasting information;

[0050] Step 103: for the normalized image obtained, an image gradient calculation algorithm is used to obtain the gradient vector of each pixel in the image, and the standard deviation of the gradient vector direction in the image and the extreme value of the gradient vector modulus in the image are summed to obtain the bottom shell sandblasting gradient noise value, wherein the extreme value of the gradient vector modulus in the image refers to the maximum value of the gradient vector modulus in the image minus the minimum value;

[0051] Step 104: for the normalized image obtained, a grayscale threshold of 1 is set. When a pixel in the image is greater than the grayscale threshold of 1, the pixel is marked as a convex region candidate point 1. The reflected light grayscale dispersion value of each pixel is calculated, wherein the reflected light grayscale dispersion value is the standard deviation of the reflected light grayscale in the area centered on the pixel in the convex region candidate point 1. When the reflected light grayscale dispersion value is greater than the set threshold PT1, the loudness point is marked as a convex region candidate point 2. The convex region candidate points 1 and 2 are merged to obtain a convex region. The difference between the highest grayscale value of the convex region and the average grayscale value of the non-convex region is obtained, and a weighted calculation is performed with the number of convex regions, and the result is multiplied by the corresponding weight factor coefficient to obtain the bottom shell sandblasting convexity value.

[0052] Step 105: for the normalized image obtained, set a grayscale threshold 2, when a pixel in the image is less than the grayscale threshold 2, mark the pixel as a candidate point 1 for the concave region, calculate the grayscale discrete degree value of the reflected light of each pixel, when the grayscale discrete degree value of the reflected light is greater than the set threshold PT2, mark the pixel as a candidate point 2 for the concave region, merge the candidate points 1 and 2 for the concave region to obtain the concave region, obtain the difference between the average grayscale value of the non-concave region and the minimum grayscale value of the concave region, and perform weighted calculation with the number of concave regions, multiply by the corresponding weight factor coefficient, and obtain the concave distance value of the bottom shell sandblasting;

[0053] Step 106: The normalized image is divided into several regions, with the pixel in each region as the center. The gray value of the center pixel is marked as zh, and the gray value of the area is marked as L. q , where q = 1, 2, ..., 8, represents the sequence number of the field pixel, according to the formula Get the grayscale value of each neighborhood pixel relative to the central pixel X q , where a1, a2 and a3 are natural constants, and a2>a1>a3. The specific values ​​of a1, a2 and a3 are 1.25, 1.35 and 0 respectively. According to the formula The pair value DZZ of the central pixel is obtained, where e is a natural constant, and its specific value is 2.718. The pair values ​​of all central pixels in the image are counted, and divided into G intervals according to the pair value distribution of the central pixel of the image, and a histogram of the pair value of the central pixel is established. According to the formula The symmetric distance DJ between the A and B intervals is obtained, where g is the interval number, which is a positive integer and has a maximum value of G. The symmetric distances between the two regions in the image are counted and summed to obtain the uniform deviation of the sandblasting texture of the bottom shell. It should be noted that the smaller the symmetric distance between the regions, the better the consistency and coherence of the texture direction between the two regions;

[0054] Step 107: Mark the bottom shell sandblasting information consistency value, bottom shell sandblasting gradient noise value, bottom shell sandblasting convex value, bottom shell sandblasting concave value and bottom shell sandblasting texture uniform deviation value as yzz, pzz, ptz, atz and wgp respectively, and perform normalization processing, substitute into the set formula Spz=ta1×yzz / (ta2×pzz+ta3×ptz+ta4×atz+ta5×wgp), and calculate to obtain the bottom shell sandblasting visual evaluation value Spz, wherein ta1, ta2, ta3, ta4 and ta5 are the set influencing factor coefficients of the bottom shell sandblasting information consistency value, the bottom shell sandblasting gradient noise value, the bottom shell sandblasting convex value, the bottom shell sandblasting concave value and the bottom shell sandblasting texture uniform deviation value respectively, and the specific values ​​are determined by personnel in this professional field;

[0055] The specific solution steps for the bottom shell sandblasting visual evaluation set H are as follows:

[0056] Step 201: Setting the visual evaluation value reference intervals NM1, NM2, and NM3 for the sandblasting of the bottom shell, and substituting the visual evaluation value of the sandblasting of the bottom shell into the preset gradient reference intervals NM1, NM2, and NM3 for comparative analysis, wherein the interval values ​​of NM1, NM2, and NM3 increase in a gradient manner;

[0057] When the visual evaluation value of the bottom shell sandblasting is in the gradient reference interval NM1, the corresponding bottom shell sandblasting will send out a bottom shell sandblasting level 3 visual evaluation judgment signal; when the visual evaluation value of the bottom shell sandblasting is in the gradient reference interval NM2, the corresponding bottom shell sandblasting will send out a bottom shell sandblasting level 2 visual evaluation judgment signal; when the visual evaluation value of the bottom shell sandblasting is in the gradient reference interval NM3, the corresponding bottom shell sandblasting will send out a bottom shell sandblasting level 1 visual evaluation judgment signal;

[0058] Step 202: Establish a set H according to the bottom shell sandblasting visual evaluation judgment signal, the first-level visual evaluation judgment signal of the bottom shell sandblasting is determined as element h1, the second-level visual evaluation judgment signal of the bottom shell sandblasting is determined as element h2, and the third-level visual evaluation judgment signal of the bottom shell sandblasting is determined as element h3, and element h1∈set H, element h2∈set H, element h3∈set H;

[0059] Among them, the specific steps for solving the deep residual corrosion value of the bottom shell sandblasting are as follows:

[0060] Step 301: The bottom shell sandblasting is deeply analyzed based on the surface physical characteristics, particle distribution characteristics, mechanical state, material performance parameters and surface optical properties of the bottom shell sandblasting. The surface physical characteristics of the bottom shell sandblasting refer to the surface friction distribution value, surface flatness, surface glossiness and surface wettability. The sum of the surface flatness, surface glossiness and surface wettability values ​​is divided by the surface friction distribution value to obtain the surface physical characteristics of the bottom shell sandblasting, wherein the surface friction distribution value refers to the standard deviation of the friction force in different areas of the bottom shell sandblasting surface, and the surface wettability value refers to the value of the contact angle of the bottom shell sandblasting; the particle distribution characteristics of the bottom shell sandblasting refer to the roundness of the sandblasting particles, the electrostatic distribution value of the sandblasting particles and the breakage rate of the sandblasting particles. The roundness of the sandblasting particles is divided by the sandblasting particles. The sum of the electrostatic distribution value and the sandblasting particle breakage rate is used to obtain the particle distribution value of the bottom shell sandblasting. The electrostatic distribution value of the sandblasting particles refers to the static electricity of the sandblasting particles; the mechanical state of the bottom shell sandblasting refers to the addition of the particle attachment force and the residual stress to obtain the force residual value of the bottom shell sandblasting, where the particle attachment force refers to the value of the particles on the tape when the standard tape is pasted on the sandblasting surface of the bottom shell and the tape is torn off from the bottom shell surface at a certain speed; the material performance parameters of the bottom shell sandblasting refer to the weighted calculation of the sandblasting chemical activity, the galvanic corrosion sensitivity of the bottom shell, and the biocompatibility, and multiplied by the corresponding influencing factor coefficient to obtain the material corrosion capacity value of the bottom shell sandblasting, where the galvanic corrosion sensitivity of the bottom shell refers to the polarization curve measurement of the bottom shell after sandblasting to obtain the corrosion current The value of density, biocompatibility refers to the number of itching, erythema, rash, swelling, redness and fever that occur when wearing the watch for a certain period of time. The higher the sandblasting accuracy, the gentler the physical contact between the skin and the bottom plate of the watch, and the fewer the number of adverse reactions. The surface optical performance of the sandblasted bottom shell refers to the sum of the surface light scattering characteristic value, surface color stability value, and optical reflectivity anisotropy value to obtain the sandblasted surface light variation value of the bottom shell, among which the surface light scattering characteristic value refers to the light scattering spectrum obtained by a light scattering instrument, the range of scattered light intensity and scattering angle in the spectrum, and the sum of the values. The surface color stability value refers to the chromaticity change value of the sandblasted bottom shell after a certain period of time in the salt spray test, and the optical reflectivity anisotropy value refers to the optical reflectivity obtained by multi-angle spectroscopy. The optical reflectivity of the bottom shell sandblasting at each angle is measured by the meter to obtain the standard deviation of the optical reflectivity; Step 302: The surface material characteristic value of the bottom shell sandblasting, the particle distribution value of the bottom shell sandblasting, the force bias residual value of the bottom shell sandblasting, the material corrosion capacity value of the bottom shell sandblasting, and the surface light variation value of the bottom shell sandblasting are normalized, and the force bias residual value of the bottom shell sandblasting is used as the circle radius of the upper surface of the truncated cone, the material corrosion capacity value of the bottom shell sandblasting is used as the circle radius of the lower surface of the truncated cone, and the surface light variation value of the bottom shell sandblasting is used as the height of the truncated cone, and the center point of the truncated cone is used as the center of the sphere, and a sphere is constructed inside the truncated cone, and the sum of the surface material characteristic value of the bottom shell sandblasting and the particle distribution value of the bottom shell sandblasting is used as the sphere radius, and the non-overlapping volume formed by the truncated cone and the sphere is identified, and marked as the deep residual corrosion value of the bottom shell sandblasting;

[0061] Among them, the specific solution steps of the bottom shell sandblasting depth evaluation set W are as follows:

[0062] Step 401: setting reference thresholds TG1 and TG2 for the deep residual corrosion value of the bottom shell sandblasting, wherein the reference threshold TG1>TG2, and comparing and analyzing the deep residual corrosion value of the bottom shell sandblasting with the reference thresholds TG1 and TG2;

[0063] Step 402: when the deep residual corrosion value of the bottom shell sandblasting is greater than or equal to the reference threshold value TG1, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 3 depth assessment judgment signal; when the deep residual corrosion value of the bottom shell sandblasting is less than the reference threshold value TG1 and greater than the reference threshold value TG2, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 2 depth assessment judgment signal; when the deep residual corrosion value of the bottom shell sandblasting is less than or equal to the reference threshold value TG2, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 1 depth assessment judgment signal;

[0064] Step 403: Establish a set W according to the bottom shell sandblasting depth assessment and determination signal, the bottom shell sandblasting first-level depth assessment and determination signal is determined as element w1, the bottom shell sandblasting second-level depth assessment and determination signal is determined as element w2, the bottom shell sandblasting third-level depth assessment and determination signal is determined as element w3, and element w1∈set W, element w2∈set W, element w3∈set W;

[0065] Among them, the specific solution steps for the storage operation and label display corresponding to the comprehensive evaluation signals at each level are as follows:

[0066] Step 501: performing a union process on the sets H and W, and when H∪W={h1, w1}, a first-level comprehensive evaluation signal for determining the sandblasting accuracy of the bottom shell is generated;

[0067] When H∪W={h1, w2} or {h1, w3} or {h2, w1} or {h2, w2} or {h3, w1}, the secondary comprehensive evaluation signal for the sandblasting accuracy of the bottom shell is obtained;

[0068] When H∪W={h2, w3} or {h3, w2}, a three-level comprehensive evaluation signal for determining the accuracy of bottom shell sandblasting is generated;

[0069] When H∪W={h3, w3}, a four-level comprehensive evaluation signal for determining the accuracy of bottom shell sandblasting is generated;

[0070] Step 502: If the sandblasting of the bottom shell generates a first-level comprehensive evaluation signal for determining the accuracy of the sandblasting of the bottom shell, the sandblasting of the bottom shell is reclassified into set Y1, put into storage, and a label is displayed indicating "the sandblasting accuracy of the bottom shell is good"; if the sandblasting of the bottom shell generates a second-level comprehensive evaluation signal for determining the accuracy of the sandblasting of the bottom shell, the sandblasting of the bottom shell is reclassified into set Y2, put into storage, and a label is displayed indicating "the sandblasting accuracy of the bottom shell is qualified"; if the sandblasting of the bottom shell generates a third-level comprehensive evaluation signal for determining the accuracy of the sandblasting of the bottom shell, the sandblasting of the bottom shell is reclassified into set Y3, and storage is not allowed for the sandblasting, and the sandblasting process parameters and procedures at this time need to be checked and adjusted"; if the sandblasting of the bottom shell generates a fourth-level comprehensive evaluation signal for determining the accuracy of the sandblasting of the bottom shell, the sandblasting of the bottom shell is reclassified into set Y4, and storage is not allowed for the sandblasting, and a label is displayed indicating "the sandblasting accuracy of the bottom shell is seriously unqualified, and it is scrapped, and the sandblasting process parameters and procedures at this time need to be checked and adjusted".

[0071] The above is an explanation of the present invention and should not be considered as a limitation thereof. Although several exemplary embodiments of the present invention have been described, it will be readily appreciated by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present invention. Therefore, all such modifications are intended to be included within the scope of the present invention as defined in the claims. It should be understood that the above is an explanation of the present invention and should not be considered as being limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present invention is defined by the claims and their equivalents.

Claims

1. A sandblasting accuracy detection method for a wearable watch bottom shell, characterized in that: The following steps are involved: Step 1: Use a camera to collect an image of the sandblasted bottom shell, and determine the window mutual information value, pixel gradient vector, pixel convex state, pixel concave state, and sandblasted texture of the visual image to obtain the visual evaluation value of the sandblasted bottom shell; Step 2: Analyze and determine the bottom shell sandblasting visual evaluation value and the reference interval to obtain the bottom shell sandblasting visual evaluation set H; Step 3: Conduct in-depth analysis of the bottom shell sandblasting based on the surface physical characteristics, particle distribution characteristics, mechanical state, material performance parameters and surface optical properties of the bottom shell sandblasting to obtain the deep residual corrosion value of the bottom shell sandblasting; Step 4: Analyze and determine the bottom shell sandblasting depth residual corrosion value and the reference threshold value to obtain the bottom shell sandblasting depth evaluation set W; Step 5: By performing a union process on the sets H and W, and analyzing them, the storage operations and label displays corresponding to the comprehensive evaluation signals at each level are obtained.

2. A sandblasting accuracy detection method for a bottom shell of a wearable watch according to claim 1, characterized in that: The bottom shell sandblasting visual evaluation analysis steps are as follows: Step 106: The normalized image is divided into several regions, with the pixel in each region as the center. The gray value of the center pixel is marked as zh, and the gray value of the area is marked as L. q , where q = 1, 2, ..., 8, represents the sequence number of the field pixel, according to the formula Get the grayscale value of each neighborhood pixel relative to the central pixel X q , where a1, a2 and a3 are natural constants, according to the formula The pair value DZZ of the central pixel is obtained, where e is a natural constant. The pair values ​​of all central pixels in the image are counted, and divided into G intervals according to the pair value distribution of the central pixel of the image. The pair value histogram of the central pixel is established according to the formula The symmetric distance DJ between the A and B intervals is obtained, where g is the interval number, which is a positive integer and has a maximum value of G. The symmetric distances between each two regions in the image are counted and summed to obtain the uniform deviation of the sandblasting texture of the bottom shell; Step 107: Calculate the bottom shell sandblasting information consistency value, the bottom shell sandblasting gradient value, the bottom shell sandblasting convex value, the bottom shell sandblasting concave value and the bottom shell sandblasting texture deviation value to obtain the bottom shell sandblasting visual evaluation value Spz.

3. A sandblasting accuracy detection method for a bottom shell of a wearable watch according to claim 2, characterized in that: The analysis steps of the bottom shell sandblasting convex stroke value and the bottom shell sandblasting concave stroke value are as follows: Step 104: for the normalized image obtained, a grayscale threshold of 1 is set. When a pixel in the image is greater than the grayscale threshold of 1, the pixel is marked as a convex region candidate point 1. The reflected light grayscale dispersion value of each pixel is calculated. The reflected light grayscale dispersion value is the standard deviation of the reflected light grayscale in the area centered on the pixel in the convex region candidate point 1. When the reflected light grayscale dispersion value is greater than the set threshold PT1, the loudness point is marked as a convex region candidate point 2. The convex region candidate points 1 and 2 are merged to obtain a convex region. The difference between the highest grayscale value of the convex region and the average grayscale value of the non-convex region is obtained, and a weighted calculation is performed with the number of convex regions, and the result is multiplied by the corresponding weight factor coefficient to obtain the bottom shell sandblasting convexity value. Step 105: For the normalized image obtained, set grayscale threshold 2. When a pixel in the image is less than grayscale threshold 2, mark the pixel as candidate point 1 for the concave area. Calculate the grayscale discreteness value of the reflected light of each pixel. When the grayscale discreteness value of the reflected light is greater than the set threshold PT2, mark the pixel as candidate point 2 for the concave area. Merge candidate points 1 and 2 for the concave area to obtain the concave area. Obtain the difference between the average grayscale value of the non-concave area and the minimum grayscale value of the concave area. Perform weighted calculation on the difference with the number of concave areas and multiply by the corresponding weight factor coefficient to obtain the bottom shell sandblasting concave value.

4. The sandblasting accuracy detection method for the bottom shell of a wearable watch according to claim 2, characterized in that: The steps for analyzing the consistent value of the bottom shell sandblasting information and the bottom shell sandblasting gradient noise value are as follows: Step 101: The light source is irradiated onto the surface of the bottom case of the watch, and image information is obtained by using a camera. The image is normalized and gray-scaled. The normalized image is gradually expanded from a 1×1 pixel window to an n×n pixel window in sequence, and the gray value of each point in each window is calculated as the reflected light intensity of each point in the window; Step 102: Divide the reflected light intensity range of two different window scales into m intervals, calculate the probability of the emitted light intensity of each window scale in each interval, marked as edge probability, and the probability that the intensity of the two window sizes falls into their respective specific intervals at the same time, marked as joint probability, for each pair of interval combinations that appear, there are m×m, divide the joint probability by the product of the two edge probabilities, marked as division value one, take the logarithm of the division value one and multiply it by the joint probability to obtain the product value, add all the product values ​​to obtain the mutual information value of the two different window scales, repeat the above steps to obtain all the mutual information values ​​of the two different window scales, and add them to obtain the consistent value of the bottom shell sandblasting information; Step 103: For the normalized image obtained, an image gradient calculation algorithm is used to obtain the gradient vector of each pixel in the image, and the standard deviation of the gradient vector direction in the image and the extreme value of the gradient vector modulus in the image are summed to obtain the bottom shell sandblasting gradient noise value.

5. The sandblasting accuracy detection method for the bottom shell of a wearable watch according to claim 2, characterized in that: The steps of storage operation and label display analysis corresponding to the comprehensive evaluation signals of each level are as follows: Step 501: Perform a union process on the bottom shell sandblasting visual evaluation set H and the bottom shell sandblasting depth evaluation set W. When H∪W={h1, w1}, a first-level comprehensive evaluation signal for bottom shell sandblasting accuracy determination is generated; when H∪W={h1, w2} or {h1, w3} or {h2, w1} or {h2, w2} or {h3, w1}, a second-level comprehensive evaluation signal for bottom shell sandblasting accuracy determination is generated; when H∪W={h2, w3} or {h3, w2}, a third-level comprehensive evaluation signal for bottom shell sandblasting accuracy determination is generated; when H∪W={h3, w3}, a fourth-level comprehensive evaluation signal for bottom shell sandblasting accuracy determination is generated; Step 502: If the sandblasting of the bottom shell generates a first-level comprehensive evaluation signal for determining the accuracy of the bottom shell sandblasting, the sandblasting of the bottom shell will be reclassified into set Y1, put into storage, and the label will display label one; if the sandblasting of the bottom shell generates a second-level comprehensive evaluation signal for determining the accuracy of the bottom shell sandblasting, the sandblasting of the bottom shell will be reclassified into set Y2, put into storage, and the label will display label one; if the sandblasting of the bottom shell generates a third-level comprehensive evaluation signal for determining the accuracy of the bottom shell sandblasting, the sandblasting of the bottom shell will be reclassified into set Y3, and storage into storage is not allowed, and the label will display label one; if the sandblasting of the bottom shell generates a fourth-level comprehensive evaluation signal for determining the accuracy of the bottom shell sandblasting, the sandblasting of the bottom shell will be reclassified into set Y4, and storage into storage is not allowed, and the label will display label one.

6. A sandblasting accuracy detection method for a bottom case of a wearable watch according to claim 5, characterized in that: The bottom shell sandblasting depth assessment set W analysis steps are as follows: Step 402: when the deep residual corrosion value of the bottom shell sandblasting is greater than or equal to the reference threshold value TG1, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 3 depth assessment judgment signal; when the deep residual corrosion value of the bottom shell sandblasting is less than the reference threshold value TG1 and greater than the reference threshold value TG2, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 2 depth assessment judgment signal; when the deep residual corrosion value of the bottom shell sandblasting is less than or equal to the reference threshold value TG2, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 1 depth assessment judgment signal; Step 403: A set W is established according to the bottom shell sandblasting depth assessment and determination signal, and the bottom shell sandblasting first, second and third level depth assessment and determination signals are respectively determined as elements w1, w2 and w3.

7. A sandblasting accuracy detection method for a bottom case of a wearable watch according to claim 6, characterized in that: The bottom shell sandblasting depth assessment set W analysis steps are as follows: Step 302: The surface material characteristic value of the sandblasting of the bottom shell, the particle distribution value of the sandblasting of the bottom shell, the force bias residual value of the sandblasting of the bottom shell, the material corrosion capacity value of the sandblasting of the bottom shell, and the surface light variation value of the sandblasting of the bottom shell are normalized, and the force bias residual value of the sandblasting of the bottom shell is used as the radius of the circle on the upper surface of the frustum, the material corrosion capacity value of the sandblasting of the bottom shell is used as the radius of the circle on the lower surface of the frustum, and the surface light variation value of the sandblasting of the bottom shell is used as the height of the frustum to construct a frustum, and the center point of the frustum is used as the center of the sphere, and a sphere is constructed inside the frustum, and the sum of the surface material characteristic value of the sandblasting of the bottom shell and the particle distribution value of the sandblasting of the bottom shell is used as the radius of the sphere, and the non-overlapping volume formed by the frustum and the sphere is identified, and marked as the deep residual corrosion value of the sandblasting of the bottom shell.

8. A sandblasting accuracy detection method for a bottom case of a wearable watch according to claim 7, characterized in that: The steps for analyzing the material corrosion capacity value and the surface light variation value of the bottom shell sandblasting are as follows: The material performance parameters of the bottom shell sandblasting refer to the weighted calculation of the sandblasting chemical activity, the galvanic corrosion sensitivity of the bottom shell, and the biocompatibility, and multiply them by the corresponding influencing factor coefficients to obtain the material corrosion capacity value of the bottom shell sandblasting. The galvanic corrosion sensitivity of the bottom shell refers to the polarization curve measurement of the bottom shell after sandblasting to obtain the value of the corrosion current density. Biocompatibility refers to the number of times itching, erythema, rash, swelling, redness and fever occur after wearing the watch for a certain period of time. The surface optical performance of the bottom shell sandblasting refers to the sum of the surface light scattering characteristic value, the surface color stability value, and the optical reflectivity anisotropy value to obtain the surface light variation value of the bottom shell sandblasting. The surface light scattering characteristic value refers to the light scattering spectrum obtained by the light scattering instrument, the range of scattered light intensity and scattering angle in the spectrum, and the sum value. The surface color stability value refers to the chromaticity change value of the bottom shell sandblasting after a certain period of time in the salt spray test. The optical reflectivity anisotropy value refers to the optical reflectivity of each angle of the bottom shell sandblasting measured by a multi-angle spectrophotometer to obtain the standard deviation of the optical reflectivity.

9. The sandblasting accuracy detection method for the bottom shell of a wearable watch according to claim 7, characterized in that: The analysis steps of the surface material characteristic value, particle distribution value and force residual value of the bottom shell sandblasting are as follows: Step 301: The surface physical characteristics of the bottom shell sandblasting refer to the surface friction distribution value, surface flatness, surface glossiness, and surface wettability. The sum of the surface flatness, surface glossiness, and surface wettability values ​​is divided by the surface friction distribution value to obtain the surface physical characteristics of the bottom shell sandblasting. The surface friction distribution value refers to the standard deviation of the friction force in different areas of the bottom shell sandblasting surface. The surface wettability value refers to the value of the contact angle of the bottom shell sandblasting. The particle distribution characteristics of the bottom shell sandblasting refer to the roundness of the sandblasting particles, the static value of the sandblasting particles, and the particle size distribution of the sandblasting particles. Electrostatic distribution value and sandblasting particle breakage rate. The roundness of sandblasting particles is divided by the sum of the electrostatic distribution value of sandblasting particles and the breakage rate of sandblasting particles to obtain the particle distribution value of bottom shell sandblasting. The electrostatic distribution value of sandblasting particles refers to the static electricity of sandblasting particles. The mechanical state of bottom shell sandblasting refers to the addition of particle attachment force and residual stress to obtain the force residual value of bottom shell sandblasting. Among them, the particle attachment force refers to the value of particles on the tape when a standard tape is pasted on the sandblasting surface of the bottom shell and the tape is torn off from the bottom shell surface at a certain speed.

10. The sandblasting accuracy detection method for the bottom shell of a wearable watch according to claim 1, characterized in that: The bottom shell sandblasting visual assessment set H analysis steps are as follows: Step 201: when the visual evaluation value of the bottom shell sandblasting is in the gradient reference interval NM1, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 3 visual evaluation judgment signal; when the visual evaluation value of the bottom shell sandblasting is in the gradient reference interval NM2, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 2 visual evaluation judgment signal; when the visual evaluation value of the bottom shell sandblasting is in the gradient reference interval NM3, the corresponding bottom shell sandblasting sends a bottom shell sandblasting level 1 visual evaluation judgment signal; Step 202: A set H is established based on the bottom shell sandblasting visual evaluation judgment signal, and the first, second and third level visual evaluation judgment signals of the bottom shell sandblasting are respectively determined as elements h1, h2 and h3.

Images