Machining measurement system and method based on Internet of Things

Abstract

The invention discloses a machining measurement system and method based on the Internet of Things, and relates to the field of image processing, the machining measurement system comprises a measurement center, and the measurement center comprises a data acquisition module, a data preprocessing module, a data processing module and a data display module; the data acquisition module is used for performing data acquisition on a machining process to obtain corresponding vibration data and image data; the data preprocessing module is used for judging whether the corresponding machined parts meet requirements or not according to the collected vibration data, and if the corresponding machined parts meet the requirements, data preprocessing is conducted on image data of the corresponding machined parts to obtain corresponding visual image data; and the data processing module is used for carrying out data processing on the obtained visual image data to obtain a corresponding part edge contour. And the data display module is used for generating a corresponding measurement report according to the obtained visual feature points and the edge contour of the part. According to the invention, the measurement precision of the processed part can be improved.

Description

Technical Field

[0001] The present invention relates to the technical fields of the Internet of Things and image processing, and particularly to a mechanical processing measurement system and method based on the Internet of Things. Background Art

[0002] In the mechanical processing industry, especially in high-precision mechanical processing, dimensional and precision measurements are required for each key processing operation. It can be said that if the processing technology is not proficient enough, once an error occurs, it will surely affect the coordination of all links in the entire mechanical processing, resulting in an interruption in continuity. Eventually, it will not only easily lead to low efficiency in the mechanical processing process but also pose a potential risk to the quality of mechanical processing.

[0003] Compared with the prior art, traditional mechanical processing measurement methods usually rely on contact measurement tools such as micrometers and vernier calipers. Although these methods are simple, they are inefficient and difficult to meet the measurement requirements for complex shapes. These are the problems we need to solve. For this reason, we provide a mechanical processing measurement system and method based on the Internet of Things. Summary of the Invention

[0004] The purpose of the present invention is to provide a mechanical processing measurement system and method based on the Internet of Things.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A mechanical processing measurement system based on the Internet of Things includes a measurement center, and the measurement center includes a data acquisition module, a data preprocessing module, a data processing module, and a data display module;

[0006] The data acquisition module is used to collect data during the mechanical processing process to obtain corresponding mechanical processing data, and the mechanical processing data includes vibration data and image data;

[0007] The data preprocessing module is used to determine whether the processed parts that have completed mechanical processing meet the requirements based on the collected vibration data. If they do not meet the requirements, feedback is given. If they meet the requirements, the image data of the corresponding processed parts is preprocessed to obtain corresponding visual image data;

[0008] The data processing module is used to process the obtained visual image data to obtain corresponding part images, and perform contour extraction on the obtained part images to obtain corresponding part edge contours;

[0009] The data display module is used to generate corresponding measurement reports based on the obtained visual feature points and part edge contours.

[0010] Further, the data acquisition module consists of several Internet of Things terminals and a measurement database. The Internet of Things terminals include a vibration acquisition terminal and an image acquisition terminal, and the machining data includes vibration data and image data;

[0011] The vibration acquisition terminal is used to collect the vibration times of the cutting tool during the machining process of the machined part, and obtain the corresponding vibration data; the image acquisition terminal is used to collect images of the machined parts after machining, obtain the corresponding image data, and upload the collected machining data to the measurement database for storage.

[0012] Further, the process of the data preprocessing module for judging whether the machined parts corresponding to the collected vibration data meet the requirements includes:

[0013] Set the vibration threshold range, read the vibration data in the collected machining data, and compare the collected vibration data with the corresponding vibration threshold;

[0014] If the vibration data corresponding to the corresponding machined part is outside the vibration threshold range, mark the corresponding machined part as a defective part and feedback it to the measurement center;

[0015] If the vibration data corresponding to the corresponding machined part is within the vibration threshold range, retain the image data corresponding to the corresponding machined part and preprocess the retained image data.

[0016] Further, the process of preprocessing the image data includes:

[0017] Count the retained image data, and mark the left view image and the right view image in the corresponding image data as Lq and Rq respectively;

[0018] Obtain all the pixel points in the left view image Lq and the values corresponding to the RGB color components of the corresponding pixel points. Taking the upper left corner pixel point of the left view image as the origin, construct a corresponding two-dimensional rectangular coordinate system, and mark the coordinates of the pixel points in the corresponding left view image Lq as (i, j);

[0019] Obtain the RGB components corresponding to the pixel point at (i, j) of the left view image Lq, and obtain the average value corresponding to the corresponding RGB components, and use it as the gray value of the corresponding pixel point;

[0020] Traverse each pixel point in the left view image Lq, use the same method to obtain the gray value of the corresponding pixel point, and then obtain the corresponding gray image based on the obtained gray values;

[0021] Construct a Gaussian template window based on Gaussian filtering technology. Taking a certain pixel point as an example, place this pixel point at the center of the Gaussian template window, and perform a convolution operation on the template coefficients and the corresponding pixel points under the Gaussian template window; and use the result of the convolution operation to replace the corresponding gray value at the center position of the Gaussian template window, thereby obtaining the corresponding filtered gray image;

[0022] Process the right-view image Rq using the same method to obtain the corresponding right-view gray image;

[0023] Summarize the left-view gray image and the right-view gray image of the corresponding machined part to obtain relevant visual image data.

[0024] Further, the process of the data processing module processing the obtained visual image data to obtain the corresponding part image includes:

[0025] Read the obtained visual image data, and respectively extract feature points from the visual image data based on the SIFT algorithm to obtain the corresponding visual feature points, where the visual feature points include left-view key feature points and right-view key feature points;

[0026] For the extracted visual feature points, match the left-view key feature points and the right-view key feature points, and obtain the gray deviation value between the matched left-view key feature points and right-view key feature points, and then obtain the corresponding disparity value based on the obtained gray deviation value. The disparity value represents the pixel offset of the right-view gray image relative to the left-view gray image;

[0027] Based on the obtained disparity value, adjust the position of the right key feature points in the right-view gray image to align it with the left-view gray image to obtain the corresponding right visual compensation map;

[0028] Integrate the right visual compensation image and the left-view gray image to obtain the final part image.

[0029] Further, the process of processing the obtained part image to obtain the corresponding part edge contour includes:

[0030] Read the obtained part image, and obtain the gray values corresponding to all pixel points in the corresponding part image, and record them as Zw; mark the maximum value and the minimum value in the obtained gray values as Zmax and Zmin respectively;

[0031] Set the initial gray threshold T0;

[0032] Compare the gray values corresponding to all pixel points in the part image with the initial gray threshold;

[0033] If Zw≤T0, then mark the corresponding pixel point as a low-gray point;

[0034] If Zw > T0, mark the corresponding pixel points as high - gray - level points;

[0035] Respectively count the gray - level values corresponding to all the marked low - gray - level points and high - gray - level points, and obtain the corresponding average values, which are respectively marked as TA and TB;

[0036] Obtain the second gray - level threshold T1 based on the obtained TA and TB, use T1 to replace the set initial gray - level threshold T0, and repeat the above process for threshold iteration until Tp + 1 = Tp; end the threshold - iteration process, and take the corresponding Tp + 1 as the final gray - level threshold;

[0037] Compare the gray - level value corresponding to the pixel points in the corresponding part image with the obtained final gray - level threshold Tp + 1;

[0038] If Zw ≤ Tp + 1, then replace the gray - level value corresponding to the corresponding pixel point with '0';

[0039] If Zw > Tp + 1, then replace the gray - level value corresponding to the corresponding pixel point with '255';

[0040] Replace the pixel points in the left - view gray - level image according to the above process to obtain the corresponding binary image;

[0041] Taking the pixel point at the center of the binary image as the origin, construct a two - dimensional rectangular coordinate system, then the coordinates of the pixel points in the corresponding binary image can be expressed as (x, y), and the gray - level value corresponding to the corresponding pixel point is expressed as f(x, y);

[0042] Based on the edge - detection algorithm, determine the edge points of the binary image in the y - direction;

[0043] If f(x, y) > [f(x - 1, y)+f(x + 1, y)] / 2, then the pixel point (x, y) is the edge point of the y - th column. Calculate all the pixel points in the binary image based on the above process to obtain the boundary points between the upper and lower edges of the corresponding processed part and the background, and mark them as edge pixel points;

[0044] Use the same method to detect the edges in the x - direction of the binary image to obtain the edge pixel points between the left and right edges of the corresponding processed part and the background;

[0045] Count the obtained edge pixel points, and obtain the part - edge contour of the corresponding processed part based on them.

[0046] Furthermore, the process by which the data - display module generates the corresponding measurement report based on the obtained visual feature points and part - edge contour includes:

[0047] Obtain the pixel coordinates corresponding to the inner-edge pixel points of the edge contour of the corresponding part;

[0048] According to the obtained pixel coordinates, the left-view key feature points and the right-view key feature points, and in combination with the acquisition parameters of the image acquisition terminal, obtain the pixel size corresponding to the corresponding machined part;

[0049] Based on big data technology, obtain the pixel equivalent corresponding to the used image acquisition terminal;

[0050] According to the obtained pixel equivalent, convert the obtained pixel measurement data to obtain the actual measurement data of the corresponding machined part;

[0051] Compare the obtained actual measurement data with the preset tolerance range. If the obtained actual measurement data does not meet the preset tolerance range, generate a non-conformance report, feedback to the staff, and at the same time screen and remove the corresponding machined part;

[0052] If the obtained actual measurement data meets the preset tolerance range, indicating that the corresponding machined part meets the requirements, then label the corresponding machined part, generate the corresponding measurement report, and store it in the measurement center.

[0053] Furthermore, a mechanical processing measurement method based on the Internet of Things includes the following steps:

[0054] Step 1: Collect data during the mechanical processing to obtain the corresponding mechanical processing data, where the mechanical processing data includes vibration data and image data;

[0055] Step 2: According to the collected vibration data, judge whether the machined part that has completed mechanical processing meets the requirements. If it does not meet the requirements, give feedback. If it meets the requirements, perform data preprocessing on the image data of the corresponding machined part to obtain the corresponding visual image data;

[0056] Step 3: Extract features from the obtained visual image data to obtain the corresponding visual feature points. At the same time, extract the contour from the obtained visual image data to obtain the edge contour of the corresponding part;

[0057] Step 4: Convert the obtained visual feature points and part contour feature data to obtain the actual measurement data of the corresponding machined part, and based on the obtained actual measurement data, judge whether the corresponding machined part meets the requirements. If it meets the requirements, generate the corresponding measurement report.

[0058] Compared with the prior art, the beneficial effects of the present invention are:

[0059] 1. By automating the measurement and data analysis of machined parts, the cost and time of manual measurement are reduced, and the production cost is lowered.

[0060] 2. Through image acquisition and Internet of Things transmission, non-contact measurement is achieved, avoiding damage that may be caused by physical contact, and at the same time facilitating the improvement of measurement efficiency.

[0061] 3. By using binocular vision to acquire images of machined parts, it helps to obtain more accurate and comprehensive measurement data of the machined parts, improving the measurement accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is the system schematic diagram of the present invention.

[0063] Figure 2 It is the system flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0064] As Figure 1 shown, an Internet of Things-based mechanical processing measurement system includes a measurement center, and the measurement center includes a data acquisition module, a data preprocessing module, a data processing module, and a data display module;

[0065] The data acquisition module is used to acquire data during the mechanical processing to obtain corresponding mechanical processing data, and the mechanical processing data includes vibration data and image data;

[0066] The data preprocessing module is used to judge whether the machined parts that have completed the mechanical processing meet the requirements based on the acquired vibration data. If they do not meet the requirements, feedback is given. If they meet the requirements, the image data of the corresponding machined parts is preprocessed to obtain corresponding visual image data;

[0067] The data processing module is used to process the obtained visual image data to obtain corresponding part images, and extract the contours of the obtained part images to obtain corresponding part edge contours;

[0068] The data display module is used to generate corresponding measurement reports based on the obtained visual feature points and part edge contours.

[0069] It should be further noted that in the specific implementation process, the process of the data acquisition module acquiring data from the machined parts of the mechanical processing to obtain corresponding mechanical processing data includes:

[0070] The data acquisition module consists of several Internet of Things terminals and a measurement database. The Internet of Things terminals include a vibration acquisition terminal and an image acquisition terminal. Data is collected from the machining process through the Internet of Things terminals to obtain corresponding machining data, which includes vibration data and image data.

[0071] The vibration acquisition terminal is used to collect the vibration times of the tool during the machining process of the machined part to obtain corresponding vibration data. The image acquisition terminal is used to collect images of the machined parts after machining to obtain corresponding image data, which includes left-view images and right-view images.

[0072] The collected machining data is uploaded to the measurement database for storage.

[0073] It should be further noted that in the specific implementation process, the process of the data preprocessing module for preprocessing the obtained machining data includes:

[0074] Set the vibration threshold range, read the vibration data in the collected machining data, and compare the collected vibration data with the corresponding vibration threshold.

[0075] If the vibration data corresponding to the corresponding machined part is outside the vibration threshold range, it indicates that there is a defect in the machined part during the machining process. Then, mark the corresponding machined part as a defective part and feedback it to the measurement center. After receiving the feedback, the measurement center will delete the corresponding image data of the machined part from the measurement database, and at the same time, adaptively adjust the vibration frequency of the corresponding machining tool and implement supervision. If the vibration data of the corresponding tool meets the requirements within several acquisition cycles, stop the supervision. If it does not meet the requirements, feedback the tool failure information to the staff, and at the same time, mark the position of the corresponding tool and feedback it to the staff.

[0076] If the vibration data corresponding to the corresponding machined part is within the vibration threshold range, retain the corresponding image data of the machined part.

[0077] Statistically analyze the retained image data and number it, denoted as q, where q = 1, 2,..., Q. Then, the left-view image and right-view image in the corresponding image data are respectively marked as Lq and Rq.

[0078] Obtain all the pixel points in the left-view image Lq and the values corresponding to the RGB color components of the corresponding pixel points. Taking the upper-left pixel point of the left-view image as the origin, construct a corresponding two-dimensional rectangular coordinate system, and mark the coordinates of the pixel points in the corresponding left-view image Lq as (i, j).

[0079] Obtain the RGB components corresponding to the pixel at (i, j) of the left-view image Lq, obtain the average value corresponding to the corresponding RGB component, and use it as the grayscale value of the corresponding pixel. The corresponding mathematical formula is as follows:

[0080]

[0081] In the formula, F(i, j) represents the grayscale value at (i, j) of the target image;

[0082] Traverse each pixel in the left-view image Lq, obtain the grayscale value of the corresponding pixel using the same method, and then obtain the corresponding grayscale image based on the obtained grayscale values;

[0083] Construct a Gaussian template window based on the Gaussian filtering technology. Taking a certain pixel as an example, place this pixel at the center of the Gaussian template window, and perform a convolution operation on the template coefficient and the corresponding pixel under the Gaussian template window; and use the convolution operation result to replace the grayscale value corresponding to the center position of the Gaussian template window, thereby obtaining the corresponding filtered grayscale image;

[0084] In an embodiment of the present invention, a 3×3 Gaussian template window is constructed based on the Gaussian filtering algorithm. The corresponding Gaussian template window can be expressed as Obtain the grayscale values corresponding to some pixels of the corresponding grayscale image, then it can be expressed as

[0085] where kt is the template coefficient, t = 0, 1, ……, 8 and Coincide the center point k0 of the Gaussian template window with the pixel with the grayscale value a0 of the grayscale image. Then the corresponding output G of the Gaussian template window at a0 can be expressed as In the formula, G is the grayscale value at the position a0 of the filtered grayscale image. By analogy, traverse the entire grayscale image based on the Gaussian template window to obtain the corresponding filtered grayscale image, and denote it as the left-view grayscale image;

[0086] Process the right-view image Rq using the same method to obtain the corresponding right-view grayscale image;

[0087] Summarize the left-view grayscale image and the right-view grayscale image of the corresponding machined part to obtain the relevant visual image data;

[0088] It should be further noted that in the specific implementation process, the data processing module is used to extract features from the obtained visual image data, and at the same time, the process of extracting the contour from the obtained visual image data includes:

[0089] Read the obtained visual image data, extract feature points from the visual image data respectively based on the SIFT algorithm to obtain corresponding visual feature points, and the visual feature points include left-view key feature points and right-view key feature points;

[0090] For the extracted visual feature points, match the left-view key feature points and the right-view key feature points, and obtain the gray-scale deviation value between the matched left-view key feature points and right-view key feature points. Furthermore, obtain the corresponding disparity value based on the obtained gray-scale deviation value, and the disparity value represents the pixel offset of the right-view gray-scale image relative to the left-view gray-scale image;

[0091] Based on the obtained disparity value, adjust the position of the right key feature points in the right-view gray-scale image to align it with the left-view gray-scale image to obtain the corresponding right visual compensation map;

[0092] Integrate the right visual compensation image with the left-view gray-scale image to obtain the final part image;

[0093] Read the obtained part image, obtain the gray-scale values corresponding to all pixel points in the corresponding part image, and number them, and denote them as Zw, where w = 1, 2,..., W, W > 0 and W is an integer;

[0094] Mark the maximum value and the minimum value in the obtained gray-scale values as Zmax and Zmin respectively;

[0095] Set the initial gray-scale threshold T0, where,

[0096] Compare the gray-scale values corresponding to all pixel points in the part image with the initial gray-scale threshold;

[0097] If Zw ≤ T0, mark the corresponding pixel point as a low-gray-scale point;

[0098] If Zw > T0, mark the corresponding pixel point as a high-gray-scale point;

[0099] Statistically analyze the gray-scale values corresponding to all marked low-gray-scale points and high-gray-scale points respectively, and obtain the corresponding average values, and mark them as TA and TB respectively;

[0100] Obtain the second gray-scale threshold T1 based on the obtained TA and TB, where, Use T1 to replace the set initial gray-scale threshold T0, and repeat the above process for threshold iteration until Tp+1 = Tp; end the threshold iteration process, and use the corresponding Tp+1 as the final gray-scale threshold;

[0101] Compare the gray-scale values corresponding to the pixel points in the corresponding part image with the obtained final gray-scale threshold Tp+1;

[0102] If Zw ≤ Tp + 1, then replace the gray value corresponding to the corresponding pixel point with '0';

[0103] If Zw > Tp + 1, then replace the gray value corresponding to the corresponding pixel point with '255';

[0104] Replace the pixel points in the left-view gray image according to the above process to obtain the corresponding binary image;

[0105] Taking the pixel point at the center of the binary image as the origin, construct a two-dimensional rectangular coordinate system. Then the coordinates of the pixel points in the corresponding binary image can be expressed as (x, y), and the gray value corresponding to the corresponding pixel point is expressed as f(x, y);

[0106] Based on the edge detection algorithm, determine the edge points of the binary image in the y direction;

[0107] If f(x, y) > [f(x - 1, y) + f(x + 1, y)] / 2, then the pixel point (x, y) is the edge point of the y-th column. Calculate all the pixel points in the binary image according to the above process to obtain the boundary points between the upper and lower edges of the corresponding machined part and the background, and mark them as edge pixel points;

[0108] Use the same method to detect the edge in the x direction of the binary image to obtain the edge pixel points between the left and right edges of the corresponding machined part and the background;

[0109] Count the obtained edge pixel points and obtain the part edge contour of the corresponding machined part based on them.

[0110] It should be further noted that in the specific implementation process, the process in which the data display module is used to measure according to the obtained part edge contour and output the corresponding measurement report includes:

[0111] Obtain the pixel point coordinates (m, n) corresponding to the inner edge pixel points in the corresponding part edge contour, where (m, n) ∈ (x, y);

[0112] Obtain the corresponding pixel dimensions of the machined part according to the obtained pixel point coordinates, the left-view key feature points and the right-view key feature points, and in combination with the acquisition parameters of the image acquisition terminal. The pixel dimensions include the pixel distance between the feature points, as well as the pixel perimeter and pixel area corresponding to the corresponding machined part;

[0113] Based on big data technology, obtain the pixel equivalent corresponding to the used image acquisition terminal. The pixel equivalent refers to the conversion coefficient between the pixel size of an object in the acquired image information and the actual object size under the same acquisition settings;

[0114] Convert the obtained pixel measurement data according to the obtained pixel equivalent to obtain the actual measurement data of the corresponding machined part;

[0115] Compare the obtained actual measurement data with the preset tolerance range; the preset tolerance range refers to the allowable error range relative to the standard machined part formulated by the staff based on historical work experience;

[0116] If the obtained actual measurement data does not meet the preset tolerance range, generate a non-conformance report, feedback to the staff, and at the same time screen and eliminate the corresponding machined part;

[0117] If the obtained actual measurement data meets the preset tolerance range, indicating that the corresponding machined part meets the requirements, then label the corresponding machined part, generate the corresponding measurement report, and store it in the measurement center for the staff to view; among them, the measurement data report includes the actual measurement data, pixel size, and corresponding image information.

[0118] As Figure 2 shown, a measurement method of a mechanical processing measurement system based on the Internet of Things includes the following steps:

[0119] Step 1: Collect data during the mechanical processing process to obtain the corresponding mechanical processing data, where the mechanical processing data includes vibration data and image data;

[0120] Step 2: Judge whether the machined part that has completed mechanical processing meets the requirements according to the collected vibration data. If it does not meet the requirements, give feedback. If it meets the requirements, perform data preprocessing on the image data of the corresponding machined part to obtain the corresponding visual image data;

[0121] Step 3: Extract features from the obtained visual image data to obtain the corresponding visual feature points, and at the same time extract the contour from the obtained visual image data to obtain the corresponding part edge contour;

[0122] Step 4: Convert the obtained visual feature points and part contour feature data to obtain the actual measurement data of the corresponding machined part, and judge whether the corresponding machined part meets the requirements based on the obtained actual measurement data. If it meets the requirements, generate the corresponding measurement report.

[0123] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An Internet of Things-based machining measurement system, comprising a measurement center, characterized in that, The measurement center includes a data acquisition module, a data preprocessing module, a data processing module, and a data display module; The data acquisition module is used to acquire data during the machining process to obtain corresponding machining data, and the machining data includes vibration data and image data; The data preprocessing module is used to determine whether the machined parts that have completed machining meet the requirements based on the acquired vibration data. If they do not meet the requirements, feedback is provided. If they meet the requirements, the image data of the corresponding machined parts is preprocessed to obtain corresponding visual image data; The data processing module is used to process the obtained visual image data to obtain corresponding part images, and perform contour extraction on the obtained part images to obtain corresponding part edge contours; The data display module is used to generate corresponding measurement reports based on the obtained visual feature points and part edge contours.

2. The machining measurement system based on the Internet of Things according to claim 1, characterized in that, The data acquisition module consists of several Internet of Things terminals and a measurement database. The Internet of Things terminals include vibration acquisition terminals and image acquisition terminals, and the machining data includes vibration data and image data; The vibration acquisition terminal is used to acquire the number of vibrations of the cutting tool during the machining process of the machined parts to obtain corresponding vibration data; the image acquisition terminal is used to acquire images of the machined parts that have completed machining to obtain corresponding image data, and upload the acquired machining data to the measurement database for storage.

3. The mechanical processing measurement system based on the Internet of Things according to claim 2, characterized in that, The process by which the data preprocessing module determines whether the machined parts that have completed machining meet the requirements based on the acquired vibration data includes: Set a vibration threshold range, read the vibration data in the acquired machining data, and compare the acquired vibration data with the corresponding vibration threshold; If the vibration data corresponding to the corresponding machined part is outside the vibration threshold range, mark the corresponding machined part as a defective part and provide feedback to the measurement center; If the vibration data corresponding to the corresponding machined part is within the vibration threshold range, retain the image data corresponding to the corresponding machined part and preprocess the retained image data.

4. An Internet of Things-based machining measurement system according to claim 3, wherein The process of preprocessing the retained image data includes: Count the retained image data, and the image data includes left-view images and right-view images; Obtain all the pixel points in the left-view image and the values corresponding to the RGB color components of the corresponding pixel points; Taking the upper-left pixel point of the left-view image as the origin, construct a corresponding two-dimensional rectangular coordinate system; Obtain the RGB components corresponding to the pixel points in the left-view image, obtain the average value corresponding to the corresponding RGB components, perform mean summation, and use it as the gray value of the corresponding pixel point; Traverse each pixel point in the left-view image, use the same method to obtain the gray value of the corresponding pixel point, and then obtain the corresponding gray image based on the obtained gray value; Construct a Gaussian template window based on Gaussian filtering technology, traverse all pixel points in the grayscale image based on the Gaussian template window, and perform a convolution operation on the template coefficients and the corresponding pixel points under the Gaussian template window; and use the result of the convolution operation to replace the corresponding grayscale value at the center position of the Gaussian template window, thereby obtaining the corresponding filtered grayscale image, and mark the grayscale image after filtering as the left-view grayscale image; Process the right-view image using the same method to obtain the corresponding right-view grayscale image; Summarize the left-view grayscale image and the right-view grayscale image of the corresponding processed part to obtain relevant visual image data.

5. The mechanical processing measurement system based on the Internet of Things according to claim 4, wherein The process of the data processing module processing the obtained visual image data to obtain the corresponding part image includes: Read the obtained visual image data, and extract feature points from the visual image data respectively based on the SIFT algorithm to obtain the corresponding visual feature points, and the visual feature points include left-view key feature points and right-view key feature points; For the extracted visual feature points, match the left-view key feature points and the right-view key feature points, and obtain the grayscale deviation value between the matched left-view key feature points and right-view key feature points, and then obtain the corresponding disparity value based on the obtained grayscale deviation value; Based on the obtained disparity value, adjust the position of the right key feature points in the right-view grayscale image to align it with the left-view grayscale image to obtain the corresponding right visual compensation map; Integrate the right visual compensation image with the left-view grayscale image to obtain the final part image.

6. The mechanical processing measurement system based on the Internet of Things according to claim 5, characterized in that, The process of extracting the contour of the obtained part image to obtain the corresponding part edge contour includes: Read the obtained part image, and obtain the grayscale value Zw corresponding to all pixel points in the corresponding part image, and set the initial grayscale threshold T0; Compare the grayscale value corresponding to all pixel points in the part image with the initial grayscale threshold; If Zw≤T0, mark the corresponding pixel point as a low-grayscale point; If Zw>T0, mark the corresponding pixel point as a high-grayscale point; Read the grayscale value at the pixel point marked as a low-grayscale point, and obtain the corresponding average value TA, and use the same method to obtain the average value TB of the corresponding high-grayscale points; Obtain the second grayscale threshold T1 based on the obtained TA and TB, use T1 to replace the set initial grayscale threshold T0, and repeat the above process for threshold iteration until Tp+1 = Tp; end the threshold iteration process, and use the corresponding Tp+1 as the final grayscale threshold; Convert the corresponding part image into a binary image based on the obtained final grayscale threshold; Taking the pixel point at the center of the binary image as the origin, construct a two-dimensional rectangular coordinate system; Based on the edge detection algorithm, determine the edge points of the binary image in the y direction to obtain the boundary points between the upper and lower edges of the corresponding processed part and the background, and mark them as edge pixels; Use the same method to detect the edges in the x direction of the binary image to obtain the edge pixels between the left and right edges of the corresponding processed part and the background; Count the obtained edge pixels, and obtain the part edge contour of the corresponding processed part based on them.

7. A mechanical processing measurement system based on the Internet of Things according to claim 6, characterized in that, The process of the data display module generating a corresponding measurement report based on the obtained part edge contour includes: Obtaining the pixel coordinates corresponding to the inner edge pixel points of the corresponding part edge contour; Obtaining the pixel size of the corresponding machined part based on the obtained pixel coordinates, the left-view key feature points, the right-view key feature points, and in combination with the acquisition parameters of the image acquisition terminal; Obtaining the pixel equivalent corresponding to the used image acquisition terminal based on big data technology; Converting the obtained pixel measurement data according to the obtained pixel equivalent to obtain the actual measurement data of the corresponding machined part; Comparing the obtained actual measurement data with the preset tolerance range. If the obtained actual measurement data does not meet the preset tolerance range, an unqualified report is generated and feedback is given to the staff. At the same time, the corresponding machined parts are screened and removed; If the obtained actual measurement data meets the preset tolerance range, indicating that the corresponding machined part meets the requirements, the corresponding machined part is labeled, a corresponding measurement report is generated, and it is stored in the measurement center.

8. A method for machining measurement based on the Internet of Things according to any one of claims 1 to 7, characterized in that, It includes the following steps: Step 1: Collect data during the machining process to obtain the corresponding machining data, where the machining data includes vibration data and image data; Step 2: Determine whether the machined part that has completed machining meets the requirements based on the collected vibration data. If it does not meet the requirements, feedback is given. If it meets the requirements, the image data of the corresponding machined part is preprocessed to obtain the corresponding visual image data; Step 3: Extract features from the obtained visual image data to obtain the corresponding visual feature points. At the same time, extract the contour from the obtained visual image data to obtain the corresponding part edge contour; Step 4: Convert the obtained visual feature points and part contour feature data to obtain the actual measurement data of the corresponding machined part, and determine whether the corresponding machined part meets the requirements based on the obtained actual measurement data. If it meets the requirements, a corresponding measurement report is generated.

Images