Method, device and system for identifying continuous lines of ceramic tiles

Abstract

The invention, which relates to the technical field of tile detection, discloses a tile connection line identification method, device and system, the method comprising: acquiring a surface texture image of a target tile; the surface texture image is preprocessed; comparing the preprocessed surface texture image with a preset template image so as to calculate the offset between the preprocessed surface texture image and the preset template image; and according to a preset calibration parameter, an offset threshold and the offset, judging whether the target ceramic tile meets a standard, and if the target ceramic tile does not meet the standard, driving calibration equipment to mark the target ceramic tile. By adopting the method and the device, the target ceramic tile can be shot to obtain the high-resolution surface texture image of the surface of the target ceramic tile, and the high-precision recognition of the ceramic tile lines is realized in combination with an accurate ceramic tile line recognition method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ceramic tile detection, and in particular to a ceramic tile continuous line identification method, device and system. BACKGROUND

[0002] With the continuous development of modern ceramic tile technology, the requirements for texture of ceramic tiles are becoming higher and higher. For example, the process of infinite continuous lines and parquet requires multiple ceramic tiles to be spliced into a complete pattern, which requires the texture of ceramic tiles to have high stability and consistency.

[0003] Currently, ceramic factories generally use inkjet printing to print ceramic tile texture. During the printing process, slight deviation in the feeding of ceramic tiles will cause deviation in texture printing, thereby failing to achieve accurate alignment. Since ceramic tiles are produced one by one on the assembly line, workers cannot detect the deviation of ceramic tile texture by naked eye or other measurement methods. Therefore, professional detection equipment is needed for accurate detection.

[0004] However, there is a lack of high-precision ceramic tile texture detection equipment on the market, and existing detection methods mostly rely on manual visual inspection, which is inefficient and prone to errors.

[0005] Therefore, it is of great practical significance to develop a device that can accurately detect the deviation of ceramic tile texture. SUMMARY

[0006] The technical problem to be solved by the present application is to provide a ceramic tile continuous line identification method, device and system, which can accurately detect the deviation of ceramic tile texture and automatically mark ceramic tiles that do not meet the standard.

[0007] To solve the above technical problems, the present application provides a ceramic tile continuous line identification method, comprising: acquiring a surface texture image of a target ceramic tile; pre-processing the surface texture image; comparing the pre-processed surface texture image with a preset template image to calculate the offset between the pre-processed surface texture image and the preset template image; determining whether the target ceramic tile meets the standard according to the preset calibration parameters, the offset threshold and the offset; and if it does not meet the standard, driving a calibration device to mark the target ceramic tile.

[0008] As an improvement of the above-mentioned scheme, the step of pre-processing the surface texture image comprises: positioning processing the surface texture image to correct the real-time position of the surface texture image; gray-scale processing the surface texture image after positioning processing; and filtering the surface texture image after gray-scale processing.

[0009] As an improvement of the above scheme, the step of positioning the surface texture image to correct the real-time position of the surface texture image comprises: performing histogram equalization processing on the surface texture image; performing binaryzation processing on the surface texture image after histogram equalization processing; performing edge detection on the surface texture image after binaryzation processing to extract edge lines; detecting reference straight lines in the edge lines through Hough transform; filtering reference straight lines in the interior of the edge lines; identifying position information of the target tile according to the filtered reference straight lines; and adjusting the real-time position of the surface texture image according to the position information to make the real-time position of the surface texture image consistent with the reference position.

[0010] As an improvement of the above scheme, the step of comparing the preprocessed surface texture image with the preset template image to calculate the offset between the preprocessed surface texture image and the preset template image comprises: obtaining an ROI region of the preprocessed surface texture image; extracting reference key points in the ROI region; filtering the reference key points to extract target key points; comparing the target key points with preset key points of the preset template image to calculate the offset between the target key points and the preset key points; and calculating the average of all offsets to generate a final offset.

[0011] As an improvement of the above scheme, the step of comparing the preprocessed surface texture image with the preset template image to calculate the offset between the preprocessed surface texture image and the preset template image further comprises: labeling target key points and offsets on the surface texture image.

[0012] As an improvement of the above scheme, the step of judging whether the target tile meets the standard according to the preset calibration parameter, the offset threshold and the offset comprises: calculating an actual deviation size according to the preset calibration parameter and the offset; comparing the actual deviation size with the preset offset threshold to judge whether the actual deviation size is greater than the offset threshold, and if yes, it indicates that the target tile does not meet the standard, and if no, it indicates that the target tile meets the standard.

[0013] As an improvement of the above scheme, the generation method of the calibration parameter comprises: obtaining a calibration image of a calibration board collected by a camera, the calibration board being provided with a mark segment with a known actual size; measuring a reference size of the mark segment in the calibration image; calculating a pixel actual size corresponding to each pixel point in the calibration image according to the actual size and the reference size; and taking the pixel actual size as the calibration parameter.

[0014] As an improvement of the above scheme, the offset comprises a horizontal offset, a vertical offset and a rotation offset.

[0015] Correspondingly, the application also provides a ceramic tile joint line identification device, comprising a memory and a processor, the memory stores a computer program, wherein the processor implements the steps of the above-mentioned ceramic tile joint line identification method when executing the computer program.

[0016] Correspondingly, the application also provides a ceramic tile joint line identification system, comprising: a camera for collecting a surface texture image of a target ceramic tile and a calibration image of a calibration plate; a calibration device for marking the target ceramic tile; and the above-mentioned ceramic tile joint line identification device.

[0017] The application has the following beneficial effects:

[0018] The application determines the relative offset between the surface texture image and the preset template image by comparing the best matching position between the surface texture image and the preset template image and by calculating the similarity or difference between the surface texture image and the preset template image, and automatically marks the ceramic tile that does not meet the standard, thereby filling the gap in the prior art. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is an embodiment flowchart of the ceramic tile joint line identification method of the application;

[0020] Figure 2 is an embodiment structural schematic diagram of the ceramic tile joint line identification system of the application. DETAILED DESCRIPTION

[0021] To make the object, technical scheme and advantages of the application clearer, the application will be further described in detail below with reference to the drawings.

[0022] Referring to Figure 1 , Figure 1 shows an embodiment flowchart of the ceramic tile joint line identification method of the application, which comprises:

[0023] S101, obtaining a surface texture image of a target ceramic tile;

[0024] When the ceramic tile to be detected passes through the detection device, the surface texture image of the target ceramic tile is captured in real time by a high-resolution camera.

[0025] S102, pre-processing the surface texture image;

[0026] Correspondingly, the step of pre-processing the surface texture image comprises:

[0027] (1) performing positioning processing on the surface texture image to correct the real-time position of the surface texture image;

[0028] (2) performing grayscale processing on the surface texture image after the positioning processing;

[0029] (3) filtering the surface texture image after the grayscale processing.

[0030] Therefore, by preprocessing the surface texture image, the influence of environmental light, camera noise and other factors on the image quality can be effectively eliminated, and the accuracy of subsequent comparison is ensured.

[0031] Further, the step of positioning the surface texture image to correct the real-time position of the surface texture image comprises:

[0032] (1.1) histogram equalization processing of the surface texture image;

[0033] (1.2) binaryzation processing of the surface texture image after the histogram equalization processing;

[0034] (1.3) edge detection of the surface texture image after the binaryzation processing to extract edge lines;

[0035] (1.4) detecting reference straight lines in the edge lines by Hough transform;

[0036] (1.5) filtering the reference straight lines in the edge lines that are inside;

[0037] (1.6) identifying the position information of the target tile according to the filtered reference straight lines;

[0038] (1.7) adjusting the real-time position of the surface texture image according to the position information, so that the real-time position of the surface texture image is consistent with the reference position.

[0039] S103, comparing the preprocessed surface texture image with a preset template image to calculate the offset between the preprocessed surface texture image and the preset template image;

[0040] It should be noted that a preset template image needs to be established before comparison. The preset template image is obtained by shooting a tile sample that meets the texture requirements by a high-resolution camera; the preset template image is stored in the industrial computer as a reference for subsequent detection.

[0041] Correspondingly, the step of comparing the preprocessed surface texture image with the preset template image to calculate the offset between the preprocessed surface texture image and the preset template image comprises:

[0042] (1) obtaining the ROI region of the preprocessed surface texture image;

[0043] (2) extracting reference key points in the ROI region;

[0044] (3) filtering the reference key points to extract target key points;

[0045] (4) Compare the target key points with the preset key points of the preset template image to calculate the offset between the target key points and the preset key points;

[0046] The offset includes horizontal offset, vertical offset, and rotational offset.

[0047] (5) Calculate the average of all offsets to generate the final offset.

[0048] Therefore, by comparing the best matching position between the surface texture image and the preset template image, and by calculating the similarity or difference between the surface texture image and the preset template image, the relative displacement between the surface texture image and the preset template image is determined.

[0049] Furthermore, target key points and offsets can be annotated on the surface texture image.

[0050] S104 determines whether the target tile meets the standard based on the preset calibration parameters, offset threshold and offset amount. If it does not meet the standard, the calibration device is driven to mark the target tile.

[0051] Accordingly, the steps for determining whether the target tile meets the standard based on preset calibration parameters, offset thresholds, and offset amounts include:

[0052] (1) Calculate the actual deviation size based on the preset calibration parameters and offset;

[0053] Furthermore, the actual deviation dimensions can be displayed on the monitor for user viewing; at the same time, the actual deviation dimensions can also be stored in the industrial control computer for subsequent data analysis and quality traceability.

[0054] (2) Compare the actual deviation size with the preset offset threshold to determine whether the actual deviation size is greater than the offset threshold.

[0055] Before the equipment is put into operation, users can preset the threshold for texture deviation (i.e., offset threshold) according to production requirements.

[0056] (3) If the judgment is yes, it means that the target tile does not meet the standard;

[0057] At this point, a signal can be sent to drive the calibration device to mark the target tile for subsequent processing.

[0058] (4) If the judgment is negative, it means that the target tile meets the standard.

[0059] It should be noted that for each type of offset (such as horizontal offset, vertical offset, and rotational offset), its corresponding pixel value needs to be calculated separately and converted into the actual deviation size; thus, the texture can be judged to meet the requirements based on the actual similarity or difference between the surface texture image and the preset template image.

[0060] To convert the offsets in the surface texture image into actual deviation dimensions, the actual size of each pixel needs to be calculated beforehand. This conversion relationship is obtained through a calibration process. Specifically, the methods for generating calibration parameters include:

[0061] (1) Obtain the calibration image of the calibration board captured by the camera. The calibration board has marker segments with known actual dimensions.

[0062] The calibration plate has clear scales or markings, and its actual dimensions are known.

[0063] (2) Measure the reference dimensions of the marked segments in the calibration image;

[0064] Reference dimensions are in pixels.

[0065] (3) Calculate the actual pixel size corresponding to each pixel in the calibration image based on the actual size and the reference size;

[0066] For example, if the actual length of the marker segment in the calibration plate is 100 mm, and the reference length of the marker segment in the calibration image is 1000 pixels, then the actual size of each pixel is 0.1 mm / pixel.

[0067] (4) Use the actual pixel size as the calibration parameter.

[0068] Therefore, the tile pattern recognition method of the present invention can accurately detect deviations in tile texture and automatically mark tiles that do not meet the standards, filling the gap in the prior art.

[0069] Accordingly, the present invention also provides a tile pattern recognition device, including a memory and a processor. The memory stores a computer program, wherein the processor executes the computer program to implement the steps of the above-described tile pattern recognition method.

[0070] See Figure 2 , Figure 2 The specific structure of the tile pattern recognition system of the present invention is shown, which includes:

[0071] A camera used to acquire surface texture images of the target ceramic tile and calibration images of the calibration plate;

[0072] Calibration equipment used to mark target tiles;

[0073] A tile pattern recognition device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the tile pattern recognition method described above.

[0074] Therefore, by cooperating with the camera, calibration equipment, and tile pattern recognition equipment, the target tile can be photographed to obtain a high-resolution surface texture image of the target tile surface. Combined with a precise tile pattern recognition method, high-precision recognition of tile patterns can be achieved.

[0075] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for identifying continuous patterns in ceramic tiles, characterized in that, include: Obtain the surface texture image of the target tile; The surface texture image is preprocessed; The preprocessed surface texture image is compared with the preset template image to calculate the offset between the preprocessed surface texture image and the preset template image; The system determines whether the target tile meets the standard based on the preset calibration parameters, offset threshold, and offset amount. If it does not meet the standard, the calibration device is driven to mark the target tile.

2. The method for identifying continuous tile patterns as described in claim 1, characterized in that, The preprocessing step of the surface texture image includes: The surface texture image is subjected to positioning processing to correct the real-time position of the surface texture image; The surface texture image after positioning processing is then converted to grayscale. The surface texture image after grayscale processing is then filtered.

3. The method for identifying continuous tile patterns as described in claim 2, characterized in that, The step of performing positioning processing on the surface texture image to correct the real-time position of the surface texture image includes: The surface texture image is subjected to histogram equalization processing; The surface texture image after histogram equalization is binarized. Edge detection is performed on the binarized surface texture image to extract edge lines; The reference straight line in the edge line is detected by Hough transform; Filter the reference straight lines that are inside the edge lines; The position information of the target tile is identified based on the filtered reference line. The real-time position of the surface texture image is adjusted according to the position information so that the real-time position of the surface texture image is consistent with the reference position.

4. The method for identifying continuous tile patterns as described in claim 1, characterized in that, The step of comparing the preprocessed surface texture image with a preset template image to calculate the offset between the preprocessed surface texture image and the preset template image includes: Obtain the ROI region of the preprocessed surface texture image; Extract benchmark key points within the ROI region; Filter the benchmark key points to extract the target key points; The target key points are compared with the preset key points of the preset template image to calculate the offset between the target key points and the preset key points; Calculate the average of all offsets to generate the final offset.

5. The method for identifying continuous tile patterns as described in claim 4, characterized in that, The step of comparing the preprocessed surface texture image with the preset template image to calculate the offset between the preprocessed surface texture image and the preset template image further includes: marking target key points and offsets on the surface texture image.

6. The method for identifying continuous tile patterns as described in claim 1, characterized in that, The step of determining whether the target tile meets the standard based on preset calibration parameters, offset threshold, and the offset amount includes: The actual deviation size is calculated based on the preset calibration parameters and the offset. The actual deviation size is compared with a preset offset threshold to determine whether the actual deviation size is greater than the offset threshold. If the judgment is yes, it means that the target tile does not meet the standard. If the result is negative, it means that the target tile meets the standard.

7. The method for identifying continuous tile patterns as described in claim 1, characterized in that, The method for generating the calibration parameters includes: Acquire a calibration image of a calibration board captured by a camera, wherein the calibration board has marker segments of known actual size; Measure the reference size of the marked segment in the calibration image; Based on the actual size and the reference size, calculate the actual pixel size corresponding to each pixel in the calibration image; The actual size of the pixel is used as the calibration parameter.

8. The method for identifying continuous tile patterns as described in claim 1, characterized in that, The offset includes lateral offset, longitudinal offset, and rotational offset.

9. A tile pattern recognition device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the tile pattern recognition method according to any one of claims 1 to 8.

10. A tile pattern recognition system, characterized in that, include: A camera used to acquire surface texture images of the target ceramic tile and calibration images of the calibration plate; A calibration device for marking the target tile; The tile pattern recognition device according to claim 9.

Images