Leather texture manufacturing method and system for watchband

Abstract

The invention relates to a leather texture manufacturing method and system for a watchband, and belongs to the technical field of leather texture manufacturing. The leather texture manufacturing method comprises the steps that a high-resolution image of leather surface texture is collected, the resolution and size of the image and position information of the texture are recorded, and the texture image is preprocessed; the texture image is etched to the surface of the mold through laser equipment, and the position deviation is compensated by collecting real-time image data in the manufacturing process of the mold; according to material selection of watchband leather, softening parameters are set to soften the leather, the leather is placed between a mold and pressing equipment, and a texture image is transferred to the surface of the leather by applying pressure and temperature. According to the invention, reference point setting and alignment are respectively carried out on the mold and the texture image, and the alignment deviation is obtained through image registration for compensation, so that precise manufacturing of the mold is completed.

Description

Technical Field

[0001] The invention belongs to the technical field of leather texture manufacturing, and in particular relates to a leather texture manufacturing method and system for a watch strap. Background Art

[0002] In the watch manufacturing industry, the texture and appearance quality of the watch strap are key factors in measuring the grade and attractiveness of the product. Natural leather is often used in the manufacture of watch straps for high-end watches due to its unique texture and comfortable feel. However, natural leather is prone to problems such as uneven texture and unstable texture during processing.

[0003] Therefore, the texture replication of artificial leather through molds has gradually developed to reduce costs and improve production efficiency. However, the traditional leather texture replication process often cannot accurately restore the tiny texture details of natural leather, resulting in the accuracy of the mold surface texture being lower than the actual leather texture, affecting the aesthetics and texture of the finished product; at the same time, during the texture replication process, it is difficult to accurately align the engraving position on the mold surface and the position of the leather texture image, which is prone to deviation, thus affecting the restoration of the leather texture of the strap. Summary of the invention

[0004] In order to solve the above-mentioned problems existing in the prior art, the present invention provides a leather texture manufacturing method and system for a watch strap, which completes the precise production of the mold by respectively setting the reference points of the mold and the texture image, and then compensating for the deviation of the alignment through image registration. The deviation acquisition combines the high efficiency of the algorithm feature extraction with the rotational invariance and robustness of the descriptor, which greatly improves the efficiency of the registration, and significantly increases the number of matching points, thereby meeting the needs of real-time feature point matching.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A first aspect of the present disclosure provides a method for manufacturing a leather texture for a watch strap, comprising the following steps: S1. Acquire leather texture image: collect high-resolution image of leather surface texture, record image resolution, size and texture position information, and pre-process the texture image; S2. Etching leather texture mold: Etching the texture image onto the mold surface through laser equipment, and compensating for position deviation by collecting real-time image data during the mold making process; S3. Production of leather texture for watch strap: According to the selection of leather for watch strap, softening parameters are set to soften the leather, the leather is placed between the mold and the pressing equipment, and the texture image is transferred to the leather surface by applying pressure and temperature.

[0006] Furthermore, the preprocessing of the texture image includes grayscale conversion, contrast adjustment and sharpening processing.

[0007] Furthermore, the etching leather texture mold comprises the following steps: S21, loading image data: loading the pre-processed leather texture image into the laser etching device, setting the etching depth, power and speed parameters on the device, and setting the etching area; S22, reference setting and positioning: setting reference points on the mold surface and in the texture image respectively, for etching the initial texture image and aligning the mold; S23, image capture and matching: feature point detection is performed through the real-time collected mold surface image, and feature point matching is completed using the SURF-ORB algorithm.

[0008] Furthermore, the leather texture etching mold further comprises the following steps: S24, offset calculation and correction: according to the image registration output result, calculate the deviation between the mold reference point and the target texture image, and adjust the moving path of the laser head, etching focus distance and laser power in real time; Among them, the adjustment parameters include: X / Y axis positioning, Z axis height adjustment, laser energy intensity; S25, dynamic verification and feedback: real-time detection and correction of deviation values ​​are performed during the etching process, so that the etched texture is always consistent with the target texture image.

[0009] Furthermore, the benchmark setting and positioning includes the following steps: S221, reference point setting: defining a set of fixed reference points on the mold, forming mold reference points by processing or marking, and then selecting positions corresponding to the mold reference points in the leather texture image to mark and form image reference points; S222, initial alignment: Use a camera or positioning probe to identify the mold reference point on the mold, compare the image reference point with the mold reference point, and adjust the beam position of the equipment so that the image reference point and the mold reference point completely overlap.

[0010] Furthermore, the image capturing and matching comprises the following steps: S231, feature point detection: calculate the determinant of the Hessian matrix, detect feature points in the mold surface image and texture image, set a threshold to filter out significant key points; Apply the FAST algorithm to the mold surface image and texture image to quickly detect corner points and assign the main direction to each feature point; S232, descriptor generation: construct SURF descriptors around each feature point to generate 64-dimensional descriptors to capture local image features; generate binary descriptors of mold surface images and texture images through the BRIEF algorithm; S233, benchmark point matching: Use SURF descriptor to calculate the Euclidean distance between the descriptors of the mold surface image and the texture image, and select the descriptor pair with the smallest distance as the preliminary match; use ORB descriptor to calculate the Hamming distance between binary descriptors to further screen matching points.

[0011] Furthermore, the image capturing and matching further comprises the steps of: S234, removing false matches: for each matching point, calculating the ratio of the minimum distance to the second minimum distance, if the ratio is less than a set threshold, retaining the match; Randomly select matching point pairs, estimate the transformation matrix, calculate the number of inliers, repeat this process to find the best transformation model, and remove outliers; S235, image registration output: according to the obtained transformation matrix from the mold surface image to the texture image, the mold surface image is transformed to obtain an image aligned with the texture image, and the registered mold surface image is output.

[0012] Furthermore, the production of the leather texture of the watch strap further includes the steps of: After pressing, the material is cooled to stabilize the texture and then post-processed, including coating, dyeing and polishing.

[0013] A second aspect of the present disclosure provides a leather texture manufacturing system for a watch strap, which is used to implement a leather texture manufacturing method for a watch strap as described above, including a mold manufacturing module, a texture manufacturing module, and a quality inspection module; The mold manufacturing module is used to etch the texture image onto the surface of the mold by laser equipment to complete the mold manufacturing. During the manufacturing process, the parameters of the laser equipment are controlled by a computer program, and the etching process is monitored in real time, including: The reference points in the mold surface and texture image are set respectively to complete the alignment of the initial texture image and the mold; feature point detection is performed through the real-time acquired mold surface image, and feature point matching is completed to obtain image registration output to achieve deviation calibration.

[0014] As a preferred technical solution of the present invention, the texture manufacturing module is used to transfer the texture image to the strap leather through the mold, and the leather material is softened before the transfer, and then the mold etching pattern is pressed onto the leather surface through a hot pressing process; The quality inspection module is used to inspect the leather texture by machine vision and screen out defective watch straps, wherein inspecting the leather texture by machine vision includes the following steps: The texture image of the leather surface is collected by a scanner, and the texture area is separated from the background by threshold segmentation after preprocessing the texture image; Extract geometric features, statistical features, and frequency domain features from texture images, including the length, width, and spacing of texture lines; extract texture contrast and directionality through gray-level co-occurrence matrices; and use Fourier transforms to analyze texture periodicity and directionality. Compare the actual texture image with the standard texture image to identify deviation areas; use template matching algorithms to determine whether the texture deviates from the design standard, including detecting scratches, stains, holes or irregular textures in the texture; use edge break detection algorithms to identify discontinuous texture areas; check texture density to discover blank or over-compressed areas; Defective texture leather is screened based on the identification deviation.

[0015] The beneficial effects of the present invention are: The present invention first pre-processes the leather texture image and then loads it into the laser etching equipment, then respectively sets the reference points for the mold and the texture image, and then obtains the alignment deviation through image registration to compensate for it, so as to accurately manufacture the mold, and finally transfers the texture image to the watch strap leather through the manufactured mold, and completes the finished product screening by defect identification through machine vision; in the deviation compensation, the present invention adopts the SURF-ORB algorithm, which combines the high efficiency of the algorithm feature extraction with the rotation invariance and robustness of the descriptor, greatly improves the efficiency of the registration, and significantly increases the number of matching points, thereby meeting the needs of real-time feature point matching and completing the deviation compensation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0017] Figure 1 A schematic flow chart of a method for manufacturing leather texture for a watch strap provided by an embodiment of the present invention; Figure 2 A schematic structural diagram of a leather texture manufacturing system for a watch strap provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0019] This embodiment provides a method for manufacturing leather texture for a watch strap, such as Figure 1 As shown, the following steps are included: S1. Acquire leather texture image: collect high-resolution images of leather surface texture, record image resolution, size and texture location information, and pre-process the texture image, including grayscale conversion, contrast adjustment and sharpening.

[0020] It is understandable that during the preprocessing process, the color image is converted into a grayscale image in order to highlight the light and dark contrast of the texture and simplify the calculation complexity of subsequent engraving. Specifically, the weighted average method is used to convert the RGB color components into grayscale values; the details of the texture are enhanced by contrast adjustment, and the histogram equalization technology is applied to make the light and dark contrast of the concave and convex parts clearer; finally, the clarity of the texture edge is further highlighted through sharpening processing, and image filtering algorithms such as Laplacian filtering, Unsharp Mask, etc. are used to customize sharpening parameters to balance the relationship between texture clarity and noise, so as to be suitable for precise processing of molds.

[0021] S2. Etching leather texture mold: Etching the texture image onto the mold surface by laser equipment, and compensating for the position deviation by collecting real-time image data during the mold manufacturing process, including the following steps: S21. Loading image data: Loading the pre-processed leather texture image into the laser etching device, setting the etching depth, power and speed parameters on the device, and setting the etching area to ensure that the operation is performed only in the specified mold surface area.

[0022] S22, reference setting and positioning: setting reference points on the mold surface and in the texture image respectively, for etching the initial texture image and aligning the mold, including the following steps: S221, reference point setting: define a set of fixed reference points (usually specific marks at the corners or center positions) on the mold, form the mold reference points by processing or marking, and then select the positions corresponding to the mold reference points in the leather texture image to mark and form image reference points.

[0023] It is understandable that the mold processing or marking method is such as small holes, engraved lines or optical marking points, and the reference points after image annotation are usually marked with obvious colors or symbols (such as crosses, dots); by marking the reference points of both, the position of the reference points in the image is accurately aligned with the actual reference points on the mold.

[0024] S222, initial alignment: Use a camera or positioning probe to identify the mold reference point on the mold, compare the image reference point with the mold reference point, and adjust the beam position of the equipment so that the image reference point and the mold reference point completely overlap.

[0025] It should be noted that after the initial alignment, due to the irregularities of the mold surface, material properties or texture complexity, and deviations in the laser equipment, real-time feature point matching is required to further achieve higher etching accuracy and stability.

[0026] S23, image capture and matching: feature point detection is performed through the real-time collected mold surface image, and feature point matching is completed using the SURF-ORB algorithm, including the following steps: S231, feature point detection: calculate the determinant of the Hessian matrix, detect the feature points in the mold surface image and texture image, set the threshold to filter out significant key points, and ensure that the extracted feature points have high stability; The FAST algorithm is applied to the mold surface image and texture image to quickly detect corner points and assign a main direction to each feature point to ensure the rotation robustness of the descriptor.

[0027] S232, descriptor generation: construct a SURF descriptor around each feature point and generate a 64-dimensional descriptor to capture local image features; generate binary descriptors of the mold surface image and texture image through the BRIEF algorithm to ensure that the descriptor is rotationally invariant.

[0028] S233, benchmark point matching: Use SURF descriptor to calculate the Euclidean distance between the descriptors of the mold surface image and the texture image, and select the descriptor pair with the smallest distance as the preliminary match; use ORB descriptor to calculate the Hamming distance between binary descriptors to further screen matching points.

[0029] S234, removing false matches: for each matching point, calculating the ratio of the minimum distance to the second minimum distance, if the ratio is less than a set threshold, retaining the match; Randomly select matching point pairs, estimate the transformation matrix (such as homography matrix), calculate the number of internal points, and repeat this process to find the best transformation model and remove outliers.

[0030] S235, image registration output: according to the obtained transformation matrix from the mold surface image to the texture image, the mold surface image is transformed to obtain an image aligned with the texture image, and the registered mold surface image is output.

[0031] It should be noted that in this embodiment, the SURF-ORB algorithm is used, which combines the high efficiency of SURF (SpeededUp Robust Features) feature extraction with the rotation invariance and robustness of ORB (Oriented FAST and Rotated BRIEF) descriptors. By detecting feature points and generating descriptors through SURF, and then using the ORB algorithm for fast matching, the efficiency of registration is greatly improved, and the number of matching points is significantly increased, thereby meeting the needs of real-time feature point matching and completing deviation compensation.

[0032] S24, offset calculation and correction: according to the image registration output results, calculate the deviation between the mold reference point and the target texture image, and adjust the moving path of the laser head, etching focus distance and laser power in real time.

[0033] Among them, the adjustment parameters include: X / Y axis positioning, Z axis height adjustment, and laser energy intensity.

[0034] S25, dynamic verification and feedback: real-time detection and correction of deviation values ​​are performed during the etching process, so that the etched texture is always consistent with the target texture image.

[0035] It should be noted that the deviations are compensated through image registration, which ensures the accuracy and stability of mold etching and lays the foundation for the subsequent production of leather texture of the watch strap.

[0036] S3. Production of leather texture for watch strap: according to the selection of leather for watch strap, softening parameters are set to soften the leather, and the leather is placed between the mold and the pressing equipment. By applying pressure and temperature, the mold transfers the texture image to the leather surface; After pressing, the material is cooled to stabilize the texture structure and then post-processed, such as coating, dyeing, polishing, etc., to further enhance the visual effect and durability.

[0037] It is understandable that the texture image is copied onto the leather using a heat pressing process, and this pressure can deform the surface layer of the material to fit closely to the texture structure of the mold, so that the leather surface forms a fine texture consistent with the mold.

[0038] This embodiment also provides a leather texture manufacturing system for a watch strap, such as Figure 2 As shown, it includes a mold manufacturing module, a texture manufacturing module and a quality inspection module.

[0039] The mold manufacturing module is used to etch the texture image onto the surface of the mold by laser equipment to complete the mold manufacturing. During the manufacturing process, the parameters of the laser equipment are controlled by a computer program, and the etching process is monitored in real time, including: The reference points in the mold surface and texture image are set respectively to complete the alignment of the initial texture image and the mold; feature point detection is performed through the real-time acquired mold surface image, and feature point matching is completed to obtain image registration output to achieve deviation calibration.

[0040] The texture manufacturing module is used to transfer the texture image to the watch strap leather through a mold. Before the transfer, the leather material is softened, and then the mold etching pattern is pressed onto the leather surface through a hot pressing process.

[0041] The quality inspection module is used to inspect the leather texture by machine vision and screen out defective watch straps, wherein inspecting the leather texture by machine vision includes the following steps: The texture image of the leather surface is collected by a scanner, and the texture area is separated from the background by threshold segmentation after preprocessing the texture image; Extract geometric, statistical, and frequency domain features from texture images, including the length, width, and spacing of texture lines; extract texture contrast and directionality through the gray-level co-occurrence matrix (GLCM); and use Fourier transform to analyze the periodicity and directionality of textures. Compare the actual texture image with the standard texture image to identify deviation areas; use template matching algorithms to determine whether the texture deviates from the design standard, including detecting scratches, stains, holes or irregular textures in the texture; use edge break detection algorithms to identify discontinuous texture areas; check texture density to discover blank or over-compressed areas; Defective texture leather is screened based on the identification deviation.

[0042] The present invention first pre-processes the leather texture image and then loads it into the laser etching equipment, then respectively sets the reference points for the mold and the texture image, and then obtains the alignment deviation through image registration to compensate for it, so as to accurately manufacture the mold, and finally transfers the texture image to the watch strap leather through the manufactured mold, and completes the finished product screening by defect identification through machine vision; in the deviation compensation, the present invention adopts the SURF-ORB algorithm, which combines the high efficiency of the algorithm feature extraction with the rotation invariance and robustness of the descriptor, greatly improves the efficiency of the registration, and significantly increases the number of matching points, thereby meeting the needs of real-time feature point matching and completing the deviation compensation.

[0043] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A method for manufacturing leather texture for a watch strap, characterized in that: The following steps are involved: Obtain leather texture image: collect high-resolution images of leather surface texture, record image resolution, size and texture location information, and pre-process the texture image; Etching leather texture mold: The texture image is etched onto the mold surface by laser equipment, and the position deviation is compensated by collecting real-time image data during the mold making process; Watch strap leather texture production: According to the selection of watch strap leather, set the softening parameters to soften the leather, place the leather between the mold and the pressing equipment, and transfer the texture image to the leather surface by applying pressure and temperature.

2. The method for manufacturing leather texture for watch strap according to claim 1, characterized in that: The preprocessing of the texture image includes grayscale conversion, contrast adjustment and sharpening.

3. The method for manufacturing leather texture for watch strap according to claim 1, characterized in that: The leather texture etching mold comprises the following steps: Loading image data: Load the pre-processed leather texture image into the laser etching device, set the etching depth, power and speed parameters on the device, and set the etching area; Benchmark setting and positioning: setting the benchmark points on the mold surface and in the texture image respectively, for etching the initial texture image and aligning the mold; Image capture and matching: Feature point detection is performed through real-time acquisition of mold surface images, and feature point matching is completed using the SURF-ORB algorithm.

4. The method for manufacturing leather texture for watch strap according to claim 3, characterized in that: The leather texture etching mold further comprises the following steps: Offset calculation and correction: Based on the image registration output results, calculate the deviation between the mold reference point and the target texture image, and adjust the laser head movement path, etching focus distance and laser power in real time; Among them, the adjustment parameters include: X / Y axis positioning, Z axis height adjustment, laser energy intensity; Dynamic verification and feedback: During the etching process, the deviation value is detected in real time for correction, so that the etched texture is always consistent with the target texture image.

5. The method for manufacturing leather texture for watch strap according to claim 3, characterized in that: The benchmark setting and positioning includes the following steps: Reference point setting: define a set of fixed reference points on the mold, form the mold reference points by processing or marking, and then select the positions corresponding to the mold reference points in the leather texture image to mark and form the image reference points; Initial alignment: Use a camera or positioning probe to identify the mold reference point on the mold, compare the image reference point with the mold reference point, and adjust the equipment's beam position so that the image reference point completely overlaps with the mold reference point.

6. The method for manufacturing leather texture for watch strap according to claim 3, characterized in that: The image capturing and matching comprises the following steps: Feature point detection: calculate the determinant of the Hessian matrix, detect feature points in the mold surface image and texture image, and set the threshold to filter out significant key points; Apply the FAST algorithm to the mold surface image and texture image to quickly detect corner points and assign the main direction to each feature point; Descriptor generation: Build SURF descriptors around each feature point to generate 64-dimensional descriptors to capture local image features; generate binary descriptors of mold surface images and texture images through the BRIEF algorithm; Benchmark point matching: Use the SURF descriptor to calculate the Euclidean distance between the descriptors of the mold surface image and the texture image, and select the descriptor pair with the smallest distance as the initial match; use the ORB descriptor to calculate the Hamming distance between binary descriptors to further screen the matching points.

7. The method for manufacturing leather texture for watch strap according to claim 6, characterized in that: The image capturing and matching further comprises the steps of: Remove false matches: For each matching point, calculate the ratio of the minimum distance to the second minimum distance. If the ratio is less than the set threshold, the match is retained. Randomly select matching point pairs, estimate the transformation matrix, calculate the number of inliers, repeat this process to find the best transformation model, and remove outliers; Image registration output: According to the obtained transformation matrix from the mold surface image to the texture image, the mold surface image is transformed to obtain an image aligned with the texture image, and the registered mold surface image is output.

8. The method for manufacturing leather texture for watch strap according to claim 1, characterized in that: The strap leather texture production also includes the steps of: After pressing, the material is cooled to stabilize the texture and then post-processed, including coating, dyeing and polishing.

9. A leather texture manufacturing system for a watch strap, used to implement a leather texture manufacturing method for a watch strap as claimed in any one of claims 1 to 8, characterized in that: Including mold manufacturing module, texture manufacturing module and quality inspection module; The mold manufacturing module is used to etch the texture image onto the surface of the mold by laser equipment to complete the mold manufacturing. During the manufacturing process, the parameters of the laser equipment are controlled by a computer program, and the etching process is monitored in real time, including: The reference points in the mold surface and texture image are set respectively to complete the alignment of the initial texture image and the mold; feature point detection is performed through the real-time acquired mold surface image, and feature point matching is completed to obtain image registration output to achieve deviation calibration.

10. The leather texture manufacturing system for a watch strap according to claim 9, characterized in that: The texture manufacturing module is used to transfer the texture image to the watch strap leather through the mold, soften the leather material before the transfer, and then press the mold etching pattern onto the leather surface through a hot pressing process; The quality inspection module is used to inspect the leather texture by machine vision and screen out defective watch straps, wherein inspecting the leather texture by machine vision includes the following steps: The texture image of the leather surface is collected by a scanner, and the texture area is separated from the background by threshold segmentation after preprocessing the texture image; Extract geometric features, statistical features, and frequency domain features from texture images, including the length, width, and spacing of texture lines; extract texture contrast and directionality through gray-level co-occurrence matrices; and use Fourier transforms to analyze texture periodicity and directionality. Compare the actual texture image with the standard texture image to identify deviation areas; use template matching algorithms to determine whether the texture deviates from the design standard, including detecting scratches, stains, holes or irregular textures in the texture; use edge break detection algorithms to identify discontinuous texture areas; check texture density to discover blank or over-compressed areas; Defective texture leather is screened based on the identification deviation.

Images