A keyboard leather cover production quality inspection method

Abstract

The present invention belongs to the field of detection, and provides a keyboard leather cover production quality detection method, which includes obtaining a standard image of the keyboard leather cover; determining a first layout according to the standard image; comparing a real-time image of the keyboard leather cover in production with the standard image to identify the position deviation of the key layout and the leather cover fold; based on a preset error range, if the key layout or the leather cover fold position deviation is within the error range of the standard layout, the layout is considered qualified, and if it exceeds the preset error range, a calibration prompt is given; according to the first layout, the expected folds and key parts in the real-time image are shielded to obtain a second image, and the surface texture detection is performed on the second image; when the texture detection result is greater than a preset threshold, a calibration prompt is given. The above scheme can improve the accuracy of keyboard leather cover production quality detection.

Description

Technical Field

[0001] The invention belongs to the field of detection, and in particular relates to a keyboard leather cover production quality detection method. Background Art

[0002] With the widespread use of tablet computers in daily office, study and entertainment, more and more users choose to equip their tablet computers with keyboard leather cases. This type of keyboard leather case not only provides protection to prevent the tablet computer from being scratched, squeezed, dropped or damaged during use or carrying, but also provides physical keyboard input function when unfolded, improving the user's typing efficiency and experience. Due to the widespread demand for keyboard leather cases, a large number of manufacturers have emerged in the market, and keyboard leather cases of various materials, designs and structures have emerged one after another.

[0003] At present, the production process of keyboard leather cases is constantly improving, but there are some unavoidable quality problems in the production process. For example, the joints between the keyboard keys and the leather case are prone to position deviation, which affects the keyboard experience; the surface of the leather case may have appearance defects such as creases, scratches, indentations, bubbles, etc. due to material processing or subsequent treatment. Therefore, how to effectively detect the production quality of keyboard leather cases, especially the accuracy of the combination of the keyboard and the leather case, the appearance quality of the leather case surface, and the functional performance of the keyboard keys, has become a technical problem that needs to be solved in the production process.

[0004] Existing detection methods mostly use manual inspection or simple image detection technology for quality control. However, manual inspection is inefficient, with large subjective errors, and it is difficult to ensure the consistency and accuracy of detection; although traditional image detection technology can partially replace manual inspection, it has limited detection effect on subtle appearance defects (such as creases and scratches). In particular, the surface creases of the keyboard leather cover itself are part of its design and material characteristics, not defects, but during the detection process, due to the irregularity of the position and shape of these creases, traditional image processing algorithms are difficult to accurately judge the difference between the normal state and abnormal deformation of the creases. Similarly, the text and logos on the keyboard keys will produce irregular optical effects in image detection due to changes in shape or light reflection, further increasing the probability of false detection. Summary of the invention

[0005] In order to solve the problems in the prior art, the present invention provides a keyboard leather cover production quality inspection method, the method comprising the following steps:

[0006] Acquire a standard image of the keyboard leather cover, wherein the standard image includes a keyboard key portion and a leather cover portion, wherein the keyboard key portion and the leather cover portion are integrally formed and are pre-stored in the system as a reference image;

[0007] Determine a first layout according to the standard image, the first layout being a standard layout of the keyboard leather cover, the layout comprising a key layout and a leather cover fold position, the key layout defines a key arrangement of the keyboard, and the leather cover fold position is an expected fold area;

[0008] Compare the real-time image of the keyboard leather cover in production with the standard image to identify the position deviation of the key layout and the fold of the leather cover; based on the preset error range, if the position deviation of the key layout or the fold of the leather cover is within the error range of the standard layout, the layout is considered qualified, and if it exceeds the preset error range, a calibration prompt is given;

[0009] According to the first layout, the expected folds and button parts in the real-time image are shielded to obtain a second image, and surface texture detection is performed on the second image; when the texture detection result is greater than a preset threshold, a calibration prompt is given.

[0010] Furthermore, the method of obtaining a standard image of the keyboard leather cover includes placing a sample keyboard leather cover in a standardized shooting device, wherein the shooting device includes at least one high-definition camera and multiple adjustable light sources, and the light source uses multi-angle lighting to ensure that various surface features of the keyboard leather cover are clearly presented during the shooting process.

[0011] Furthermore, the first layout is determined manually.

[0012] Further, the first layout is determined using an automated method, the automated method comprising:

[0013] Acquired standard images;

[0014] Use edge detection algorithm to extract features from the keyboard keys in the standard image and identify the outline of each key;

[0015] Based on the extracted key contours, morphological operations are used to further accurately define the key boundaries;

[0016] Extract and recognize the text on the key surface through optical character recognition technology;

[0017] The geometric features of all buttons are saved as button layout data;

[0018] Perform texture analysis on the leather case part in the standard image;

[0019] By detecting the natural creases on the surface of the leather case or the crease areas expected by the design;

[0020] For the identified crease position, further analyze the characteristic parameters of the crease depth, length and width;

[0021] The expected position and morphological feature data of the leather case crease are saved as crease layout data to form a standard layout including the crease position, morphology and geometric parameters.

[0022] Furthermore, comparing the real-time image of the keyboard leather cover in production with the standard image to identify the position deviation of the key layout and the fold of the leather cover includes:

[0023] Preprocessing the real-time image;

[0024] The features of key boundaries and crease areas are enhanced through morphological processing;

[0025] Compare the key layout extracted from the real-time image with the pre-stored standard key layout one by one;

[0026] The deviation of the position coordinates of each key is calculated, and the deviation distance is calculated using the Euclidean distance formula.

[0027] Furthermore, comparing the real-time image of the keyboard leather cover in production with the standard image to identify the position deviation of the key layout and the fold of the leather cover also includes:

[0028] The characteristics of leather case creases are extracted from real-time images by using curvature analysis method;

[0029] Compare the crease positions extracted from the real-time image with those in the standard image;

[0030] By comparing the position coordinates and morphological features of the crease with the preset crease parameters in the standard layout, it is determined whether it meets the standards.

[0031] Furthermore, shielding the expected fold and key portion in the real-time image includes:

[0032] Based on the coordinates and size of the key, a binary mask of the key area is created on the image, and the pixel value of the key area is set to zero, indicating that the area does not participate in subsequent texture detection.

[0033] Further, performing surface texture detection on the second image includes:

[0034] The texture features are extracted from the effective surface area in the second image.

[0035] Further, performing surface texture detection on the second image further includes:

[0036] The contrast is further calculated, which reflects the severity of local changes.

[0037] Furthermore, when the texture detection result is greater than a preset threshold, the prompt calibration includes:

[0038] When the contrast is greater than the preset threshold, a calibration prompt is displayed.

[0039] The present invention can produce the following beneficial effects through the above technical solution:

[0040] By combining image processing algorithms with intelligent texture detection technology, the entire process of automated detection from image acquisition, comparison, shielding to texture analysis is realized, which reduces the subjective errors of manual detection, greatly improves detection efficiency, and is suitable for real-time quality control in large-scale production environments.

[0041] By introducing grayscale-based contrast calculation and comparing it with the preset threshold, the system can flexibly adjust the detection sensitivity to adapt to keyboard leather covers of different materials and designs, ensuring quality stability under various working conditions.

[0042] By shielding the keyboard key part and the expected leather cover crease area, the interference of these inherent structures in image detection is effectively avoided, ensuring that the subsequent surface texture detection is more accurate, thereby improving the overall accuracy of detection and reducing false alarms and missed detections. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0045] Below, the invention is preferably described in conjunction with the accompanying drawings and specific implementation methods.

[0046] This embodiment solves the above problem through the following steps:

[0047] In one embodiment, reference Figure 1 The present invention provides a keyboard leather cover production quality inspection method, specifically relates to a detection technology scheme for detecting the appearance quality of the keyboard leather cover, which aims to effectively improve the quality control efficiency in the production of keyboard leather covers through automated image processing and intelligent analysis means, reduce quality defects caused by human detection errors or limitations of traditional methods, and ensure the functional reliability and appearance consistency of the keyboard leather cover during use.

[0048] A standard image of the keyboard leather cover is obtained, wherein the standard image includes a keyboard key portion and a leather cover portion, wherein the keyboard key portion and the leather cover portion are integrally formed and are pre-stored in the system as a reference image.

[0049] First, place the sample keyboard leather cover in a standardized shooting device, which includes at least one high-definition camera and multiple adjustable light sources. The light sources use multi-angle lighting to ensure that the various surface features of the keyboard leather cover are clearly presented during the shooting process. Then, control the camera to shoot the keyboard leather cover from multiple angles and in all directions to obtain multiple high-resolution image data including the keyboard key part and the leather cover part. The keyboard key part includes multiple physical keys arranged in a regular pattern, while the leather cover part covers the entire tablet area and is integrally formed with the key part to ensure that the captured image fully reflects the appearance and structural characteristics of the keyboard leather cover.

[0050] After the shooting is completed, the image processing algorithm is used to stitch and optimize multiple images to generate a complete standard image to ensure that the image is free of distortion or aberration. The standard image includes the specific layout of the keyboard keys, the surface details of the keys, the connection area between the keys and the leather case, and the creases and texture features on the surface of the leather case. Subsequently, the standard image is stored in the database of the inspection system and used as a reference image in the subsequent quality inspection process.

[0051] The reference image is preset as the detection benchmark in the system, and subsequent real-time detection will be based on comparative analysis of the standard image to ensure that the appearance and structure of each keyboard cover in the production process are consistent with the standard image, thereby achieving efficient and accurate quality inspection.

[0052] A first layout is determined according to the standard image, wherein the first layout is a standard layout of a keyboard leather cover, wherein the layout includes a key layout and a leather cover crease position, wherein the key layout defines a key arrangement of a keyboard, and the leather cover crease position is an expected crease area.

[0053] First, the key layout is the standard arrangement of the keys in the keyboard leather case. The key layout is determined by analyzing the position, size and shape information of each key in the keyboard part in the standard image to ensure that the physical arrangement of the keyboard keys is consistent with the standard design. The key layout not only defines the spatial position of each key relative to the leather case, but also includes detailed information such as the distance between the keys, the row and column arrangement relationship, and the position and size of the text or symbols on the key surface. The key layout is accurately extracted from the standard image through an image processing algorithm and recorded by the system as a reference benchmark for calibration and detection of the key arrangement during the production process.

[0054] Secondly, the leather cover crease position is the crease area expected to be formed during the use of the keyboard leather cover. By analyzing the surface features of the leather cover part in the standard image, the system automatically determines the crease position and range that may be formed when the leather cover is folded, unfolded or used for a long time. The crease position is part of the expected design, and its shape, depth and distribution are defined by the crease feature information in the standard image, ensuring that the crease area on the leather cover surface under normal use will not be mistakenly judged as a quality defect.

[0055] The first layout may be determined manually or by using an automated method.

[0056] When using an automated method for determination, the present invention provides a specific implementation step based on image processing and feature analysis, wherein the first layout includes a button layout and a leather case fold position, and the specific implementation scheme is as follows:

[0057] First, the acquired standard image needs to be preprocessed to ensure the quality of the image data. The preprocessing step includes operations such as image denoising, brightness and contrast adjustment, and image distortion correction to ensure the clarity and accuracy of the image. The standard image is denoised by using a filtering algorithm to eliminate noise caused by the shooting environment or hardware equipment, thereby improving the image's fineness and feature fidelity.

[0058] Use edge detection algorithms (such as the Canny edge detection algorithm or the Sobel operator) to extract features from the keyboard keys in the standard image and identify the outline of each key. This step separates the key area from the background area in the standard image and generates clear key boundaries.

[0059] Based on the extracted key contours, morphological operations (such as dilation and erosion operations) are used to further accurately define the boundaries of the keys. Then, geometric analysis methods are used to calculate the position coordinates of each key, the distance between keys, and the row and column arrangement of keys.

[0060] If there are characters or symbols (such as letters, numbers, etc.) on the surface of the key, the text on the surface of the key is extracted and recognized through optical character recognition technology (OCR) to ensure that the character position and size of the key are consistent with the design in the standard layout.

[0061] The geometric features of all keys (including position coordinates, arrangement order, spacing and character information) are saved as key layout data. The key layout data includes the absolute position and relative position of each key for subsequent real-time detection and comparison.

[0062] The texture analysis of the leather case in the standard image is performed, and the texture analysis algorithm (such as gray level co-occurrence matrix method or wavelet transform) is used to extract the surface texture features of the leather case. In particular, the texture changes of the area where the leather case may have creases are analyzed to identify the potential location and morphological features of the creases.

[0063] By detecting the natural creases on the surface of the leather case or the crease areas expected by the design, the curvature analysis method is used to determine the deformation amplitude of the local area of ​​the leather case surface, and these areas are marked as the location of the crease. This process locates the crease by analyzing the curvature changes and brightness gradient changes of the leather case surface.

[0064] For the identified crease position, the characteristic parameters such as the depth, length and width of the crease are further analyzed. The shape contour of the crease is calculated through the grayscale difference in the image and compared with the expected crease standard in the design to ensure that the crease shape is consistent with the expected design.

[0065] The expected position and morphological feature data of the leather case crease are saved as crease layout data to form a standard layout including the crease position, morphology and its geometric parameters as a reference for quality inspection.

[0066] Finally, the key layout and the leather cover fold position layout are combined to form a first layout. The first layout includes not only the arrangement information of the keys and the characteristic parameters of each key, but also the position information of the expected folds on the leather cover surface. The first layout is stored by the system as standard reference data, which serves as a benchmark for the quality inspection of the keyboard leather cover in the subsequent production process.

[0067] The real-time image of the keyboard leather cover in production is compared with the standard image to identify the position deviation of the key layout and the leather cover fold; based on the preset error range, if the position deviation of the key layout or the leather cover fold is within the error range of the standard layout, the layout is considered qualified; if it exceeds the preset error range, a calibration prompt is given.

[0068] During the production process, a high-precision camera device is used to obtain real-time images of the keyboard leather cover in production. The real-time image includes the keyboard key part and the leather cover part, ensuring that the acquired image covers the structural features of the entire keyboard leather cover, including the arrangement of the keys and the surface creases of the leather cover.

[0069] To ensure that the real-time image can be accurately compared with the standard image, the real-time image is first preprocessed. The preprocessing steps include: contrast enhancement, denoising, grayscale conversion, and distortion correction to ensure that the image clarity is consistent with the standard image. Morphological processing (such as dilation and erosion) is used to further enhance the features of the key boundary and crease area.

[0070] The same edge detection algorithm as the standard image (such as Canny edge detection or Sobel operator) is used to extract the outline and position features of the keys from the real-time image. The system identifies the coordinate information of each key by analyzing the shape, edge features and arrangement order of the keys.

[0071] The key layout extracted from the real-time image is compared with the pre-stored standard key layout one by one. The position coordinates (x, y) of each key are calculated for deviation, and the deviation distance is calculated using the Euclidean distance formula.

[0072] Compare the deviation distance of each button with the preset error range. The preset error range can be set according to production requirements, for example, the tolerance range of button position deviation is ±0.5 mm. If the deviation is less than or equal to the error range, the button position is qualified; if the deviation exceeds the error range, a calibration prompt is issued.

[0073] The features of leather case creases are extracted from the real-time image by curvature analysis. The curvature changes, grayscale differences or brightness gradients of the creases are used to identify the possible crease locations and contours on the leather case surface.

[0074] The crease positions extracted from the real-time image are compared with those in the standard image. By comparing the position coordinates and morphological features of the crease (such as crease length, depth and width) with the preset crease parameters in the standard layout, it is determined whether the standard is met. The deviation calculation formula is also based on the Euclidean distance, combined with the curvature change for comprehensive morphological evaluation.

[0075] If the deviations of the key layout and the position of the leather cover fold are within the preset error range, the system determines that the keyboard leather cover layout is qualified; if the deviation of any key or fold exceeds the error range, the system will prompt for calibration. The calibration prompt can notify the operator through visual, sound or software interface to manually or automatically correct the relevant positions.

[0076] According to the first layout, the expected folds and button parts in the real-time image are shielded to obtain a second image, and surface texture detection is performed on the second image; when the texture detection result is greater than a preset threshold, a calibration prompt is given.

[0077] First, based on the key positions and crease positions determined in the first layout, the system shields the corresponding areas in the real-time image. The main purpose of shielding is to exclude these normal, expected structural features and ensure that the surface texture detection will not be interfered by key protrusions or creases. The specific shielding implementation process includes:

[0078] According to the key outline data stored in the first layout, the system removes the key area from the image in the form of a mask. The specific implementation method is to create a binary mask of the key area on the image based on the coordinates and size of the key, and set the pixel value of the key area to zero or other fixed values, indicating that the area does not participate in the subsequent texture detection. The formula of the binary mask is:

[0079] ;

[0080] in, The mask image representing the key part, are pixel coordinates in the image.

[0081] Similarly, according to the expected fold position and morphological information in the first layout, a binary mask of the fold area is generated, and the fold area is shielded from the image in the same way. The mask calculation formula of the fold area is:

[0082] ;

[0083] in, Mask image showing the folded part

[0084] By masking the pixel values ​​of all key areas and crease areas in the real-time image to zero, a second image with interference structures removed is obtained.

[0085] After the above processing, the second image The expression is:

[0086] ;

[0087] in, is the pixel value of the real-time image.

[0088] After the shielding process is completed, the second image obtained only contains the effective area of ​​the keyboard leather cover surface (that is, excluding the keys and folds). Surface texture detection is performed in this effective area to ensure that there are no scratches, indentations, bubbles and other defects on the leather cover surface. The specific texture detection process is as follows:

[0089] Extract texture features from the effective surface area in the second image:

[0090] ;

[0091] in Represents the grayscale level, which is the joint probability of the pixel coordinate (i, j) in the direction of distance d and angle θ. The resolution of the image is 𝑀×𝑁, where 𝑀 represents the number of pixels in the vertical direction (height) of the image, and 𝑁 represents the number of pixels in the horizontal direction (width). and It is the horizontal and vertical coordinate offset of adjacent pixels (x, y) and (i, j).

[0092] Further calculation of contrast

[0093] ;

[0094] Where K represents the detection range, that is, the range within the circle with (i, j) as the center and K as the radius.

[0095] The contrast reflects the severity of local changes. Since the present invention removes the buttons and expected folds in the previous steps, when the contrast is greater than the preset threshold, it means that an area beyond the expected area has appeared, and the area is a possible abnormal area, such as the presence of abnormal texture features, such as scratches, bubbles and other defects.

[0096] When the result of the surface texture detection exceeds the preset threshold, the system will automatically generate a prompt to notify the operator to calibrate or further inspect the keyboard cover. The prompt can be issued through a display, an alarm sound, or other means to remind the operator on the production line to deal with possible quality problems.

[0097] Through the above steps, the present invention can effectively shield the inherent structural interference of keys and folds, and accurately detect the texture of the keyboard leather cover surface, ensuring the accuracy and reliability of the detection results. When abnormal surface texture is detected, a calibration prompt is issued in time to ensure the consistency of product quality during the production process.

[0098] The prior art mentioned in the aforementioned background technology part and specific embodiment part of the present invention may be regarded as a part of the present invention, and is used to understand the meaning of some technical features or parameters.

Claims

1. A keyboard leather cover production quality inspection method, characterized in that: The method comprises the following steps: Acquire a standard image of the keyboard leather cover, wherein the standard image includes a keyboard key portion and a leather cover portion, wherein the keyboard key portion and the leather cover portion are integrally formed and are pre-stored in the system as a reference image; Determine a first layout according to the standard image, the first layout being a standard layout of the keyboard leather cover, the layout comprising a key layout and a leather cover fold position, the key layout defines a key arrangement of the keyboard, and the leather cover fold position is an expected fold area; Compare the real-time image of the keyboard leather cover in production with the standard image to identify the position deviation of the key layout and the fold of the leather cover; based on the preset error range, if the position deviation of the key layout or the fold of the leather cover is within the error range of the standard layout, the layout is considered qualified, and if it exceeds the preset error range, a calibration prompt is given; According to the first layout, the expected folds and the key parts in the real-time image are shielded to obtain a second image, and surface texture detection is performed on the second image; When the texture detection result is greater than the preset threshold, a calibration prompt will be displayed; The first layout is determined using an automated method, the automated method comprising: Acquired standard images; Use edge detection algorithm to extract features from the keyboard keys in the standard image and identify the outline of each key; Based on the extracted key contours, morphological operations are used to further accurately define the key boundaries; Extract and recognize the text on the key surface through optical character recognition technology; The geometric features of all buttons are saved as button layout data; Perform texture analysis on the leather case part in the standard image; By detecting the natural creases on the surface of the leather case or the crease areas expected by the design; For the identified crease position, further analyze the characteristic parameters of the crease depth, length and width; The expected position and morphological feature data of the leather case crease are saved as crease layout data to form a standard layout including the crease position, morphology and its geometric parameters; Wherein, performing surface texture detection on the second image includes: Extract texture features from the effective surface area in the second image: ； in represents the gray level, is the joint probability of the pixel pair i and j in the direction of distance d and angle θ, M and N are the image sizes, and is the horizontal and vertical coordinate offset of adjacent pixels, represents a second image; Performing surface texture detection on the second image further includes: Further calculation of contrast ; Where K represents the detection range, and the contrast reflects the severity of local changes.

2. A keyboard leather cover production quality inspection method according to claim 1, characterized in that: The method of obtaining a standard image of the keyboard leather cover includes placing a sample keyboard leather cover in a standardized shooting device, wherein the shooting device includes at least one high-definition camera and multiple adjustable light sources, and the light source uses multi-angle lighting to ensure that various surface features of the keyboard leather cover are clearly presented during the shooting process.

3. A keyboard leather cover production quality inspection method according to claim 1, characterized in that: The first layout is determined manually.

4. A keyboard leather cover production quality inspection method according to claim 1, characterized in that: Comparing the real-time image of the keyboard leather cover in production with the standard image, identifying the position deviation of the key layout and the leather cover fold includes: Preprocessing the real-time image; The features of key boundaries and crease areas are enhanced through morphological processing; Compare the key layout extracted from the real-time image with the pre-stored standard key layout one by one; The deviation of the position coordinates of each key is calculated, and the deviation distance is calculated using the Euclidean distance formula.

5. A keyboard leather cover production quality inspection method according to claim 4, characterized in that: Comparing the real-time image of the keyboard leather cover in production with the standard image to identify the position deviation of the key layout and the leather cover fold also includes: The characteristics of leather case creases are extracted from real-time images by using curvature analysis method; Compare the crease positions extracted from the real-time image with those in the standard image; By comparing the position coordinates and morphological features of the crease with the preset crease parameters in the standard layout, it is determined whether it meets the standards.

6. A keyboard leather cover production quality inspection method according to claim 1, characterized in that: The expected fold and key portion in the real-time image is partially shielded including: Based on the coordinates and size of the key, a binary mask of the key area is created on the image, and the pixel value of the key area is set to zero, indicating that the area does not participate in subsequent texture detection.

7. A keyboard leather cover production quality inspection method according to claim 1, characterized in that: Performing surface texture detection on the second image includes: The texture features are extracted from the effective surface area in the second image.

8. A keyboard leather cover production quality inspection method according to claim 7, characterized in that: Performing surface texture detection on the second image further includes: The contrast is further calculated, which reflects the severity of local changes.

9. A keyboard leather cover production quality inspection method according to claim 8, characterized in that: When the texture detection result is greater than the preset threshold, the calibration prompt includes: When the contrast is greater than the preset threshold, a calibration prompt is displayed.

Images