Galvanized automobile sheet black spot detection method and related equipment

Abstract

The invention discloses a galvanized automobile sheet black spot detection method and related equipment, and relates to the technical field of metallurgical machinery, and the method comprises the steps: obtaining the specification information of strip steel; determining an upper surface detection area and a lower surface detection area of the strip steel based on the specification information and a preset edge offset; acquiring a first surface image and a second surface image; performing feature extraction on the first surface image and the second surface image to generate first black spot feature data and second black spot feature data; identifying the first black spot feature data and the second black spot feature data based on a deep learning model, and counting the number of first black spots in the upper surface detection area and the number of second black spots in the lower surface detection area; and determining the quality grade of the strip steel based on the first black spot number, the second black spot number and a preset black spot threshold value. According to the invention, automatic and full-process detection of black spot defects is realized.

Description

Technical Field

[0001] The present application relates to the field of metallurgical machinery technology, and in particular to a method for detecting black spots on galvanized automobile sheets and related equipment. Background Art

[0002] Galvanized automotive sheet is an important material for automobile manufacturing, and its surface quality directly affects the subsequent processing performance and product appearance. In the production process of cold-rolled strip galvanizing, black spot defects are a common quality problem, mainly caused by factors such as zinc liquid inclusions, oxidation residues or process fluctuations. Such defects can lead to uneven coating, and in severe cases, cause product degradation or even scrapping. At present, traditional black spot detection mainly relies on manual visual inspection or offline sampling analysis. It is necessary to cut the sample after the production line is stopped, and use a microscope to locally count the defects. However, the manual method is inefficient and highly subjective, and it is difficult to achieve full coverage in high-intensity continuous production. There are problems such as high missed detection rate and great safety hazards. Therefore, there is an urgent need for a method for detecting black spots on galvanized automotive sheets to solve the above-mentioned technical problems. Summary of the Invention

[0003] The Summary of the Invention introduces a series of simplified concepts that will be further described in the Detailed Description of the Invention. The Summary of the Invention of this application is not intended to limit the key features and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0004] In a first aspect, the present application provides a method for detecting black spots on galvanized automobile sheets, comprising:

[0005] Obtain the specification information of the strip steel;

[0006] Based on the specification information and the preset edge offset, determining the upper surface detection area and the lower surface detection area of ​​the strip, wherein the upper surface detection area includes a first driving side detection area, a first middle detection area, and a first operating side detection area, and the lower surface detection area includes a second driving side detection area, a second middle detection area, and a second operating side detection area;

[0007] Acquire a first surface image of the upper surface detection area and a second surface image of the lower surface detection area;

[0008] Performing feature extraction on the first surface image and the second surface image to generate first black dot feature data corresponding to the first surface image and second black dot feature data corresponding to the second surface image;

[0009] Identify the first black dot feature data and the second black dot feature data based on a deep learning model, and count the number of first black dots in the upper surface detection area and the number of second black dots in the lower surface detection area;

[0010] The quality grade of the steel strip is determined based on the first black spot number, the second black spot number, and a preset black spot threshold.

[0011] In some embodiments, determining the upper surface inspection area of ​​the steel strip based on the specification information and the preset edge offset includes:

[0012] Based on the strip length and the preset edge offset according to the specification information, the upper surface is divided into a first driving side area, a first operating side area, and a first middle area along the length direction, wherein the preset edge offset includes the driving side offset and the operating side offset;

[0013] In the first driving side area, a first driving side detection area is selected along the length direction based on a preset fixed side length value, and the first driving side detection area includes the first driving side center point;

[0014] In the first operating side area, a first operating side detection area is selected along the length direction based on a preset fixed side length value, wherein the first operating side detection area includes the first operating side center point, wherein the first driving side center point and the first operating side center point are located on the same horizontal straight line;

[0015] Determining a first central point of the first central detection area based on the first driving side center point, the first operating side center point, and the first central area;

[0016] Based on the first central point, a first central detection area is determined.

[0017] In some embodiments, based on the strip length and the preset edge offset in the specification information, the upper surface is divided into a first driving side area, a first operating side area, and a first middle area along the length direction, including:

[0018] Based on the strip length and the drive side offset, sequentially determining a drive side starting position and a drive side ending position of the first drive side region along the first direction;

[0019] determining a first driving side region of the upper surface based on the driving side starting position and the driving side ending position;

[0020] Based on the strip length and the operating side offset, sequentially determining the operating side starting position and the operating side ending position of the first operating side region along a second direction opposite to the first direction;

[0021] Determining a first operation side area of ​​the upper surface based on the operation side starting position and the operation side ending position;

[0022] Based on the driving side end position and the operating side end position, a first middle area of ​​the upper surface is determined, wherein the middle starting position of the first middle area is the driving side end position and the middle end position is the operating side end position.

[0023] In some embodiments, the first surface image includes a first driving side image, a first operating side image, and a first middle image, and acquiring the first surface image of the upper surface detection area includes:

[0024] Determining a first driving side image acquisition range of the first driving side detection area based on the first driving side center point and a preset fixed side length value;

[0025] Determining a first operating side image acquisition range of the first operating side detection area based on the first operating side center point and a preset fixed side length value;

[0026] Determining a first central image acquisition range of the first central detection area based on the first central center point and a preset fixed side length value;

[0027] Based on the preset scanning resolution, the first driving side image acquisition range, the first operating side image acquisition range and the first central image acquisition range are scanned line by line by a line scanning camera to generate the first driving side image, the first operating side image and the first central image.

[0028] In some embodiments, performing feature extraction on the first surface image to generate first black spot feature data corresponding to the first surface image includes:

[0029] Based on the first driving side image, the first operating side image, and the first central image, adaptive light correction and distortion correction are performed respectively to generate corresponding first driving side correction image, first operating side correction image, and first central correction image;

[0030] Denoising the first driving-side corrected image, the first operating-side corrected image, and the first central corrected image based on a preset filtering algorithm to generate a first driving-side denoised image, a first operating-side denoised image, and a first central denoised image;

[0031] performing black dot region enhancement on the first driving side denoised image, the first operating side denoised image and the first central denoised image based on a contrast enhancement algorithm to generate a first driving side enhanced image, a first operating side enhanced image and a first central enhanced image;

[0032] Based on a preset segmentation rule, the first driving side enhanced image, the first operating side enhanced image, and the first central enhanced image are respectively segmented into a plurality of sub-images, wherein the size of each sub-image is smaller than a preset fixed side length value;

[0033] Extracting black spot candidate regions corresponding to the first driving side image, the first operating side image, and the first central image from the plurality of sub-images based on a size threshold, shape characteristics, and color difference of the black spots;

[0034] First black dot feature data is determined according to the grayscale distribution and edge gradient characteristics of the black dot candidate area, wherein the first black dot feature data includes black dot position, black dot size and black dot shape parameters.

[0035] In some embodiments, identifying the first black dot feature data and the second black dot feature data based on a deep learning model, and counting the number of first black dots in the upper surface detection area and the number of second black dots in the lower surface detection area, includes:

[0036] Based on the first black spot feature data and the second black spot feature data, feature matching is performed on the black spot candidate area through a deep learning model to determine the confidence parameters of the candidate black spots;

[0037] Based on the comparison result between the confidence parameter and the preset confidence threshold, the valid black spots are screened out;

[0038] Based on the positions and sizes of the valid black spots, the valid black spots in the upper surface detection area and the lower surface detection area are deduplicated to determine the target black spot position set;

[0039] Based on the number of black dots in the target black dot position set, a first number of black dots in the upper surface detection area and a second number of black dots in the lower surface detection area are determined.

[0040] In some embodiments, the preset black spot threshold includes a first preset threshold and a second preset threshold, and determining the quality grade of the steel strip based on the first black spot number, the second black spot number, and the preset black spot threshold includes:

[0041] determining a quality state of the upper surface of the upper surface detection area based on a comparison result of the first black spot quantity and a first preset threshold;

[0042] determining a lower surface quality state of the lower surface detection area based on a comparison result of the second black spot quantity and a second preset threshold;

[0043] The quality grade of the strip is determined based on the logical combination relationship between the upper surface quality status and the lower surface quality status.

[0044] In a second aspect, the present application proposes a device for detecting black spots on galvanized automobile sheets, comprising:

[0045] A specification information acquisition unit, used to acquire specification information of the strip steel;

[0046] a detection area determination unit, which determines an upper surface detection area and a lower surface detection area of ​​the steel strip based on the specification information and the preset edge offset, wherein the upper surface detection area includes a first drive side detection area, a first middle detection area, and a first operating side detection area, and the lower surface detection area includes a second drive side detection area, a second middle detection area, and a second operating side detection area;

[0047] a surface image acquisition unit, configured to acquire a first surface image of an upper surface detection area and a second surface image of a lower surface detection area;

[0048] A black spot feature generating unit, configured to extract features from the first surface image and the second surface image, and generate first black spot feature data corresponding to the first surface image and second black spot feature data corresponding to the second surface image;

[0049] a black dot number counting unit, which identifies the first black dot feature data and the second black dot feature data based on a deep learning model, and counts the first black dot number in the upper surface detection area and the second black dot number in the lower surface detection area;

[0050] The quality grade determining unit determines the quality grade of the steel strip based on the first black spot number, the second black spot number and a preset black spot threshold.

[0051] In a third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the method for detecting black spots on galvanized automobile sheets according to any one of the first aspects when executing the computer program stored in the memory.

[0052] In a fourth aspect, the present application proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any method for detecting black spots on galvanized automobile sheets according to the first aspect.

[0053] In summary, this application achieves automated, full-process detection of black spot defects by dynamically adjusting the multi-area detection range on the upper and lower surfaces of the steel strip, combining the efficient image acquisition of line scan cameras with the precise recognition of deep learning models. Specifically, the detection area is adaptively determined based on the steel strip specification information, avoiding manual intervention; double-sided synchronous image processing and black spot number statistics improve detection efficiency and coverage; and the intelligent analysis of black spot feature data using deep learning models effectively reduces the false detection rate and missed detection rate, ensuring the objectivity and accuracy of quality judgment results. This application not only improves production efficiency, but also reduces the safety risks of manual operation, providing reliable technical support for the quality control of galvanized automotive sheets. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present description. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0055] Figure 1A schematic flow chart of a method for detecting black spots on galvanized automobile sheets provided in an embodiment of the present application;

[0056] Figure 2 A schematic diagram of the detection results of black spots on galvanized automobile sheets provided in an embodiment of the present application;

[0057] Figure 3 A schematic structural diagram of a device for detecting black spots on galvanized automobile sheets provided in an embodiment of the present application;

[0058] Figure 4 This is a schematic diagram of the structure of an electronic device for detecting black spots on galvanized automobile sheets provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the embodiments described are only part of the embodiments of the present application, not all of the embodiments.

[0060] See also Figure 1 , which is a flow chart of a method for detecting black spots on galvanized automobile sheets provided in an embodiment of the present application, which may specifically include:

[0061] S110, obtaining specification information of the strip steel;

[0062] For example, gauge information is basic parameters recorded during the strip production process, including key physical properties such as width, length, and thickness. This information is typically collected in real time by sensors or control systems on the production line and stored in a database. The primary purpose of acquiring gauge information is to provide a basis for the dynamic division of subsequent inspection areas, ensuring that the inspection range can adapt to different strip sizes and avoiding blind spots or overlapping coverage caused by gauge differences.

[0063] By acquiring this specification information, combined with pre-set parameters such as edge offset, the initial positions and distribution patterns of the upper and lower surface inspection zones can be quickly determined. For example, the strip width directly influences the division of the drive-side, operating-side, and center inspection zones, while the length parameter dynamically adjusts the sampling interval of the inspection zones along the strip's extension. This step provides the necessary data foundation for subsequent image acquisition and black spot identification, ensuring that the inspection process accurately matches the actual strip production status.

[0064] S120. Determine an upper surface detection area and a lower surface detection area of ​​the steel strip based on the specification information and the preset edge offset, wherein the upper surface detection area includes a first driving side detection area, a first middle detection area, and a first operating side detection area, and the lower surface detection area includes a second driving side detection area, a second middle detection area, and a second operating side detection area;

[0065] For example, the inspection areas on the upper and lower surfaces are dynamically divided based on the strip's specifications and preset edge offsets. Based on the strip's actual dimensions, the drive-side offset and operator-side offset define the start and end positions of the edge inspection area. Through the synergistic effect of these two, the drive-side, middle, and operator-side inspection areas are sequentially divided along the strip's length, ensuring that the inspection range covers critical quality risk areas on the strip's surface and avoiding blind spots caused by width variations.

[0066] Furthermore, the lower surface inspection area is divided according to the same logic as the upper surface, with the locations of the drive-side, center, and operator-side inspection zones determined through symmetrical mapping. This dynamic division mechanism adapts to the production needs of varying strip sizes, ensuring synchronization and consistency between the upper and lower surface inspection areas, and providing a standardized spatial reference for subsequent image acquisition and black spot identification.

[0067] S130, acquiring a first surface image of the upper surface detection area and a second surface image of the lower surface detection area;

[0068] For example, a line scan camera captures high-precision images of pre-defined inspection areas on the upper and lower surfaces of the strip. Based on the dynamic movement of the strip, the line scan camera scans the inspection area line by line along its length, generating surface images covering the drive, center, and operating sides. Synchronous acquisition of upper and lower surface images ensures consistent black spot data on both sides, preventing detection errors caused by strip movement or process fluctuations.

[0069] The first and second surface images serve as raw data, providing input for subsequent black spot feature extraction and identification. Preset scanning resolution and dynamic exposure control ensure image clarity and detail capture, adapting to inspection requirements under varying lighting conditions. The captured images are directly transmitted to the processing system, where they are associated with strip specifications and inspection area coordinates to establish a mapping relationship for black spot location and counting.

[0070] S140, performing feature extraction on the first surface image and the second surface image to generate first black dot feature data corresponding to the first surface image and second black dot feature data corresponding to the second surface image;

[0071] Exemplarily, the feature extraction process first pre-processes the collected surface image to eliminate environmental interference and enhance the recognition of the black spot area. Through adaptive light correction and distortion correction technology, the brightness difference and geometric deformation of the image caused by shooting angle or uneven lighting are adjusted to generate a standardized corrected image. Subsequently, a filtering algorithm is used to remove image noise, and contrast enhancement technology is used to enhance the brightness and darkness difference between the black spot and the background to ensure the visibility of small defects. On this basis, the black spot candidate area is further extracted through image segmentation and feature screening. The enhanced image is divided into multiple sub-images to expand the relative size proportion of the black spots for subsequent analysis. Combined with the size threshold, shape characteristics and color difference of the black spots, potential black spot areas are screened out from the sub-image, and black spot feature data containing position, size and morphological parameters is generated based on the grayscale distribution and edge gradient features. This process provides structured input for the deep learning model, supporting accurate statistics of the number of black spots and quality judgment.

[0072] S150: Identify the first black dot feature data and the second black dot feature data based on a deep learning model, and count the number of first black dots in the upper surface detection area and the number of second black dots in the lower surface detection area;

[0073] For example, a pre-trained deep learning model analyzes black spot feature data to accurately identify and count black spot defects. Based on a convolutional neural network architecture, the model performs feature matching and classification on candidate black spot regions, calculating confidence parameters for each candidate region. Valid black spots are screened using a pre-set confidence threshold, eliminating false positives due to noise or interference. Furthermore, combining the location and size of the black spots, duplicate or overlapping black spots in the upper and lower surface inspection areas are removed to generate a unique set of black spot locations, ensuring the accuracy of the statistical results. This process achieves robustness against complex background noise, ensuring the reliable detection of minor defects while avoiding subjective biases caused by human intervention.

[0074] Based on the number of black spots in the target black spot location set, the total number of black spots in the upper and lower surface inspection areas is counted. The statistical results are directly linked to the strip quality determination logic, providing core data support for subsequent quality grading. The efficient computation and adaptive learning capabilities of the deep learning model improve the coverage and accuracy of black spot identification, reducing reliance on manual re-inspection.

[0075] S160: Determine the quality grade of the steel strip based on the first number of black spots, the second number of black spots, and a preset black spot threshold.

[0076] Exemplarily, the objective determination of the strip's quality grade is achieved by comparing a preset black spot threshold with the number of black spots in the upper and lower surface inspection areas. Specifically, the upper surface's quality status (e.g., qualified, unqualified, or critical) is determined based on the comparison of the first black spot count with the first preset threshold. Simultaneously, the lower surface's quality status is determined based on the comparison of the second black spot count with the second preset threshold. Preset logical combination rules are used to combine the quality statuses of the upper and lower surfaces, ultimately outputting a comprehensive strip quality grade (e.g., level 1 qualified, level 2 warning, or level 3 unqualified).

[0077] This judgment mechanism, through a layered threshold design and a state combination strategy, avoids the risk of misjudgment that could result from a single threshold. Simultaneously, by combining simultaneous inspection results from both the upper and lower surfaces, it ensures that the quality grade reflects the overall defect distribution characteristics of the strip, providing a reliable basis for subsequent production decisions.

[0078] In summary, in the embodiment of the present application, the multiple detection areas of the upper and lower surfaces are automatically divided by dynamically adapting to the strip specification information, and the high-efficiency image acquisition of the line scan camera and the analysis of the deep learning model are combined to realize the full-process automated detection of black spot defects on galvanized automobile plates. The embodiment of the present application dynamically adjusts the position and range of the detection areas on the drive side, middle side and operating side based on the actual size of the strip to ensure that the detection covers key quality risk areas and avoids missed detection or repeated detection due to width changes; through double-sided synchronous image acquisition and preprocessing technology, it eliminates the interference of uneven illumination and geometric distortion, enhances the contrast between black spots and background, and provides a high-quality data foundation for subsequent feature extraction; uses a deep learning model to identify and deduplicate black spot candidate areas, effectively distinguishes between real defects and noise interference, and reduces the false detection rate and missed detection rate; finally, based on the hierarchical threshold judgment rule, it comprehensively evaluates the number of black spots on the upper and lower surfaces and outputs an objective quality grade result. The embodiment of the present application improves the detection efficiency and accuracy, reduces the safety hazards of manual intervention, and at the same time provides real-time quality monitoring capabilities for continuous production of the production line, meeting the control needs of automobile outer panels for surface defects.

[0079] In some examples, determining the upper surface inspection area of ​​the steel strip based on the specification information and the preset edge offset includes:

[0080] Based on the strip length and the preset edge offset according to the specification information, the upper surface is divided into a first driving side area, a first operating side area, and a first middle area along the length direction, wherein the preset edge offset includes a driving side offset and an operating side offset, including:

[0081] Based on the strip length and the drive side offset, sequentially determining a drive side starting position and a drive side ending position of the first drive side region along the first direction;

[0082] determining a first driving side region of the upper surface based on the driving side starting position and the driving side ending position;

[0083] Based on the strip length and the operating side offset, sequentially determining the operating side starting position and the operating side ending position of the first operating side region along a second direction opposite to the first direction;

[0084] Determining a first operation side area of ​​the upper surface based on the operation side starting position and the operation side ending position;

[0085] Based on the driving side end position and the operating side end position, a first middle area of ​​the upper surface is determined, wherein the middle starting position of the first middle area is the driving side end position and the middle end position is the operating side end position.

[0086] Exemplarily, based on the strip length recorded in the specification information of the strip, combined with the preset drive side offset (i.e., a fixed distance indented inward from the drive side edge of the strip), the drive side starting position and the drive side ending position of the first drive side area are calculated in sequence along the first direction of the strip length (such as the direction from the head of the strip to the tail). The drive side offset is a preset fixed value or proportional value, which is used to define the starting offset distance of the drive side area relative to the edge of the strip. For example, if the strip length is L and the drive side offset is D1, the drive side starting position is the position where the starting point of the strip head is offset by D1 along the first direction, and the drive side ending position is the end point after the drive side starting position is extended by a preset fixed distance (such as 70 cm) along the first direction. Through the above steps, it is ensured that the coverage of the drive side area matches the actual size of the strip, and invalid detection caused by exceeding the actual size is avoided.

[0087] Based on the drive-side starting and ending positions obtained in the above steps, the first drive-side region of the upper surface is determined. This first drive-side region extends along the length of the strip, and its width is determined by the strip width as determined by the strip specification information. Specifically, the first drive-side region is a rectangular area extending along the strip width (i.e., perpendicular to the length) from the drive-side starting position to the drive-side ending position. This region's logical division ensures that the drive-side region remains at the edge of the strip, covering the critical risk area prone to black spot defects during the galvanizing process.

[0088] Based on the strip length and the preset operating side offset, the operating side starting position and the operating side ending position of the first operating side area are determined in sequence along a second direction opposite to the first direction (e.g., the direction from the tail of the strip to the head). The operating side offset D2 is a preset value independent of the drive side offset and is used to define the starting offset distance of the operating side area relative to the other side edge of the strip. For example, the operating side starting position is the position where the end point of the strip tail is offset by D2 along the second direction, and the operating side ending position is the end point after the operating side starting position is extended along the second direction by the same preset fixed distance (e.g., 70 cm). The above steps cover the potential defect area on the other side edge of the strip.

[0089] Based on the operating side starting and ending positions obtained in the above steps, a first operating side region is determined on the upper surface. The width of this first operating side region is also determined by the strip width information from the strip specifications, and its extension along the strip length is defined by the operating side starting and ending positions. Through these steps, the logical division of the operating side region ensures that the operating side region always remains on the other side of the strip, covering another risk area prone to black spot defects during the galvanizing process.

[0090] Based on the drive-side and operator-side end positions, the first central region of the top surface is determined. The middle starting position of the first central region is the drive-side end position, and the middle end position is the operator-side end position. This region covers the middle area between the drive-side and operator-side regions along the strip's length. This step ensures that the drive-side, central, and operator-side regions collectively cover the entire width of the strip's top surface, providing a spatial reference for subsequent inspection area demarcation and comprehensive identification of black spot defects.

[0091] In the first driving side area, a first driving side detection area is selected along the length direction based on a preset fixed side length value, and the first driving side detection area includes the first driving side center point;

[0092] Exemplarily, in the first driving side area, the first driving side detection area is selected based on a preset fixed side length value (for example, 10 cm) along the length direction of the strip (i.e., the first direction, such as the direction extending from the head to the tail of the strip). Specifically, the first driving side detection area is a square area with a side length equal to the preset fixed side length value, and its starting position along the length direction of the strip is determined by the driving side starting position and the preset driving side offset. By offsetting the first preset step length along the first direction from the driving side starting position, the coordinates of the upper left corner vertex of the first driving side detection area are determined, and the square area is generated by extending along the length direction and the width direction based on the preset fixed side length value. The first driving side center point of the first driving side detection area is located at the geometric center of the square area, and its coordinates are determined by calculating the mean of the vertex coordinates of the square area. The position of the first driving side center point is associated with the specification information of the strip and the preset offset to ensure that the detection area covers the key quality risk area of ​​the driving side edge.

[0093] In the first operating side area, a first operating side detection area is selected along the length direction based on a preset fixed side length value, wherein the first operating side detection area includes the first operating side center point, wherein the first driving side center point and the first operating side center point are located on the same horizontal straight line;

[0094] Exemplarily, in the first operating side area, the first operating side detection area is selected based on the same preset fixed side length value (for example, 10 cm) along the second direction opposite to the first direction (such as the direction extending from the tail of the strip to the head). The first operating side detection area is also a square area with a side length equal to the preset fixed side length value, and its starting position is determined by the operating side starting position and the preset operating side offset. By offsetting the same first preset step length from the operating side starting position along the second direction, the coordinates of the upper left corner vertex of the first operating side detection area are determined, and the square area is generated by extending along the length direction and the width direction based on the preset fixed side length value. The first operating side center point of the first operating side detection area is located at the geometric center of the square area, and its coordinates are determined by calculating the mean of the vertex coordinates. The first drive side center point and the first operating side center point are ensured to be on the same horizontal straight line through coordinate calibration, and the horizontal straight line is parallel to the length direction of the strip, thereby ensuring the symmetry and consistency of the upper and lower surface detection areas. By combining the preset offset with dynamic calculation, a single square detection area is positioned to avoid the efficiency loss of traditional multi-area detection.

[0095] Determine the first central point of the first central detection area based on the first driving side center point, the first operating side center point and the first central area

[0096] Determining a central axis of the first central area based on a line connecting the center point of the first driving side and the center point of the first operating side;

[0097] Determine the position of the first central point according to the central axis and the first central region, where the first central point is the intersection of the central axis and the symmetry center line of the first central region, and the symmetry center line is the vertical midline;

[0098] Determine a first central detection area based on the first central point, wherein each detection area is a square with a side length that is a preset fixed side length value;

[0099] Exemplarily, the central axis of the first central area is determined based on the line connecting the center point of the first driving side and the center point of the first operating side. The central axis is a straight line connecting the center point of the first driving side and the center point of the first operating side, and its direction is parallel to the length direction of the strip. The first central center point of the first central detection area is determined by calculating the intersection of the central axis and the symmetry center line of the first central area (that is, a straight line perpendicular to the central axis and bisecting the first central area). The first central detection area is a square area with a side length equal to a preset fixed side length value, and its starting position is determined by the coordinates of the first central center point. Specifically, with the first central center point as the center, half of the preset fixed side length value is extended in the length direction and the width direction respectively to generate a square area. The position of the first central center point is bound to the spatial relationship between the center points of the driving side and the operating side to ensure that the central detection area covers the potential defect area in the middle of the strip and does not overlap with the edge detection area.

[0100] The process of determining the lower surface detection area is completely symmetrical with the division logic of the upper surface detection area, and is based on the same preset parameters and dynamic adjustment mechanism. Specifically, based on the specification information of the strip and the preset edge offset, the second drive side detection area, the second middle detection area and the second operating side detection area of ​​the lower surface are dynamically divided along the length direction of the strip. First, according to the strip length and the drive side offset, the drive side starting position and the drive side ending position of the second drive side area are determined along the first direction to generate a rectangular area covering the drive side edge of the lower surface; synchronously, based on the strip length and the operating side offset, the operating side starting position and the operating side ending position of the second operating side area are determined along the second direction opposite to the first direction to form a rectangular area of ​​the operating side edge of the lower surface; subsequently, according to the drive side ending position and the operating side ending position, the middle starting position and the ending position of the second middle area are determined to cover the middle area of ​​the lower surface between the drive side and the operating side areas. On this basis, based on the preset fixed side length value, a unique square detection area is selected in the second drive side area and the second operating side area along the length direction, and the coordinates of the upper left corner vertex and the geometric center point of each detection area are determined using the same offset step size to ensure that both are located on the same horizontal line. Furthermore, by connecting the central axis of the second drive side center point and the second operating side center point, combined with the intersection of the symmetrical center lines of the middle area of ​​the lower surface, the center point of the second middle detection area is determined, and the preset fixed side length value is expanded along the length and width direction with this as the center to generate a unique square detection area. The above process ensures strict consistency in the spatial distribution, coverage range and detection accuracy of the lower surface detection area and the upper surface by symmetrically mapping the coordinates and parameter configuration of the upper surface detection area, thereby achieving simultaneous and accurate identification of double-sided defects.

[0101] In some examples, the first surface image includes a first driving side image, a first operating side image, and a first middle image, and acquiring the first surface image of the upper surface detection area includes:

[0102] Determining a first driving side image acquisition range of the first driving side detection area based on the first driving side center point and a preset fixed side length value;

[0103] Determining a first operating side image acquisition range of the first operating side detection area based on the first operating side center point and a preset fixed side length value;

[0104] Determining a first central image acquisition range of the first central detection area based on the first central center point and a preset fixed side length value;

[0105] Based on the preset scanning resolution, the first driving side image acquisition range, the first operating side image acquisition range and the first central image acquisition range are scanned line by line by a line scanning camera to generate the first driving side image, the first operating side image and the first central image.

[0106] Exemplarily, when acquiring the first surface image of the upper surface inspection area, the image acquisition range for each inspection area is first defined based on a preset fixed side length and the coordinates of the first drive-side center point, the first operating-side center point, and the first central center point. Specifically, for the first drive-side inspection area, a square image acquisition range with a side length of 10 cm is formed, centered at the first drive-side center point and extending by half the preset fixed side length in both the length and width directions. The vertex coordinates of this range are calculated using the center point coordinates and an offset to ensure full surface coverage of the drive-side inspection area. Similarly, the first operating-side image acquisition range of the first operating-side inspection area extends the same distance in both the second and width directions, centered at the first operating-side center point, to form a square area with equal side lengths. The first central image acquisition range of the first central inspection area is symmetrically extended in both the length and width directions based on the first central center point, generating a separate square coverage area. This standardized configuration of preset fixed side lengths ensures consistent spatial coverage for each inspection area, avoiding black dot identification errors due to size differences.

[0107] After determining the image acquisition range of each detection area, the line scan camera performs a line-by-line scan of the first drive side, first operating side, and first central image acquisition range. The line scan camera moves along the length of the strip based on the preset scanning resolution (50 pixels per centimeter), and synchronously triggers the image sensor to continuously shoot the target area at high speed. For each detection area, the line scan camera starts from the starting vertex of the acquisition range and collects pixel data line by line along the length direction with a preset scanning step size (0.02mm per line) until the entire square area is covered. During the scanning process, the camera's dynamic exposure control module adjusts the lighting parameters in real time to adapt to local changes in the reflectivity of the strip surface and ensure the uniformity of the image brightness. In this way, the line scan camera generates the first drive side image, the first operating side image, and the first central image respectively. The resolution of each image is determined by the preset scanning resolution and the side length of the detection area (10cm×10cm area corresponds to 500×500 pixels).

[0108] The generated surface images are strictly associated with the coordinate information of the detection area. Specifically, the pixel coordinates of the first drive side image are mapped to the physical position of the first drive side detection area, and the first operating side image and the first central image are also bound to the corresponding detection area through the coordinate conversion matrix. This mapping relationship is dynamically calculated based on the strip specification information (such as width and length) and the preset offset to ensure that the position of each pixel in the image can accurately reflect the actual defect distribution on the strip surface. In addition, the line scan camera is linked to the strip motion control system through a time synchronization module to capture images in real time during the continuous movement of the strip, avoiding image blur or dislocation caused by strip displacement.

[0109] In summary, the first drive-side image, the first operating-side image, and the first center image are input into the image processing system as raw data, providing a foundation for subsequent feature extraction and black spot identification. The acquisition range, resolution, and coordinate association information of each image are stored in the configuration file of the detection system, which supports the adjustment of preset parameters (such as side length and scanning resolution) according to the process requirements of different production lines, thereby achieving flexible adaptation of the detection method. Through the above steps, high-precision image data of the upper surface detection area can be efficiently acquired, laying a reliable data foundation for full-process automated black spot detection.

[0110] The second surface image acquisition process of the lower surface detection area is the same as that of the upper surface detection area. Based on the coordinates of the second driving side center point, the second operating side center point and the second middle center point, the second driving side, second operating side and second middle image acquisition ranges are defined with preset fixed side length values ​​respectively; the above ranges are scanned line by line by a line scan camera according to a preset scanning resolution, and the second driving side image, the second operating side image and the second middle image are generated synchronously, and dynamically associated with the physical coordinates of the lower surface detection area to ensure that the image data accurately maps the actual defect distribution on the lower surface of the strip.

[0111] In some examples, performing feature extraction on the first surface image to generate first black spot feature data corresponding to the first surface image includes:

[0112] Based on the first driving side image, the first operating side image, and the first central image, adaptive light correction and distortion correction are performed respectively to generate corresponding first driving side correction image, first operating side correction image, and first central correction image;

[0113] Denoising the first driving-side corrected image, the first operating-side corrected image, and the first central corrected image based on a preset filtering algorithm to generate a first driving-side denoised image, a first operating-side denoised image, and a first central denoised image;

[0114] performing black dot region enhancement on the first driving side denoised image, the first operating side denoised image and the first central denoised image based on a contrast enhancement algorithm to generate a first driving side enhanced image, a first operating side enhanced image and a first central enhanced image;

[0115] Based on a preset segmentation rule, the first driving side enhanced image, the first operating side enhanced image, and the first central enhanced image are respectively segmented into a plurality of sub-images, wherein the size of each sub-image is smaller than a preset fixed side length value;

[0116] Extracting black spot candidate regions corresponding to the first driving side image, the first operating side image, and the first central image from the plurality of sub-images based on a size threshold, shape characteristics, and color difference of the black spots;

[0117] First black dot feature data is determined according to the grayscale distribution and edge gradient characteristics of the black dot candidate area, wherein the first black dot feature data includes black dot position, black dot size and black dot shape parameters.

[0118] Exemplarily, adaptive light correction and distortion correction are performed based on the first driving side image, the first operating side image and the first central image, respectively. Adaptive light correction adjusts the global brightness distribution of the image through a histogram equalization algorithm, and adopts the local contrast limited adaptive histogram equalization (CLAHE) technology to eliminate local illumination unevenness. For distortion correction, based on the pre-calibrated line scan camera intrinsic parameter matrix (including focal length, principal point coordinates and radial distortion coefficient), the image is geometrically corrected through a perspective transformation model to generate a first driving side correction image, a first operating side correction image and a first central correction image. The corrected image eliminates pixel offset caused by shooting angle or lens distortion, ensuring the geometric accuracy of subsequent processing.

[0119] A preset filtering algorithm is used to suppress noise in the rectified images. Specifically, spatial filtering is performed on the first driven-side rectified image, the first operating-side rectified image, and the first central rectified image using a median filter with a preset window size of 5×5 pixels. By replacing pixel values ​​with the median within a neighborhood window, median filtering effectively removes salt-and-pepper noise and isolated noise points while preserving black point edge details. After filtering, the first driven-side denoised image, the first operating-side denoised image, and the first central denoised image are generated, providing clean input data for subsequent contrast enhancement.

[0120] The denoised image is enhanced for black spots based on a contrast enhancement algorithm. Using CLAHE technology, the image is divided into multiple 8×8 pixel sub-blocks. Histogram equalization is performed independently on each sub-block, and a preset contrast limit threshold is used to prevent noise amplification. Gamma correction is also used to enhance the contrast between dark spots and the bright background, generating the first driving side enhanced image, the first operating side enhanced image, and the first central enhanced image. In the enhanced images, the grayscale values ​​of the black spots are significantly lower than the background, and the edge gradient features are more pronounced.

[0121] The enhanced image is divided into multiple sub-images according to a preset segmentation rule. This segmentation rule defines each enhanced image (e.g., 500×500 pixels) as 25 sub-images (each 100×100 pixels), with each sub-image smaller than a preset fixed side length (500 pixels for 10 cm). Segmentation is performed by sliding along the rows and columns of the image with a fixed step size, without overlapping segments. The segmented sub-images are stored independently, and their positions within the original image are recorded using coordinate labels to ensure the integrity of the black spot location information.

[0122] Based on the size threshold (such as diameter ≥ 0.5mm), shape characteristics (circularity ≥ 0.7) and color difference (grayscale value ≤ 50) of the black spots, candidate black spot areas are extracted from the sub-images. Specifically, all sub-images are traversed through the connected domain analysis algorithm, and connected areas that meet the preset conditions are screened out as candidate black spots. For each candidate area, the center coordinates of its circumscribed rectangle are calculated as the black spot position, the diagonal length of the circumscribed rectangle is used as the black spot size, and its shape parameters are described by the Hu moment feature. This step outputs the candidate black spot set corresponding to the first driving side image, the first operating side image, and the first central image.

[0123] The first black spot feature data is generated based on the grayscale distribution and edge gradient characteristics of the candidate black spot area. For each candidate area, the average grayscale value, grayscale standard deviation and gradient amplitude histogram of its internal pixels are extracted; at the same time, the edge contour is extracted using the Canny edge detection algorithm, and the average gradient direction and gradient intensity of the edge pixels are calculated. Finally, the black spot position (physical coordinates of the strip surface), black spot size (physical length) and black spot shape parameters (circularity, aspect ratio) are integrated to form structured feature data as input to the deep learning model. The feature data is strictly associated with the coordinates of the strip detection area, supporting the subsequent accurate statistics of the number of black spots and quality judgment.

[0124] When extracting features from the second driving side image, second operating side image, and second central image of the lower surface inspection area, the same processing flow as that of the upper surface inspection area is adopted. First, adaptive light correction and distortion correction are used to eliminate uneven illumination and geometric distortion to generate the second driving side corrected image, the second operating side corrected image, and the second central corrected image. Subsequently, noise removal is performed based on a preset filtering algorithm to generate the corresponding denoised image. The contrast of the black spot area is then enhanced using the CLAHE algorithm to generate the second driving side enhanced image, the second operating side enhanced image, and the second central enhanced image. The enhanced image is further divided into multiple sub-images according to preset segmentation rules. Candidate areas are screened based on the same black spot size threshold, shape characteristics, and color differences. Edge gradient analysis and grayscale distribution calculation are used to generate second black spot feature data containing physical coordinates, size, and morphological characteristics to ensure strict consistency with the upper surface processing logic, parameters, and data format, supporting the simultaneous and accurate determination of double-sided defects.

[0125] In some examples, identifying the first black dot feature data and the second black dot feature data based on a deep learning model, and counting the number of first black dots in the upper surface detection area and the number of second black dots in the lower surface detection area, includes:

[0126] Based on the first black spot feature data and the second black spot feature data, feature matching is performed on the black spot candidate area through a deep learning model to determine the confidence parameters of the candidate black spots;

[0127] Based on the comparison result between the confidence parameter and the preset confidence threshold, the valid black spots are screened out;

[0128] Based on the positions and sizes of the valid black spots, the valid black spots in the upper surface detection area and the lower surface detection area are deduplicated to determine the target black spot position set;

[0129] Based on the number of black dots in the target black dot position set, a first number of black dots in the upper surface detection area and a second number of black dots in the lower surface detection area are determined.

[0130] Exemplarily, based on the first black spot feature data and the second black spot feature data, feature matching and classification are performed on the black spot candidate areas through a pre-trained deep learning model. The deep learning model adopts a convolutional neural network (CNN) architecture, and the input is the position, size, shape parameters and grayscale gradient features of the candidate area. The model extracts high-order features through multi-layer convolution and pooling operations, and outputs a confidence parameter (ranging from 0 to 1) that each candidate area is a real black spot. For example, the confidence parameter is calculated by normalization of the Softmax function, reflecting the probability that the candidate area belongs to a black spot defect. A labeled black spot dataset (marking the real black spot position and category) is used for model training, and the weight parameters are optimized through cross-validation to ensure sensitivity to minor defects and robustness to noise interference.

[0131] Based on a preset confidence threshold (such as 0.85), the confidence parameters of all candidate areas are screened. Specifically, if the confidence parameter of a candidate area is greater than or equal to the preset confidence threshold, it is determined to be a valid black spot; if it is less than the preset confidence threshold, it is marked as noise or a false positive and eliminated. For example, for the set of candidate black spots in the upper surface detection area, after threshold comparison, the candidate areas with confidence greater than or equal to the preset confidence threshold of 0.8 are retained to generate the set of valid black spots on the upper surface; the lower surface detection area is screened using the same threshold to generate the set of valid black spots on the lower surface. This step significantly reduces the false detection rate through threshold filtering, ensuring that subsequent statistical results only contain high-confidence defects.

[0132] The valid black dots in the first driving side detection area, the first central detection area, and the first operating side detection area on the upper surface are deduplicated separately. Based on the physical position and size of the black dots, the Euclidean distance algorithm is used to calculate the distance between adjacent black dots in the same detection area. If the distance is less than a preset deduplication threshold (for example, 3mm) and the size difference is less than or equal to 10%, they are considered to be the same black dot, and the one with the higher confidence level is retained. After deduplication, the number of black dots in the three sub-areas is counted and accumulated to obtain the first number of black dots, where the number of black dots in the first driving side detection area, the number of black dots in the first central detection area, and the number of black dots in the first operating side detection area are recorded as N1_DS, N1_CT, and N1_OS, respectively. The same operation is performed on the second driving side detection area, the second central detection area, and the second operating side detection area on the lower surface, and the number of black dots in each sub-area is counted and accumulated to obtain the second number of black dots, where the number of black dots in the second driving side detection area, the number of black dots in the second central detection area, and the number of black dots in the second operating side detection area are recorded as N2_DS, N2_CT, and N2_OS, respectively.

[0133] Based on the target black spot position set, the number of black spots in the upper surface detection area and the lower surface detection area are counted separately. Specifically, each black spot in the target black spot position set is traversed, and the total number of black spots in the corresponding area is accumulated according to the detection area label to which it belongs. For example, if the target set contains 50 black spots, of which 30 are labeled as upper surface and 20 are labeled as lower surface, then the first number of black spots N1 output is 30 and the second number of black spots N2 is 20. The statistical results are associated with the coordinate information of the strip inspection area and stored as input for quality grade determination. This step ensures that the number of black spots is consistent with the actual defect distribution through precise spatial mapping and logical counting, supporting the reliable execution of subsequent stratification threshold determination rules.

[0134] In some examples, the preset black spot threshold includes a first preset threshold and a second preset threshold, and determining the quality grade of the steel strip based on the first black spot number, the second black spot number, and the preset black spot threshold includes:

[0135] determining a quality state of the upper surface of the upper surface detection area based on a comparison result of the first black spot quantity and a first preset threshold;

[0136] determining a lower surface quality state of the lower surface detection area based on a comparison result of the second black spot quantity and a second preset threshold;

[0137] The quality grade of the strip is determined based on the logical combination relationship between the upper surface quality status and the lower surface quality status.

[0138] Exemplarily, in the quality grade determination process, first, independent preset black dot thresholds are set for each sub-area of ​​the upper surface and lower surface detection areas. For the upper surface detection area, the first drive side detection area threshold A_DS, the first middle detection area threshold A_CT, and the first operating side detection area threshold A_OS are preset, for example, A_DS=5, A_CT=3, A_OS=5, indicating that the maximum number of black dots allowed in the drive side and operating side sub-areas is 5, and the maximum number of black dots allowed in the middle sub-area is 3. Similarly, the second drive side detection area threshold B_DS, the second middle detection area threshold B_CT, and the second operating side detection area threshold B_OS are preset for the lower surface detection area, for example, B_DS=5, B_CT=3, B_OS=5. The thresholds of each sub-area are independently configured according to the actual process requirements to adapt to the sensitivity of different areas to defects.

[0139] Based on the statistically obtained number of black spots in each sub-area (upper surface: N1_DS, N1_CT, N1_OS; lower surface: N2_DS, N2_CT, N2_OS), the sub-area quality status judgment is performed separately. For the upper surface detection area, if the number of black spots N1_DS in the first drive side detection area is less than or equal to A_DS, the sub-area is judged to be qualified; if N1_DS exceeds A_DS, it is marked as unqualified. Similarly, the first middle detection area and the first operating side detection area are judged based on the comparison results of N1_CT and A_CT, and N1_OS and A_OS respectively. If all sub-areas of the upper surface meet the threshold requirements, the overall quality status of the upper surface is marked as "qualified"; if any sub-area exceeds the limit, the overall status of the upper surface is marked as "unqualified". The lower surface detection area uses the same logic, and the overall status of the lower surface is judged as "qualified" or "unqualified" by comparing N2_DS, N2_CT, N2_OS with B_DS, B_CT, and B_OS.

[0140] The overall quality grade of the strip is further determined based on the combined quality status of the upper and lower surfaces. If the number of black spots in all sub-areas of the upper and lower surfaces is within the limit (i.e., N1_DS ≤ A_DS, N1_CT ≤ A_CT, N1_OS ≤ A_OS, and N2_DS ≤ B_DS, N2_CT ≤ B_CT, N2_OS ≤ B_OS), the strip quality grade is determined to be "Level 1 qualified." If one sub-area on the upper or lower surface exceeds the limit (for example, N1_DS = 6 and A_DS = 5 on the upper surface, but all other sub-areas and the lower surface are qualified), the strip is determined to be "Level 2 warning." If at least one sub-area on each of the upper and lower surfaces exceeds the limit, or if two or more sub-areas on the same surface exceed the limit (for example, N1_DS = 6 and N1_CT = 4 on the upper surface, or N1_DS = 6 on the upper surface and N2_CT = 4 on the lower surface), the strip is determined to be "Level 3 unqualified."

[0141] The final quality grade result is associated with the strip number, detection time, and the number of black spots in each sub-area and stored in the database, and transmitted to the production control system in real time. For example, the output content includes the quality grade (such as "Secondary Warning"), the location of the sub-area that exceeds the limit (such as "Unqualified upper surface drive side: N1_DS=6 / A_DS=5") and detailed statistical data. Through independent judgment of each area and hierarchical combination rules, this application accurately quantifies the impact of local defects on overall quality, provides high-granularity data support for process optimization and product downgrade decisions, and ensures the comprehensiveness and traceability of the judgment logic. Please refer to Figure 2 , is a schematic diagram of the detection results of black spots on a galvanized automobile sheet provided in an embodiment of the present application. The detection results of the first driving side detection area, the first middle detection area, the first operating side detection area, the second driving side detection area, the second middle detection area and the second operating side detection area are all qualified.

[0142] See also Figure 3 , is a schematic structural diagram of a device for detecting black spots on galvanized automobile sheets provided in an embodiment of the present application, comprising:

[0143] Specification information acquisition unit 21, used to obtain specification information of the strip;

[0144] The detection area determination unit 22 determines an upper surface detection area and a lower surface detection area of ​​the steel strip based on the specification information and the preset edge offset, wherein the upper surface detection area includes a first drive side detection area, a first middle detection area, and a first operating side detection area, and the lower surface detection area includes a second drive side detection area, a second middle detection area, and a second operating side detection area;

[0145] a surface image acquisition unit 23 for acquiring a first surface image of an upper surface detection area and a second surface image of a lower surface detection area;

[0146] A black spot feature generating unit 24 is configured to extract features from the first surface image and the second surface image to generate first black spot feature data corresponding to the first surface image and second black spot feature data corresponding to the second surface image;

[0147] a black dot number counting unit 25 , which identifies the first black dot feature data and the second black dot feature data based on a deep learning model, and counts the first black dot number in the upper surface detection area and the second black dot number in the lower surface detection area;

[0148] The quality grade determining unit 26 determines the quality grade of the steel strip based on the first number of black spots, the second number of black spots, and a preset black spot threshold.

[0149] See also Figure 4An embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any method for detecting black spots on galvanized automobile sheets are implemented.

[0150] Since the electronic device introduced in this embodiment is the equipment used to implement a device for detecting black spots on galvanized automobile plates in the embodiment of this application, based on the method introduced in the embodiment of this application, technical personnel in this field can understand the specific implementation of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of this application will not be introduced in detail here. As long as the equipment used by technical personnel in this field to implement the method in the embodiment of this application falls within the scope of protection of this application.

[0151] During the specific implementation process, when the computer program 311 is executed by the processor, any implementation method of the embodiments corresponding to the first aspect can be implemented.

[0152] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0153] Those skilled in the art will appreciate that the embodiments of the present application may provide methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-readable program code.

[0154] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0155] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0156] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0157] The present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes Figure 1 The process of a method for detecting black spots on galvanized automobile sheets in the corresponding embodiment.

[0158] A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium can be a magnetic medium, an optical medium or a semiconductor medium, etc.

[0159] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0160] In the several embodiments provided in this application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0161] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0162] In addition, the functional units in the various embodiments of the present application may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The above-mentioned integrated units may be implemented in the form of hardware and / or software functional units.

[0163] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory, a magnetic disk, or an optical disk.

[0164] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

[0165] Although the preferred embodiments of this specification have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of this specification.

[0166] Obviously, those skilled in the art may make various changes and modifications to this specification without departing from the spirit and scope of this specification. Thus, if such changes and modifications fall within the scope of the claims of this specification and their equivalents, this specification is intended to include such changes and modifications.

Claims

1. A method for detecting black spots on galvanized automobile sheets, characterized in that: include: Obtain the specification information of the strip steel; Based on the specification information and the preset edge offset, determining an upper surface detection area and a lower surface detection area of ​​the strip, wherein the upper surface detection area includes a first driving side detection area, a first middle detection area, and a first operating side detection area, and the lower surface detection area includes a second driving side detection area, a second middle detection area, and a second operating side detection area; Acquire a first surface image of the upper surface detection area and a second surface image of the lower surface detection area; performing feature extraction on the first surface image and the second surface image to generate first black dot feature data corresponding to the first surface image and second black dot feature data corresponding to the second surface image; Identify the first black dot feature data and the second black dot feature data based on a deep learning model, and count the number of first black dots in the upper surface detection area and the number of second black dots in the lower surface detection area; The quality grade of the steel strip is determined based on the first black spot number, the second black spot number, and a preset black spot threshold.

2. The method according to claim 1, characterized in that The determining of the upper surface detection area of ​​the steel strip based on the specification information and the preset edge offset includes: Based on the strip length and the preset edge offset of the specification information, the upper surface is divided into a first driving side area, a first operating side area, and a first middle area along the length direction, wherein the preset edge offset includes a driving side offset and an operating side offset; In the first driving side area, the first driving side detection area is selected along the length direction based on a preset fixed side length value, and the first driving side detection area includes the first driving side center point; In the first operating side area, the first operating side detection area is selected along the length direction based on the preset fixed side length value, and the first operating side detection area includes the first operating side center point, wherein the first driving side center point and the first operating side center point are located on the same horizontal straight line; determining a first central point of the first central detection area based on the first driving side central point, the first operating side central point, and the first central area; Based on the first central center point, the first central detection area is determined.

3. The method according to claim 2, characterized in that The strip length and the preset edge offset based on the specification information are used to divide the upper surface into a first driving side area, a first operating side area, and a first middle area along the length direction, including: Based on the strip length and the drive side offset, sequentially determining a drive side starting position and a drive side ending position of a first drive side region along a first direction; determining a first driving side region of the upper surface based on the driving side starting position and the driving side ending position; Based on the strip length and the operating side offset, sequentially determining an operating side starting position and an operating side ending position of a first operating side region along a second direction opposite to the first direction; determining a first operation side area of ​​the upper surface based on the operation side starting position and the operation side ending position; Based on the driving side end position and the operating side end position, a first middle area of ​​the upper surface is determined, wherein the middle starting position of the first middle area is the driving side end position, and the middle end position is the operating side end position.

4. The method according to claim 2, characterized in that The first surface image includes a first driving side image, a first operating side image, and a first middle image. Acquiring the first surface image of the upper surface detection area includes: Determining a first driving side image acquisition range of the first driving side detection area based on the first driving side center point and the preset fixed side length value; Determining a first operation side image acquisition range of the first operation side detection area based on the first operation side center point and the preset fixed side length value; Determining a first central image acquisition range of the first central detection area based on the first central center point and the preset fixed side length value; Based on a preset scanning resolution, the first driving side image acquisition range, the first operating side image acquisition range and the first central image acquisition range are scanned line by line by a line scanning camera to generate the first driving side image, the first operating side image and the first central image.

5. The method according to claim 4, characterized in that Extracting features from the first surface image to generate first black dot feature data corresponding to the first surface image includes: Based on the first driving side image, the first operating side image, and the first central image, adaptive light correction and distortion correction are performed respectively to generate corresponding first driving side corrected image, first operating side corrected image, and first central corrected image; performing noise removal on the first driving-side corrected image, the first operating-side corrected image, and the first central corrected image based on a preset filtering algorithm to generate a first driving-side denoised image, a first operating-side denoised image, and a first central denoised image; performing black dot region enhancement on the first driving side denoised image, the first operating side denoised image and the first central denoised image based on a contrast enhancement algorithm to generate a first driving side enhanced image, a first operating side enhanced image and a first central enhanced image; Based on a preset segmentation rule, the first driving-side enhanced image, the first operating-side enhanced image, and the first central enhanced image are respectively segmented into a plurality of sub-images, wherein a size of each sub-image is smaller than the preset fixed side length value; extracting black dot candidate regions corresponding to the first driving side image, the first operating side image, and the first central image from the multiple sub-images based on a size threshold, shape characteristics, and color difference of the black dots; First black dot feature data is determined according to the grayscale distribution and edge gradient characteristics of the black dot candidate area, wherein the first black dot feature data includes black dot position, black dot size and black dot shape parameters.

6. The method according to claim 5, characterized in that The identifying the first black dot feature data and the second black dot feature data based on the deep learning model, and counting the number of first black dots in the upper surface detection area and the number of second black dots in the lower surface detection area, includes: Based on the first black spot feature data and the second black spot feature data, performing feature matching on the black spot candidate area through a deep learning model to determine the confidence parameters of the candidate black spots; Screening out valid black spots based on a comparison result between the confidence parameter and a preset confidence threshold; Based on the positions and sizes of the valid black dots, deduplication is performed on the valid black dots in the upper surface detection area and the lower surface detection area to determine a target black dot position set; Based on the number of black dots in the target black dot position set, a first number of black dots in the upper surface detection area and a second number of black dots in the lower surface detection area are determined.

7. The method according to claim 1, characterized in that The preset black spot threshold includes a first preset threshold and a second preset threshold, and determining the quality grade of the steel strip based on the first black spot number, the second black spot number, and the preset black spot threshold includes: determining a quality state of the upper surface of the upper surface detection area based on a comparison result of the first black spot quantity and the first preset threshold; determining a lower surface quality state of the lower surface detection area based on a comparison result of the second black spot quantity and the second preset threshold; The quality grade of the steel strip is determined based on a logical combination relationship between the upper surface quality status and the lower surface quality status.

8. A device for detecting black spots on galvanized automobile sheets, characterized in that: include: A specification information acquisition unit, used to acquire specification information of the strip steel; a detection area determination unit, configured to determine an upper surface detection area and a lower surface detection area of ​​the steel strip based on the specification information and a preset edge offset, wherein the upper surface detection area includes a first drive side detection area, a first middle detection area, and a first operating side detection area, and the lower surface detection area includes a second drive side detection area, a second middle detection area, and a second operating side detection area; a surface image acquisition unit, configured to acquire a first surface image of the upper surface detection area and a second surface image of the lower surface detection area; a black spot feature generating unit, configured to extract features from the first surface image and the second surface image, and generate first black spot feature data corresponding to the first surface image and second black spot feature data corresponding to the second surface image; a black dot number counting unit, which identifies the first black dot feature data and the second black dot feature data based on a deep learning model, and counts the first black dot number in the upper surface detection area and the second black dot number in the lower surface detection area; The quality grade determining unit determines the quality grade of the steel strip based on the first black spot number, the second black spot number and a preset black spot threshold.

9. An electronic device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the method for detecting black spots on galvanized automobile sheets as claimed in any one of claims 1 to 7 when executing the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for detecting black spots on galvanized automobile sheets according to any one of claims 1 to 7 is implemented.

Images