A method, system and storage medium for detecting defects of grate bars of a sintering machine trolley

Abstract

The present invention relates to a method, system and storage medium for detecting defects in grate bars of a sintering machine trolley. The method comprises the following steps: obtaining an image of grate bars of a sintering machine trolley; cropping the image to obtain a feature map containing a row of grate bars; detecting whether the grate bars to be tested in the feature map have defects according to a darkness mechanism; if defects exist, drawing the detected defects in the original image. The method can realize real-time detection of defects in the grate bars of the sintering trolley, and no manual intervention is required during the detection process, which effectively solves the hysteresis existing in manual detection of grate bar defects and avoids the safety risks of manual detection. When there are defects in the grate bars, they can be detected in time. By displaying the defects in the original image, maintenance personnel can accurately locate the position of the grate bar defects and perform timely maintenance on the grate bar defects.

Description

Technical Field

[0001] The present application relates to the field of sintering car status monitoring and inspection in an ironmaking plant, and specifically to a sintering machine car grate bar defect detection method, system and storage medium. Background Art

[0002] Grate bars are one of the important parts of the sintering machine trolley in an ironmaking plant. As a material carrying device, the service life of the grate bars is an important factor affecting the operating costs and productivity of the sintering machine trolley.

[0003] During operation, the grates of the sintering machine trolley will be subjected to static pressure, and will be affected by temperature changes and corrosion from high-temperature sulfidation gases for a long time, which will cause the strength to decrease, break, and fall off. If the defects are not repaired in time, they will continue to expand. When large holes appear in the grates, the sealing of the trolley will decrease, causing air to enter the trolley, which will directly affect the quantity and quality of the sintered materials. The fallen grates and materials may cause a sudden shutdown of the machine and cause safety accidents.

[0004] At present, the working status of grates in sintering production is generally detected by manual periodic inspection, which has obvious lag and danger. By the time the defects are discovered, they may have become large. Therefore, it is of great significance to realize automatic detection of defective parts of sintering machine trolleys. Summary of the invention

[0005] In view of the above problems, the present application provides a method, system and storage medium for detecting defects in grates of a sintering machine trolley, which solve the problem that the existing method of manual periodic inspection of grates in sintering machine trolleys has obvious lag and danger.

[0006] To achieve the above object, the inventor provides a method for detecting defects of grate bars of a sintering machine, comprising the following steps:

[0007] Acquire an image of grates of a sintering machine carriage, wherein the image includes a plurality of rows of grates;

[0008] Crop the image to obtain a feature map containing a row of grate bars;

[0009] Defect detection is performed on the grate bars to be tested in the characteristic image according to the darkness mechanism;

[0010] When a defect is detected, the detected defect is drawn in the original image.

[0011] In some embodiments, the step of cropping the image further includes the following steps:

[0012] Perform image correction on the acquired images.

[0013] In some embodiments, the image correction of the acquired image specifically comprises the following steps:

[0014] The obtained image is subjected to image rotation, barrel correction and erosion filling.

[0015] In some embodiments, cropping the image specifically includes the following steps:

[0016] Perform perspective transformation on the image according to the prior value;

[0017] The perspective-transformed image is cropped to remove the preset number of rows of bars above and below and the gaps on the left and right sides;

[0018] Perform edge detection, identify the upper and lower boundaries of the grate bars, and perform upper and lower cropping to keep only the middle row of grate bars to be identified;

[0019] Then perform edge detection and line detection to determine the last corrected transformation coordinates and left and right boundaries, and perform corresponding affine transformation and left and right cropping.

[0020] In some embodiments, the edge detection is performed using a canny edge detection algorithm.

[0021] In some embodiments, the line detection is performed by using a Hough line detection algorithm.

[0022] In some embodiments, the defect detection of the grate bars to be tested in the characteristic image according to the darkness mechanism specifically includes the following steps:

[0023] Traverse the grayscale values ​​of a single column of the feature map and store them, set a vertical sliding window, and add all the grayscale values ​​in the window, filter all windows in a single column whose darkness is lower than the first threshold and mark this column as a possible defect column;

[0024] Merge the continuous possible defect columns and record them as possible defect areas;

[0025] Determining whether the darkness of the possible defective area is lower than a second threshold;

[0026] If it is lower than the second threshold, the possible defective area is determined to be a defective area.

[0027] In some embodiments, drawing the detected defects in the original image specifically comprises the following steps:

[0028] Initialize two feature maps of the same size as the original image as storage images, storing the horizontal and vertical coordinates of the pixels respectively;

[0029] For the same image, each time a defect is detected, the counter is incremented by 1, and the defect coordinates of the defect corresponding to the current feature map are recorded;

[0030] Input the recorded defect coordinates into the stored image, obtain the corresponding horizontal and vertical coordinates in the original image, and establish a mapping relationship between the defect coordinates and the coordinates of the original image;

[0031] The detected defects are drawn on the original image according to the mapping relationship.

[0032] Another technical solution is also provided, a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method for detecting defects in grate bars of a sintering machine trolley are executed.

[0033] Another technical solution is provided, a sintering machine trolley grate bar defect detection system, including a camera and a computer;

[0034] The camera is connected to the computer, and is used to collect images of the grate bars of the sintering machine trolley;

[0035] The computer includes a storage medium and a processor, the storage medium is the storage medium mentioned above, and the processor is used to process the computer program in the storage medium.

[0036] Different from the prior art, the above technical solution obtains an image of the grate bars of the sintering machine trolley. The image obtained contains several rows of grate bars. By cropping the image, the part of the grate bars that are distorted due to shooting is cut off to obtain a feature map containing a row of grate bars. Then, the grate bars to be tested in the feature map are checked for defects according to the darkness mechanism. When defects are detected, the detected defects are drawn in the corresponding position of the original image. The defects of the grate bars of the sintering trolley can be detected in real time. No manual intervention is required during the detection process, which effectively solves the lag in manual detection of grate bar defects and avoids the safety risks of manual detection. When there are defects in the grate bars, they can be detected in time. By displaying the defects in the original image, maintenance personnel can accurately locate the position of the grate bar defects and perform timely maintenance on the defects of the grate bars. In particular, the judgment of defects by darkness reduces the influence of different lighting conditions on the detection and improves the accuracy of the detection results.

[0037] The above-mentioned records related to the invention content are only an overview of the technical solution of the present application. In order to enable ordinary technicians in the field to more clearly understand the technical solution of the present application, and then implement it according to the text of the specification and the contents recorded in the drawings, and to make the above-mentioned purpose and other purposes, features and advantages of the present application easier to understand, the following is an explanation in combination with the specific implementation mode and drawings of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The drawings are only used to illustrate the principles, implementation methods, applications, characteristics and effects of the specific embodiments of the present application and other related contents, and shall not be considered as limitations of the present application.

[0039] In the drawings of the specification:

[0040] Figure 1 A schematic flow chart of a method for detecting defects in sintering machine trolley grates according to a specific implementation method;

[0041] Figure 2 A schematic diagram of a process of performing image correction on an acquired image according to a specific implementation method;

[0042] Figure 3 A schematic diagram of a process of cropping an image according to a specific implementation method;

[0043] Figure 4 A schematic diagram of a process of detecting defects of grate bars according to a characteristic diagram as described in a specific implementation method;

[0044] Figure 5 Another schematic flow chart of the method for detecting defects of sintering machine trolley grates according to the specific implementation method;

[0045] Figure 6 A schematic diagram of the structure of the storage medium described in the specific implementation method;

[0046] Figure 7 A structural schematic diagram of a sintering machine trolley grate defect detection system according to a specific implementation method.

[0047] The reference numerals in the above drawings are described as follows:

[0048] 600, Computer,

[0049] 610. Storage medium,

[0050] 620, processor,

[0051] 700. Camera. DETAILED DESCRIPTION

[0052] In order to explain in detail the possible application scenarios, technical principles, specific schemes that can be implemented, and the purposes and effects that can be achieved, the following is a detailed description of the specific embodiments listed in conjunction with the accompanying drawings. The embodiments described herein are only used to more clearly illustrate the technical solutions of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.

[0053] Reference to "embodiment" herein means that the specific features, structures or characteristics described in conjunction with the embodiment may be included in at least one embodiment of the present application. The term "embodiment" appearing in various places in the specification does not necessarily refer to the same embodiment, nor does it particularly limit its independence or association with other embodiments. In principle, in the present application, as long as there is no technical contradiction or conflict, the various technical features mentioned in the embodiments can be combined in any way to form a corresponding implementable technical solution.

[0054] Unless otherwise defined, the technical terms used in this document have the same meanings as those generally understood by those skilled in the art to which this application belongs; the use of relevant terms in this document is only for describing specific embodiments and is not intended to limit this application.

[0055] In the description of this application, the term "and / or" is an expression used to describe the logical relationship between objects, indicating that three relationships may exist, for example, A and / or B, which means: A exists, B exists, and A and B exist at the same time. In addition, the character " / " in this article generally indicates that the objects before and after are in an "or" logical relationship.

[0056] In the present application, terms such as “first” and “second” are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship of quantity, priority or sequence between these entities or operations.

[0057] Without further limitations, in this application, the words "include", "comprises", "has" or other similar expressions used in the sentences are intended to cover non-exclusive inclusion. These expressions do not exclude the presence of additional elements in the process, method or product including the elements, so that the process, method or product including a series of elements may include not only those limited elements, but also other elements not explicitly listed, or also include elements inherent to such process, method or product.

[0058] Similar to the understanding in the Examination Guidelines, in this application, expressions such as "greater than", "less than", "exceed" and the like are understood to exclude the number itself; expressions such as "above", "below", "within" and the like are understood to include the number itself. In addition, in the description of the embodiments of this application, "multiple" means more than two (including two), and similar expressions related to "multiple" are also understood in this way, such as "multiple groups", "multiple times", etc., unless otherwise clearly and specifically limited.

[0059] In the description of the embodiments of the present application, space-related expressions used, such as "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "vertical", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicate the orientation or position relationship based on the orientation or position relationship shown in the specific embodiments or drawings, and are only for the convenience of describing the specific embodiments of the present application or facilitating the reader's understanding, and do not indicate or imply that the referred device or component must have a specific position, a specific orientation, or be constructed or operated in a specific orientation, and therefore cannot be understood as a limitation on the embodiments of the present application.

[0060] Unless otherwise expressly specified or limited, in the description of the embodiments of the present application, the terms such as "install", "connect", "connect", "fix", "set", etc. used should be understood in a broad sense. For example, the "connection" can be a fixed connection, a detachable connection, or an integrated setting; it can be a mechanical connection, an electrical connection, or a communication connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For technicians in the technical field to which the present application belongs, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0061] See also Figure 1 This embodiment provides a method for detecting defects in grate bars of a sintering machine, comprising the following steps:

[0062] Step S110: acquiring an image of grate bars of a sintering machine trolley, wherein the image includes a plurality of rows of grate bars;

[0063] Step S120: cropping the image to obtain a feature map containing a row of grate bars;

[0064] Step S130: performing defect detection on the grate bars to be tested in the characteristic image according to the darkness mechanism;

[0065] Step S140: When a defect is detected, the detected defect is drawn in the original image.

[0066] By acquiring the image of the grate bars of the sintering machine trolley, the captured image contains several rows of grate bars. By cropping the image, the part of the grate bars that are distorted due to shooting is cut off to obtain a feature map containing a row of grate bars. Then, the grate bars to be tested in the feature map are checked for defects according to the darkness of the feature map. When defects are detected, the detected defects are drawn in the corresponding position of the original image, so that the defects of the grate bars of the sintering trolley can be detected in real time. No manual intervention is required during the detection process, which effectively solves the lag in manual detection of grate bar defects and avoids the safety risks of manual detection. When there are defects in the grate bars, they can be detected in time. By displaying the defects in the original image, maintenance personnel can accurately locate the position of the grate bar defects and perform timely maintenance on the grate bar defects. Among them, darkness refers to the ratio of the total grayscale value in the detection area to the total grayscale value of the image. Since images will have different grayscale values ​​under different lighting conditions, this embodiment determines whether it is a defect based on darkness, that is, the relative value of the grayscale rather than the absolute value is used as the basis for judgment, thereby avoiding misjudgment caused by the influence of lighting conditions on the grayscale value, and having higher accuracy in defect detection.

[0067] In some embodiments, the step of cropping the image further includes the following steps:

[0068] Perform image correction on the acquired images.

[0069] In order to avoid distortion of the acquired image due to camera problems when the camera is shooting the grate bars of the sintering trolley, the acquired image is corrected after the grate bars of the sintering trolley are acquired to avoid distortion of the acquired image and affect the subsequent detection of grate bar defects. Figure 2 The image correction of the acquired image specifically comprises the following steps:

[0070] S210: performing image rotation, barrel correction and erosion filling on the obtained image.

[0071] exist Figure 1 There is also a step S210 between step S110 and step S120, in which the image is corrected by rotating, barrel-shaped correction and erosion filling, wherein the barrel-shaped correction makes it easier to extract features in the image, and the erosion filling makes the features more obvious. This step is not cropped, so the current image still contains multiple rows of grate bars and the background around the grate bars.

[0072] In some embodiments, cropping the image specifically includes the following steps:

[0073] Perform perspective transformation on the image according to the prior value;

[0074] The perspective-transformed image is cropped to remove the preset number of rows of bars above and below and the gaps on the left and right sides;

[0075] Perform edge detection, identify the upper and lower boundaries of the grate bars, and perform upper and lower cropping to keep only the middle row of grate bars to be identified;

[0076] Then perform edge detection and line detection to determine the last corrected transformation coordinates and left and right boundaries, and perform corresponding affine transformation and left and right cropping.

[0077] See also Figure 3 In the process of cropping the image, the image is firstly transformed according to the prior value, and then the image is corrected. The perspective problem of the image being larger near and smaller far away is received through the perspective transformation. After that, it will be roughly cropped first, and the upper and lower rows of grate bars and the left and right gaps are cut off. The left and right gaps are the parts of the grate bars that do not belong to the grate bars, such as walls, on the left and right of the grate bars. In the process of cropping, this part is cut off to avoid its interference with the subsequent grate bar detection. Then edge detection is performed to identify the upper and lower boundaries, and the upper and lower cropping is performed, leaving only the middle row of grate bars to be identified. Subsequently, edge detection and line detection are continued to determine the transformation coordinates and left and right boundaries of the last correction, and the corresponding affine transformation and left and right cropping are performed to obtain the feature map, which is a rectangular grayscale image of a row of grate bars. Among them, edge detection is detected by the canny edge detection algorithm, which is suitable for different occasions. Its parameters allow adjustment according to the specific requirements of different implementations to identify different edge characteristics. In other embodiments, other edge detection algorithms may be used for edge detection, such as the Laplacian edge detection algorithm. The straight line detection is performed using the Hough straight line detection algorithm. The Hough straight line detection algorithm has strong anti-interference capabilities and is insensitive to noise in the image. In other embodiments, other methods may be used for straight line detection, such as the Freeman straight line detection algorithm and the looper crawling algorithm.

[0078] In some embodiments, the defect detection of the grate bars to be tested in the characteristic image according to the darkness mechanism specifically includes the following steps:

[0079] Traverse the grayscale values ​​of a single column of the feature map and store them, set a vertical sliding window, and add all the grayscale values ​​in the window, filter all windows in a single column whose darkness is lower than the first threshold and mark this column as a possible defect column;

[0080] Merge the continuous possible defect columns and record them as possible defect areas;

[0081] Determining whether the darkness of the possible defective area is lower than a second threshold;

[0082] If it is lower than the second threshold, the possible defective area is determined to be a defective area.

[0083] See also Figure 4 , when performing defect detection of grate bars in the feature map, first traverse the grayscale values ​​of a single column in the feature map and store them, then set a vertical sliding window, add all the grayscale values ​​in the window, get the total grayscale value in the window, filter all windows in the single column whose darkness is lower than the first threshold and mark this column, record it as a possible defect column. Among them, in the judgment of possible defect columns, the darkness of the filtered single column is the ratio of the total grayscale value in the window to the total grayscale value of the image. Then try to merge the marked columns. If multiple marked columns appear continuously, merge the vertical window height and the horizontal continuous columns to form a rectangular mark, which is the possible defect area. Finally judge whether the area is a defect area by the darkness of this area. When the darkness of the possible defect area is lower than the second threshold, the area is judged to be a defect area. Among them, the darkness of the possible defect area is the ratio between the total grayscale value in the possible defect area and the total grayscale value of the image. Among them, the first threshold is lower than the second threshold. As many suspected defects as possible are found through the first threshold, and then the defect area is found more accurately through the second threshold. By introducing darkness defects, the impact of different lighting conditions on the defect detection mechanism is reduced. The specific ranges of the first threshold and the second threshold can be set according to empirical values ​​or based on preset standard values.

[0084] In some embodiments, drawing the detected defects in the original image specifically comprises the following steps:

[0085] Initialize two feature maps of the same size as the original image as storage images, storing the horizontal and vertical coordinates of the pixels respectively;

[0086] For the same image, each time a defect is detected, the counter is incremented by 1, and the defect coordinates of the defect for the current feature map are recorded;

[0087] Input the recorded defect coordinates into the stored image, obtain the corresponding horizontal and vertical coordinates in the original image, and establish a mapping relationship between the defect coordinates and the coordinates of the original image;

[0088] The detected defects are drawn on the original image according to the mapping relationship.

[0089] For the same image, each time a defect is detected, the counter will be increased by 1, and the defect coordinates of the defect for the current feature map will be recorded. It is difficult to directly inversely transform the coordinates, so during the first perspective transformation, two feature maps of the same size as the original image are initialized as storage images, where one storage image stores the horizontal coordinates of the pixels, and the other storage image stores the vertical coordinates of the pixels. After that, after each transformation, the mapping relationship between the original image and the new image can be found through the horizontal and vertical coordinates of the pixels stored in the two storage images. After defect detection, the defect coordinates are input into the storage image, and the corresponding horizontal and vertical coordinates before the transformation can be obtained. The mapping relationship is established to complete the inverse transformation of the coordinates. At this point, all detected defects will be drawn on the original image, which is convenient for the staff to locate and process.

[0090] like Figure 5 As shown, in some embodiments, a method for detecting defects in grate bars of a sintering machine trolley is provided, by performing barrel correction (rotating image, barrel correction, corrosion filling, etc.), image interception and re-correction (perspective transformation, canny edge detection, Hough line detection, image cropping, etc.) on the original camera image, defect search (traversing pixels, darkness threshold comparison, etc.) and coordinate inverse transformation (auxiliary coordinate diagram), the detected defects are drawn on the original image and saved. The operating status of the trolley grate bars of the ironmaking plant is detected in real time, and the camera equipment is used as the data acquisition equipment, supplemented by light fill light. The above method is used to analyze the grate bar features in the image, and finally the location of the grate bar detachment and the estimation of the detachment degree are realized, and mapped to the corresponding trolley number, and the detection process does not require manual intervention. The real-time and accuracy of grate bar defect detection are effectively solved, the safety risks of manual detection are avoided, and the utilization efficiency of the trolley equipment is improved.

[0091] See also Figure 6 A storage medium 610 stores a computer program, and when the computer program is executed by a processor 620, the steps in the above-mentioned method for detecting defects in a sintering machine trolley grate are executed:

[0092] Acquire an image of grates of a sintering machine carriage, wherein the image includes a plurality of rows of grates;

[0093] Crop the image to obtain a feature map containing a row of grate bars;

[0094] Check whether the grate bars to be tested in the characteristic image have defects according to the darkness mechanism;

[0095] When a defect is detected, the detected defect is drawn in the original image.

[0096] By acquiring the image of the grate bars of the sintering machine trolley, the image obtained contains several rows of grate bars. By cropping the image, the part of the grate bars in the image that is distorted due to shooting is cut off to obtain a feature map containing a row of grate bars. Then, the grate bars to be tested in the feature map are checked for defects according to the darkness mechanism. When defects are detected, the detected defects are drawn in the corresponding position of the original image, which can realize real-time detection of defects in the grate bars of the sintering trolley. No manual intervention is required during the detection process, which effectively solves the lag in manual detection of grate bar defects and avoids the safety risks of manual detection. When there are defects in the grate bars, they can be detected in time. By displaying the defects in the original image, maintenance personnel can accurately locate the position of the grate bar defects and perform timely maintenance on the grate bar defects. Among them, darkness refers to the ratio of the total grayscale value in the detection area to the total grayscale value of the image. Since images will have different grayscale values ​​under different lighting conditions, this embodiment determines whether it is a defect based on darkness, that is, the relative value of the grayscale rather than the absolute value is used as the basis for judgment, thereby avoiding misjudgment caused by the influence of lighting conditions on the grayscale value, and having higher accuracy in defect detection.

[0097] In some embodiments, the step of cropping the image further includes the following steps:

[0098] Perform image correction on the acquired images.

[0099] In order to avoid distortion of the acquired image due to camera problems when the camera is shooting the grate bars of the sintering trolley, after acquiring the image of the grate bars of the sintering trolley, image correction is performed on the acquired image to avoid distortion of the acquired image and affect the subsequent detection of grate bar defects. The image correction of the acquired image specifically includes the following steps:

[0100] The obtained image is subjected to image rotation, barrel correction and erosion filling.

[0101] The image is corrected by rotating, barreling and corroding the image. The barreling makes the features in the image easier to extract, while the corroding makes the features more obvious. This step is not cropped, so the current image still contains multiple rows of grate bars and the background around the grate bars.

[0102] In some embodiments, cropping the image specifically includes the following steps:

[0103] Perform perspective transformation on the image according to the prior value;

[0104] The perspective-transformed image is cropped to remove the preset number of rows of bars above and below and the gaps on the left and right sides;

[0105] Perform edge detection, identify the upper and lower boundaries of the grate bars, and perform upper and lower cropping to keep only the middle row of grate bars to be identified;

[0106] Then perform edge detection and line detection to determine the last corrected transformation coordinates and left and right boundaries, and perform corresponding affine transformation and left and right cropping.

[0107] In the process of cropping the image, the image is firstly transformed in perspective according to the prior value, and then the image is corrected. The perspective problem of the image being larger near and smaller far away is received through the perspective transformation. After that, it will be roughly cropped first, and the upper and lower rows of grate bars and the left and right gaps will be cut off. The left and right gaps are the parts of the grate bars that do not belong to the grate bars, such as walls, on the left and right of the grate bars. In the process of cropping, this part is cut off to avoid its interference with the subsequent grate bar detection. Then edge detection is performed to identify the upper and lower boundaries, and the upper and lower cropping is performed, leaving only the middle row of grate bars to be identified. Subsequently, edge detection and line detection are continued to determine the transformation coordinates and left and right boundaries of the last correction, and the corresponding affine transformation and left and right cropping are performed to obtain the feature map, which is a rectangular grayscale image of a row of grate bars. Among them, edge detection is detected by the canny edge detection algorithm, which is suitable for different occasions. Its parameters allow adjustment according to the specific requirements of different implementations to identify different edge characteristics. In other embodiments, other edge detection algorithms may be used for edge detection, such as the Laplacian edge detection algorithm. The straight line detection is performed using the Hough straight line detection algorithm. The Hough straight line detection algorithm has strong anti-interference capabilities and is insensitive to noise in the image. In other embodiments, other methods may be used for straight line detection, such as the Freeman straight line detection algorithm and the looper crawling algorithm.

[0108] In some embodiments, the defect detection of the grate bars to be tested in the characteristic image according to the darkness mechanism specifically includes the following steps:

[0109] Traverse the grayscale values ​​of a single column of the feature map and store them, set a vertical sliding window, and add all the grayscale values ​​in the window, filter all windows in a single column whose darkness is lower than the first threshold and mark this column as a possible defect column;

[0110] Merge the continuous possible defect columns and record them as possible defect areas;

[0111] Determining whether the darkness of the possible defective area is lower than a second threshold;

[0112] If it is lower than the second threshold, the possible defective area is determined to be a defective area.

[0113] When performing defect detection on the grate bars in the feature map, first traverse the grayscale values ​​of a single column in the feature map and store them, then set a vertical sliding window, add all the grayscale values ​​in the window, get the total grayscale value in the window, filter all windows in the single column whose darkness is lower than the first threshold and mark this column as a possible defect column. Among them, in the judgment of the possible defect column, the darkness of the filtered single column is the ratio of the total grayscale value in the window to the total grayscale value of the image. Then try to merge the marked columns. If multiple marked columns appear continuously, merge the vertical window height and the horizontal continuous columns to form a rectangular mark, which is the possible defect area. Finally judge whether the area is a defect area by the darkness of this area. When the darkness of the possible defect area is lower than the second threshold, the area is judged to be a defect area. Among them, the darkness of the possible defect area is the ratio between the total grayscale value in the possible defect area and the total grayscale value of the image. Among them, the first threshold is lower than the second threshold. As many suspected defects as possible are found through the first threshold, and then the defect area is found more accurately through the second threshold. By introducing darkness defects, the impact of different lighting conditions on the defect detection mechanism is reduced.

[0114] In some embodiments, drawing the detected defects in the original image specifically comprises the following steps:

[0115] Initialize two feature maps of the same size as the original image as storage images, storing the horizontal and vertical coordinates of the pixels respectively;

[0116] For the same image, each time a defect is detected, the counter is incremented by 1, and the defect coordinates of the defect for the current feature map are recorded;

[0117] Input the recorded defect coordinates into the stored image, obtain the corresponding horizontal and vertical coordinates in the original image, and establish a mapping relationship between the defect coordinates and the coordinates of the original image;

[0118] The detected defects are drawn on the original image according to the mapping relationship.

[0119] For the same image, each time a defect is detected, the counter will be increased by 1, and the defect coordinates of the defect for the current feature map will be recorded. It is difficult to directly inversely transform the coordinates, so during the first perspective transformation, two feature maps of the same size as the original image are initialized as storage images, where one storage image stores the horizontal coordinates of the pixels, and the other storage image stores the vertical coordinates of the pixels. After that, after each transformation, the mapping relationship between the original image and the new image can be found through the horizontal and vertical coordinates of the pixels stored in the two storage images. After defect detection, the defect coordinates are input into the storage image, and the corresponding horizontal and vertical coordinates before the transformation can be obtained. The mapping relationship is established to complete the inverse transformation of the coordinates. At this point, all detected defects will be drawn on the original image, which is convenient for the staff to locate and process.

[0120] By performing barrel correction (rotating images, barrel correction, corrosion filling, etc.), image capture and re-correction (perspective transformation, canny edge detection, Hough line detection, image cropping, etc.), defect search (traversing pixels, darkness threshold comparison, etc.) and coordinate inverse transformation (auxiliary coordinate diagram) on the original camera image, the detected defects are drawn on the original image and saved. The operating status of the trolley grate bars of the ironmaking plant is detected in real time, and the camera equipment is used as the data acquisition equipment, supplemented by light fill light. The above method is used to analyze the grate bar features in the image, and finally the location of the grate bar detachment and the degree of detachment are estimated, and mapped to the corresponding trolley number. No manual intervention is required during the detection process. The real-time and accuracy of grate bar defect detection are effectively solved, the safety risks of manual detection are avoided, and the utilization efficiency of trolley equipment is improved.

[0121] See also Figure 7 , a sintering machine trolley grate defect detection system, comprising a camera 700 and a computer 600;

[0122] The camera 700 is connected to the computer 600, and the camera 700 is used to collect images of the grate bars of the sintering machine trolley;

[0123] The computer 600 includes a storage medium 610 and a processor 620 . The storage medium 610 is the storage medium in the above embodiment. The processor 620 is used to process the computer program in the storage medium 610 .

[0124] The image of the grate bars of the sintering trolley is collected by the camera 700, and the image collected by the camera 700 is sent to the computer 600. The computer 600 obtains the image of the grate bars of the sintering machine trolley, and the image obtained by the computer 600 includes several rows of grate bars. By cropping the image, the part of the grate bars that are distorted due to the shooting is cut off, and a feature map including a row of grate bars is obtained. Then, the grate bars to be tested in the feature map are checked for defects according to the darkness mechanism. When defects are detected, the detected defects are drawn in the corresponding position of the original image, so that the defects of the grate bars of the sintering trolley can be detected in real time. No manual intervention is required during the detection process, which effectively solves the hysteresis in the manual detection of grate bar defects and avoids the safety risks of manual detection. When there are defects in the grate bars, they can be detected in time. By displaying the defects in the original image, the maintenance personnel can accurately locate the position of the grate bar defects and perform timely maintenance on the grate bar defects. Among them, darkness refers to the ratio of the total gray value in the detection area to the total gray value of the image. Since images will have different grayscale values ​​under different lighting conditions, this embodiment determines whether it is a defect based on darkness, that is, the relative value of the grayscale rather than the absolute value is used as the basis for judgment, thereby avoiding misjudgment caused by the influence of lighting conditions on the grayscale value, and having higher accuracy in defect detection.

[0125] In some embodiments, the system further includes a fill light, which is used to provide fill light to the camera to make the captured image clearer.

[0126] Finally, it should be noted that although the above embodiments have been described in the specification and drawings of this application, this does not limit the scope of patent protection of this application. All technical solutions generated by replacing or modifying equivalent structures or equivalent processes based on the essential concept of this application using the contents recorded in the specification and drawings of this application, as well as directly or indirectly implementing the technical solutions of the above embodiments in other related technical fields, are included in the scope of patent protection of this application.

Claims

1. A method for detecting defects in grate bars of a sintering machine trolley. It is characterized in that The following steps are involved: Acquire an image of grates of a sintering machine carriage, wherein the image includes a plurality of rows of grates; Crop the image to obtain a feature map containing a row of grate bars; Defect detection is performed on the grate bars to be tested in the characteristic image according to the darkness mechanism; When a defect is detected, the detected defect is drawn in the original image; The defect detection of the grate bars to be tested in the characteristic graph according to the darkness mechanism specifically comprises the following steps: Traverse the grayscale values ​​of a single column of the feature map and store them, set a vertical sliding window, and add all the grayscale values ​​in the window, filter all windows in a single column whose darkness is lower than the first threshold and mark this column as a possible defect column. The darkness of a single column is the ratio of the total grayscale value in the window to the total grayscale value of the image; Merge the continuous possible defect columns and record them as possible defect areas; Determining whether the darkness of the possible defective area is lower than a second threshold; If it is lower than the second threshold, the possible defect area is judged to be a defect area, the darkness of the possible defect area is the ratio between the total grayscale value in the possible defect area and the total grayscale value of the image, and the first threshold is lower than the second threshold.

2. The method for detecting defects of grate bars of a sintering machine according to claim 1, It is characterized in that The step of cropping the image also includes the following steps: Perform image correction on the acquired images.

3. The method for detecting defects of grate bars of a sintering machine according to claim 2, It is characterized in that The image correction of the acquired image specifically comprises the following steps: The obtained image is subjected to image rotation, barrel correction and erosion filling.

4. The method for detecting defects of grate bars of a sintering machine according to claim 1, It is characterized in that The image cropping specifically comprises the following steps: Perform perspective transformation on the image according to the prior value; The perspective-transformed image is cropped to remove the preset number of rows of bars above and below and the gaps on the left and right sides; Perform edge detection, identify the upper and lower boundaries of the grate bars, and perform upper and lower cropping to keep only the middle row of grate bars to be identified; Then perform edge detection and line detection to determine the last corrected transformation coordinates and left and right boundaries, and perform corresponding affine transformation and left and right cropping.

5. The method for detecting defects of grate bars of a sintering machine according to claim 4, It is characterized in that The edge detection is performed by using a canny edge detection algorithm.

6. The method for detecting defects of grate bars of a sintering machine according to claim 4, It is characterized in that The line detection is performed by using a Hough line detection algorithm.

7. The method for detecting defects of grate bars of a sintering machine according to claim 1, It is characterized in that Drawing the detected defects in the original image specifically comprises the following steps: Initialize two feature maps of the same size as the original image as storage images, storing the horizontal and vertical coordinates of the pixels respectively; For the same image, each time a defect is detected, the counter is incremented by 1, and the defect coordinates of the defect corresponding to the current feature map are recorded; Input the recorded defect coordinates into the stored image, obtain the corresponding horizontal and vertical coordinates in the original image, and establish a mapping relationship between the defect coordinates and the coordinates of the original image; The detected defects are drawn on the original image according to the mapping relationship.

8. A storage medium storing a computer program, It is characterized in that When the computer program is executed by a processor, the steps of the method for detecting defects in grate bars of a sintering machine trolley as claimed in any one of claims 1 to 7 are executed.

9. A sintering machine trolley grate bar defect detection system, It is characterized in that Including cameras and computers; The camera is connected to the computer, and is used to collect images of the grate bars of the sintering machine trolley; The computer includes a storage medium and a processor, the storage medium is the storage medium in claim 8, and the processor is used to process the computer program in the storage medium.

Images