Material recognition method, material recognition device, material sorting apparatus, and medium
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING HONEST TECHNOLOGY CO LTD
- Filing Date
- 2025-12-30
- Publication Date
- 2026-06-19
Smart Images
Figure CN122244485A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of ore identification technology, specifically to a material identification method, a material identification device, a material sorting equipment, and a computer-readable storage medium. Background Technology
[0002] Gangue identification technology is a core component of ore sorting. Its goal is to accurately distinguish gangue from target minerals (such as coal and metal ores) to provide a basis for subsequent sorting operations.
[0003] Among related technologies, X-ray scanning is the primary method for material identification. Its core principle is to differentiate materials based on the density difference between gangue and the target mineral: different substances have different absorption coefficients for X-rays; the denser gangue absorbs X-rays more strongly, while the less dense target mineral absorbs them less. This difference is translated into a contrast in grayscale images. For pure mineral or pure gangue areas, due to their uniform internal composition and density distribution, the corresponding X-ray grayscale images exhibit consistent grayscale characteristics, clear boundaries, and stable features, facilitating rapid identification and judgment.
[0004] However, when faced with a symbiotic mixture of minerals and gangue interbedded together, the internal composition and density distribution of such materials are highly uneven, which will cause the corresponding X-ray grayscale images to show a state of mixed grayscale values, making it difficult to clearly distinguish the effective boundaries, thus affecting the recognition results. Summary of the Invention
[0005] To overcome the problems existing in the related technologies, an exemplary embodiment of this disclosure provides a material identification method, including: acquiring a ray grayscale image of the material to be identified; identifying and extracting the contour region corresponding to the material to be identified in the ray grayscale image; generating a global binary image corresponding to the contour region based on the ray grayscale image; determining at least one contour to be identified within the contour region based on the global binary image; and determining the type of the material to be identified as a symbiotic mixture type or a first type based on the contour information of the contour to be identified, wherein the symbiotic mixture type is a type formed by the mutual embedding of the first type and the second type.
[0006] In some embodiments, identifying and extracting the contour region corresponding to the material to be identified in the ray grayscale image includes: determining the material contour of the material to be identified based on the adaptive median filter window and the grayscale value of each pixel in the ray grayscale image; generating a mask corresponding to the material contour; and extracting the image region corresponding to the material contour through the mask to obtain the contour region.
[0007] In some embodiments, generating a global binary map corresponding to the contour region based on the ray grayscale image includes: dividing the contour region into grid blocks based on the ray grayscale image to obtain multiple grid regions; generating a local binary map within each grid region based on the grayscale value within each grid region; and obtaining a global binary map corresponding to the contour region based on all local binary maps.
[0008] In some embodiments, generating a local binary map within each grid region based on the gray values within each grid region includes: determining a gray threshold for the grid region based on the gray values of each pixel within the grid region; and generating a local binary map within the grid region based on the gray threshold and the gray values of each pixel within the grid region.
[0009] In some embodiments, determining the type of the material to be identified as a symbiotic mixture type or a first type based on the contour information of the contour to be identified includes: determining the aspect ratio of the minimum bounding rectangle of the contour to be identified according to the contour information of the contour to be identified; if the aspect ratio is greater than or equal to a first threshold, then determining the type corresponding to the contour to be identified as a second type and the type of the material to be identified as a symbiotic mixture type; if the aspect ratio is less than the first threshold, then determining the type of the material to be identified as a first type.
[0010] In some embodiments, determining the type of the material to be identified as a symbiotic mixture type or a first type based on the contour information of the contour to be identified further includes: determining the maximum indentation depth of the contour to be identified based on the contour information of the contour to be identified; if the maximum indentation depth is greater than or equal to a second threshold, then determining the type corresponding to the contour to be identified as the second type and the type of the material to be identified as a symbiotic mixture type; if the maximum indentation depth is less than the second threshold, then determining the type of the material to be identified as the first type.
[0011] In some embodiments, the material identification method further includes: determining the confidence level of a first type of the material to be identified based on a ray grayscale image; if the confidence level is greater than or equal to a third threshold, then determining the material as the first type; if the confidence level is less than or equal to a fourth threshold, then determining the material as the second type, wherein the fourth threshold is less than the third threshold; if the confidence level is greater than the fourth threshold and less than the third threshold, then generating a global binary image corresponding to the contour region based on the ray grayscale image.
[0012] Secondly, this disclosure also provides a material identification device, comprising: an acquisition module for acquiring a ray grayscale image of a material to be identified; an extraction module for identifying and extracting a contour region corresponding to the material to be identified in the ray grayscale image; a processing module for generating a global binary image corresponding to the contour region based on the ray grayscale image, and determining at least one contour to be identified within the contour region based on the global binary image; and an identification module for determining, based on the contour information of the contour to be identified, that the type of the material to be identified is a symbiotic mixture type or a first type, wherein the symbiotic mixture type is a type formed by the mutual embedding of the first type and the second type.
[0013] Thirdly, this disclosure also provides a material sorting device, comprising: a material identification device provided in any of the above aspects, used to determine the type of the material to be identified; and a sorting mechanism, used to sort the material to be identified based on the type.
[0014] Fourthly, this disclosure also provides a computer-readable storage medium storing computer instructions, which are used to cause a computer to perform the material identification method provided in any of the above aspects.
[0015] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure.
[0016] The technical solutions provided by the embodiments of this disclosure can include the following beneficial effects: According to the material identification method provided by this disclosure, by sequentially extracting the contour region and binarizing the grayscale image of the material to be identified, noise generated by external environmental interference factors during image acquisition can be effectively filtered out. While simplifying grayscale information, it can effectively highlight the density difference boundary inside the material to be identified, significantly narrowing the scope of subsequent data analysis and improving the real-time performance and efficiency of material identification. Furthermore, determining the type of the material to be identified based on the contour information of the contour to be identified within the contour region can more intuitively and conveniently distinguish whether the cause of the contour to be identified is due to environmental interference factors or due to the embedding of components inside the material. This can effectively improve the accuracy and reliability of type determination, effectively solving the problems of low efficiency of traditional manual judgment and the inability of threshold segmentation to adapt to complex embedding structures. This helps to shorten the material identification time, promote the material sorting process, and improve the material sorting efficiency and performance. Attached Figure Description
[0017] This disclosure can be better understood by describing exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, in which: Figure 1 This is a flowchart illustrating a material identification method according to an exemplary embodiment of a published document; Figure 2This is a schematic diagram of a ray grayscale image according to an exemplary embodiment disclosed in a publication; Figure 3 This is a schematic diagram of another ray grayscale image illustrated according to an exemplary embodiment of a disclosed document; Figure 4 This is a schematic diagram of a mask according to an exemplary embodiment disclosed in a publication; Figure 5 This is a schematic diagram illustrating a contour region according to an exemplary embodiment of a disclosed document; Figure 6 This is a schematic diagram of a binary image according to an exemplary embodiment disclosed in a book; Figure 7 This is a schematic diagram of another binary image according to an exemplary embodiment disclosed in a publication; Figure 8 This is a schematic diagram of yet another binary image illustrated according to an exemplary embodiment of a published document; Figure 9 This is a schematic diagram of another binary image shown according to an exemplary embodiment of a disclosed document; Figure 10 This is a schematic diagram of the structure of a material identification device according to an exemplary embodiment disclosed in a book; Figure 11 It is a material sorting device illustrated in an exemplary embodiment of a publication. Detailed Implementation
[0018] The following describes specific embodiments of this disclosure. It should be noted that, in order to provide a concise description, this specification cannot exhaustively describe all features of the actual embodiments. It should be understood that, in the actual implementation of any embodiment, just as in any engineering or design project, various specific decisions are often made to achieve the developer's specific goals and to meet system-related or business-related constraints, and this can change from one embodiment to another. Furthermore, it is understood that although the efforts made in this development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this disclosure, changes in design, manufacturing, or production based on the technical content disclosed in this disclosure are merely conventional technical means and should not be construed as insufficient content of this disclosure.
[0019] Unless otherwise defined, the technical or scientific terms used in this disclosure shall have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms “first,” “second,” and similar terms used in this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different components. The terms “a” or “one,” etc., do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms “comprising,” “including,” etc., mean that the element or object preceding “comprising” or “including” encompasses the elements or objects listed following “comprising” or “including” and their equivalents, and do not exclude other elements or objects. The terms “connected,” “linked,” etc., are not limited to physical or mechanical connections, nor are they limited to direct or indirect connections.
[0020] In related technologies, the identification of symbiotic mixtures in which minerals and gangue are interbedded mainly relies on two methods: manual judgment or simple threshold segmentation.
[0021] However, manual identification is not only extremely slow and cannot match the high-speed conveying rhythm of industrial belts, but also suffers from large fluctuations in identification accuracy due to subjective factors such as differences in operator experience and fatigue, making it easy to miss or misidentify.
[0022] Using a simple threshold segmentation method for identification, due to the fixed threshold parameter, makes it difficult to effectively distinguish between mineral and gangue components in the same detection area for materials with complex internal embedded structures, resulting in large final identification errors and affecting the overall identification accuracy.
[0023] To address the aforementioned problems, this disclosure provides a material identification method. For example... Figure 1 As shown, the material identification method may include the following steps: Step S110: Obtain the X-ray grayscale image of the material to be identified.
[0024] Because different types of materials have different densities, and the degree of X-ray absorption and attenuation is positively correlated with density, an X-ray camera can be used to scan the material during its transport by a material conveying mechanism to determine its type. This scan yields a grayscale image reflecting the material's internal density distribution. Higher density indicates stronger X-ray absorption and a higher grayscale value; conversely, lower density indicates weaker X-ray absorption and a lower grayscale value.
[0025] Step S120: Identify and extract the contour region corresponding to the material to be identified in the grayscale image.
[0026] To reduce the impact of background areas on material recognition, the contour region corresponding to the material to be identified in the ray grayscale image is identified and extracted to limit the recognition range of the material. In some examples, this contour region can be obtained based on image segmentation algorithms or contour recognition models. Image segmentation algorithms can include, but are not limited to, threshold segmentation, edge detection, or contour tracking, which can be determined according to requirements and are not limited here. The contour recognition model can be a pre-trained artificial intelligence (AI) model or neural network model for edge detection, which can be determined according to hardware deployment requirements and is not limited here.
[0027] Extracting the contour region helps to focus on the data within that range during subsequent analysis, effectively narrowing the analysis scope and improving recognition efficiency.
[0028] Step S130: Based on the ray grayscale image, generate a global binary image corresponding to the contour region.
[0029] By using the contour region as the boundary and binarizing the grayscale values within that region, the continuous grayscale information in the ray grayscale image can be simplified into discrete black and white features. This highlights the boundaries of density differences within the material to be identified, filters redundant grayscale noise, and thus obtains a global binary image corresponding to the contour region. For example, a grayscale threshold can be preset, marking pixels with grayscale values higher than the threshold as the first feature value and pixels with grayscale values lower than the threshold as the second feature value. Once all pixels within the region have been traversed, the desired global binary image is obtained.
[0030] Based on the distribution of feature values in the global binary graph, it is possible to quickly determine whether the density of the material to be identified is uniformly distributed. This facilitates the rapid identification of whether the material is a symbiotic mixture, simplifying the identification process and improving efficiency. The symbiotic mixture type is defined as a mixture of the first and second types interpenetrated.
[0031] Step S140: Based on the global binary map, determine at least one contour to be identified within the contour region.
[0032] In actual working conditions, the material to be identified is easily affected by environmental factors (such as light fluctuations, equipment dust accumulation, and dust covering the material surface) during the transmission process, resulting in noise in the ray grayscale image. This leads to the appearance of pseudo-feature pixels that are not the material itself in the contour area after binarization, affecting the reliability of the global binary image.
[0033] Therefore, in order to improve the accuracy of determining the type of material to be sorted, at least one contour to be identified is determined in the contour area based on the distribution of the first feature value and the second feature value in the global binary image. Based on the contour information of the contour to be identified, it is determined whether the existence of the contour to be identified is a pseudo contour caused by environmental interference factors or a real contour caused by the internal structural differences of the material to be identified itself.
[0034] Step S150: Based on the contour information of the contour to be identified, determine the type of the material to be identified as a symbiotic mixture type or the first type.
[0035] Based on the contour information of the contour to be identified, the distribution of the contour within the contour area can be determined. The contour information may include, but is not limited to, the size and location information of the contour to be identified.
[0036] If the distribution of the contours to be identified is relatively close and the size features are highly similar, then the existence of the contours to be identified can be considered as being caused by environmental interference factors. Therefore, it can be determined that the material to be identified is a pure substance and the corresponding type is the first type.
[0037] If the distribution of the contours to be identified has no obvious pattern and the size characteristics are significantly different, the existence of the contours to be identified can be considered as being caused by the differences in the internal structure of the material to be identified, and thus the type of the material to be identified can be determined to be a symbiotic mixture.
[0038] According to the material identification method provided in this disclosure, by sequentially extracting the contour region and binarizing the grayscale image of the material to be identified, noise generated by external environmental interference factors during image acquisition can be effectively filtered out. While simplifying grayscale information, it can effectively highlight the density difference boundaries within the material to be identified, significantly narrowing the scope of subsequent data analysis and improving the real-time performance and efficiency of material identification. Furthermore, by determining the type of the material to be identified based on the contour information of the contour within the contour region, it is possible to more intuitively and conveniently distinguish the cause of the contour to be identified, whether it is caused by environmental interference factors or formed by the embedding of components within the material. This can effectively improve the accuracy and reliability of type determination, and effectively solve the problems of low efficiency in traditional manual discrimination and the inability of threshold segmentation to adapt to complex embedding structures. This helps to shorten the material identification time, promote the material sorting process, and improve the efficiency and performance of material sorting.
[0039] In some embodiments, step S120 above may include the following steps: Step a1: Based on the adaptive median filter window and the grayscale values of each pixel in the ray grayscale image, determine the material outline of the material to be identified.
[0040] To ensure the accuracy of material outline recognition and eliminate noise interference, the size of the filter window is adaptively adjusted based on the preset dynamic adjustment rules of the filter window and the gray value distribution characteristics of each pixel in the ray grayscale image, thereby filtering out environmental interference noise such as salt and pepper noise and Gaussian noise in the image.
[0041] Based on the filtered and denoised ray grayscale image, the material contour of the material to be identified is determined through edge detection or contour tracking algorithms. For example, the Otsu method (OTSU) global adaptive threshold segmentation algorithm can be used to binarize the filtered ray grayscale image, and then a contour detection algorithm can be used to extract all connected component contours in the image. Subsequently, based on the contour area feature, the connected component contour with the largest area is selected as the material contour of the material to be identified, thereby eliminating false contour interference caused by environmental noise and small impurities in the image.
[0042] The process of noise reduction using an adaptive median filtering window can be as follows: Initialize the filtering window to its minimum size (e.g., 3×3), and use the center pixel within the current window as the pixel to be judged; count the gray values of all pixels within the window to obtain the original gray set, for example: [52,55,60,150,250,148,145,58,62], where the center pixel's gray value is 250. Calculate the interquartile range (IQR) or mean ± 2 standard deviations of the gray values within the window to determine the normal gray range of this local area. For example, the minimum and maximum values of the effective pixels within the window after removing isolated extrema can be selected as the normal gray range for simplified judgment, such as [52~150] in the example above. Since 250 significantly exceeds the normal gray range of the window, the center pixel can be considered an extremum, i.e., salt-and-pepper noise. If the center pixel is detected as an extreme value and the current window size has not reached the preset maximum size (e.g., 7×7), the window size is gradually increased (from 3×3 to 5×5, then to 7×7), and the median gray value within the new window is recalculated. This median value replaces the gray value of the center pixel. If the center pixel is not detected as an extreme value, or the window has reached its maximum size, the gray value of the center pixel within the current window is retained, or the median gray value of the current window is used for smoothing to avoid over-filtering and blurring of material edge details.
[0043] Step a2: Generate the mask corresponding to the material outline.
[0044] Based on the position of the material outline in the ray grayscale image, a mask corresponding to the material outline is generated so that the outline boundary range of the material outline can be locked during subsequent processing, and the pixels within the material outline can be focused for feature analysis. This can effectively reduce the interference of the background area and improve processing efficiency and accuracy.
[0045] Step a3: Extract the image region corresponding to the material contour using a mask to obtain the contour region.
[0046] By using a mask to extract the image region corresponding to the material contour, the extraction accuracy of the image region can be guaranteed, making the obtained contour region more reliable and complete, thus laying a high-quality data foundation for subsequent binarization processing and contour feature analysis.
[0047] In some embodiments, step S130 above may include the following steps: Step b1: Based on the ray grayscale image, the contour region is divided into grid blocks to obtain multiple grid regions.
[0048] Using a preset grid size (e.g., 10×10 pixels), a uniform block division strategy is employed to divide the contour region into multiple non-overlapping rectangular sub-regions. This reduces segmentation errors caused by the global threshold when determining the global binary image, thus improving the generation accuracy of the global binary image. In one example, if the contour region is irregular, edge grid regions can be filled with standard sizes consistent with other grid regions using edge grayscale interpolation or background value padding to ensure size consistency across all grid regions.
[0049] Furthermore, by using the pixel coordinate mapping relationship of the ray grayscale image, the relative position information of each grid region within the contour region can be determined, which helps to provide a positioning basis for subsequent stitching of local binary images.
[0050] Step b2: Based on the grayscale values within each grid region, generate a local binary image for each grid region.
[0051] Based on the distribution of gray values within the grid area, a corresponding local binary image is generated. This ensures that the generated local binary image better matches the actual pixel features of the corresponding grid area, thereby effectively reducing feature value loss and improving the generation accuracy of the local binary image.
[0052] In some examples, step b2 above may include: determining a grayscale threshold for the grid region based on the grayscale values of each pixel within the grid region; and generating a local binary image within the grid region based on the grayscale threshold and the grayscale values of each pixel within the grid region. That is, to make the generation of the local binary image more reasonable, the average grayscale value of the grid region is determined based on the distribution of grayscale values of each pixel within the grid region. To make the first and second feature values within the grid region more reliable and to avoid over-segmentation, the minimum grayscale value within the grid region plus one-third of the average grayscale value is used as the grayscale threshold for that grid region. Pixels within the grid region with grayscale values greater than or equal to the grayscale threshold are marked as the first feature value, and pixels with grayscale values less than the grayscale threshold are marked as the second feature value, thereby obtaining a local binary image within the grid region.
[0053] Step b3: Based on all local binary maps, obtain the global binary map corresponding to the contour region.
[0054] Since each local binary image is generated independently based on the gray value distribution characteristics of the corresponding grid region, its threshold selection is more flexible and targeted, and can accurately match the actual pixel gray value characteristics of each grid. Therefore, by stitching and merging all local binary images sequentially according to their original coordinates based on the relative position information of the grid region corresponding to each local binary image within the contour region, the resulting global binary image can be consistent with the size of the contour region.
[0055] Furthermore, this stitching method can ensure the integrity and continuity of the global binary image, avoid the problem of local feature loss caused by global threshold binarization, and enable the final output global binary image to accurately present the density embedding structure inside the material to be identified, providing reliable data support for subsequent contour analysis and type determination.
[0056] In some optional implementation scenarios, to improve the extraction efficiency of the contour to be identified, the pixel values of the background region of the contour area can be uniformly set to the same value as the second feature value. The first feature value and the second feature value are relative colors (for example, the first feature value is 255, corresponding to white; the second feature value is 0, corresponding to black). Thus, based on the significant color difference between the two, the contour to be identified within the contour area can be quickly distinguished and located, reducing the interference of background pixels on the contour extraction algorithm.
[0057] In some embodiments, step S150 above may include the following steps: Step c1: Determine the aspect ratio of the minimum bounding rectangle of the contour to be identified based on the contour information of the contour to be identified. Step c2: If the aspect ratio is greater than or equal to the first threshold, then the type of the contour to be identified is determined to be the second type, and the type of the material to be identified is the symbiotic mixture type. Step c3: If the aspect ratio is less than the first threshold, then the type of the material to be identified is determined to be the first type.
[0058] Specifically, the material to be identified includes at least substances of the first type. To determine whether the material to be identified is a pure substance, the length and width of the minimum bounding rectangle of the outline to be identified are determined based on the outline information. A first threshold is preset. The first threshold can be the minimum aspect ratio used to determine whether the material type corresponding to the outline is the second type. The value of the first threshold can be determined based on the actual characteristics of the second type of material.
[0059] The length-to-width ratio (aspect ratio) is compared with a first threshold. If the aspect ratio is greater than or equal to the first threshold, it indicates that the type corresponding to the profile to be identified is type two, and the material to be identified is not a pure substance but also includes substances of type two. Therefore, the type of the material to be identified is determined to be a symbiotic mixture. If the aspect ratio is less than the first threshold, it indicates that the type corresponding to the profile to be identified is not type two, and the profile to be identified is caused by environmental interference factors. Therefore, it can be determined that the material to be identified is a pure substance, and the type of the material to be identified is type one.
[0060] In some other embodiments, step S150 above may further include the following steps: Step d1: Determine the maximum indentation depth of the contour to be identified based on the contour information of the contour to be identified. Step d2: If the maximum indentation depth is greater than or equal to the second threshold, then the type of the contour to be identified is determined to be the second type, and the type of the material to be identified is the symbiotic mixture type. Step d3: If the maximum indentation depth is less than the second threshold, then the type of the material to be identified is determined to be the first type.
[0061] The second threshold can be the minimum indentation depth used to determine whether the material type corresponding to the contour is of the second type. The value of the second threshold can be determined based on the actual characteristics of the material of the second type.
[0062] If the maximum indentation depth is greater than or equal to the second threshold, it indicates that the type corresponding to the contour to be identified is the second type, and the material to be identified is not a pure substance, but also includes substances of the second type. Therefore, the type of the material to be identified is determined to be a symbiotic mixture. If the maximum indentation depth is less than the second threshold, it indicates that the type corresponding to the contour to be identified is not the second type, and the contour to be identified is caused by environmental interference factors. Therefore, it can be determined that the material to be identified is a pure substance, and the type of the material to be identified is the first type.
[0063] In some embodiments, the material identification method provided in this disclosure may further include: determining the confidence level of a first type of the material to be identified based on a ray grayscale image. To improve the efficiency of material identification, after obtaining the ray grayscale image of the material to be identified, the type of the material to be identified is first determined by an identification model to determine the confidence level of the first type. If the confidence level is greater than or equal to a third threshold, the material is determined to be of the first type. The third threshold is the minimum confidence level for determining the first type as a reliable result. If the confidence level is less than or equal to a fourth threshold, the material is determined to be of the second type. The fourth threshold is less than the third threshold, and the fourth threshold is the maximum confidence level for determining the first type as an unreliable result, and can also be used as the minimum confidence level for determining the second type as a reliable result.
[0064] If the confidence level is greater than the fourth threshold and less than the third threshold, it indicates that it may be a symbiotic hybrid type. To improve the recognition accuracy, the above step S130 is executed to generate a global binary map corresponding to the contour region based on the ray grayscale map, so as to determine the type of material to be identified based on the contour to be identified in the global binary map.
[0065] In some optional application scenarios, taking the material's primary type as coal, secondary type as gangue, and symbiotic mixture type as interbedded gangue coal as an example, if the material to be identified is interbedded gangue coal, the acquired X-ray grayscale image can be as follows: Figure 2 As shown, the distribution location of the gangue area can be as follows: Figure 3 As shown.
[0066] The process of identifying the type of material to be sorted using this disclosure can be as follows: Obtain a grayscale ray image of the material to be identified, which may be coal. Denoise the grayscale image and use OTSU adaptive thresholding to filter out the material contour with the largest area. Generate... Figure 4 The mask corresponding to the material outline shown is displayed (outline area is white, background is black). Extract the outline area corresponding to the material to be identified from the ray grayscale image, and configure the background as follows: Figure 5 The color shown is black.
[0067] The contour region is divided into multiple grid regions using a 10×10 grid size. Based on the grayscale values of non-black pixels within each grid region, a grayscale threshold is determined for that region. Then, based on the grayscale threshold and the grayscale values of each pixel within the corresponding grid region, a local binary image is generated for each grid region (e.g., ...). Figure 6 The local binary images shown are stitched together based on their relative positions within the contour region to obtain the result shown below. Figure 7 The global binary graph shown.
[0068] Perform morphological operations on the global binary image, including opening, to remove noise while preserving the larger main structures in the original image, resulting in: Figure 8 The results are shown. Using the denoised global binary image, at least one contour to be identified within the contour region is determined and extracted, such as... Figure 9 The black area shown is highlighted.
[0069] Based on the contour information of the contour to be identified, the aspect ratio of the minimum bounding rectangle of the contour to be identified is determined. If the aspect ratio is greater than or equal to the first threshold, the type of the contour to be identified is determined to be gangue, and the type of the material to be identified is intercalated coal; if the aspect ratio is less than the first threshold, the type of the material to be identified is determined to be coal.
[0070] Based on the contour information of the contour to be identified, the maximum indentation depth of the contour is determined. If the maximum indentation depth is greater than or equal to the second threshold, the contour is identified as gangue, and the material type is identified as interbedded coal. If the maximum indentation depth is less than the second threshold, the material type is identified as coal.
[0071] Based on the same inventive concept, this disclosure also provides a material identification device. For example... Figure 10 As shown, the material identification device 200 may include: The acquisition module 210 is used to acquire the grayscale image of the material to be identified; Extraction module 220 is used to identify and extract the contour region corresponding to the material to be identified in the grayscale image; The processing module 230 is used to generate a global binary map corresponding to the contour region based on the ray grayscale map, and to determine at least one contour to be identified within the contour region based on the global binary map. The identification module 240 is used to determine the type of the material to be identified as a symbiotic mixture type or a first type based on the contour information of the contour to be identified, wherein the symbiotic mixture type is a type formed by the interpenetration of the first type and the second type.
[0072] In some embodiments, the extraction module 220 may include: a first determining unit, configured to determine the material outline of the material to be identified based on an adaptive median filtering window and the gray values of each pixel in the ray grayscale image; a mask generating unit, configured to generate a mask corresponding to the material outline; and an extraction unit, configured to extract the image region corresponding to the material outline through the mask to obtain the outline region.
[0073] In some embodiments, the processing module 230 may include: a first processing unit, configured to perform gridding processing on the contour region based on the ray grayscale image to obtain multiple grid regions; a second processing unit, configured to generate a local binary image in each grid region based on the grayscale value in each grid region; and a third processing unit, configured to obtain a global binary image corresponding to the contour region based on all local binary images.
[0074] In some embodiments, the second processing unit may include: a first execution unit, configured to determine a grayscale threshold for the grid region based on the grayscale values of each pixel within the grid region; and a second execution unit, configured to generate a local binary map within the grid region based on the grayscale threshold and the grayscale values of each pixel within the grid region.
[0075] In some embodiments, the identification module 240 may include: a second determining unit, configured to determine the aspect ratio of the minimum bounding rectangle of the contour to be identified based on the contour information of the contour to be identified; a third determining unit, configured to determine that the type corresponding to the contour to be identified is a second type and the type of the material to be identified is a symbiotic mixture type if the aspect ratio is greater than or equal to a first threshold; and a fourth determining unit, configured to determine that the type of the material to be identified is a first type if the aspect ratio is less than the first threshold.
[0076] In some embodiments, the identification module 240 may further include: a fifth determining unit, configured to determine the maximum indentation depth of the contour to be identified based on the contour information of the contour to be identified; a sixth determining unit, configured to determine that the type corresponding to the contour to be identified is a second type and the type of the material to be identified is a symbiotic mixture type if the maximum indentation depth is greater than or equal to a second threshold; and a seventh determining unit, configured to determine that the type of the material to be identified is a first type if the maximum indentation depth is less than the second threshold.
[0077] In some embodiments, the processing module 230 may further include: an eighth determining unit, configured to determine the confidence level of a first type of material to be identified based on a ray grayscale image; a fourth processing unit, configured to determine that the material is of the first type if the confidence level is greater than or equal to a third threshold; a fifth processing unit, configured to determine that the material is of the second type if the confidence level is less than or equal to a fourth threshold, wherein the fourth threshold is less than the third threshold; and a sixth processing unit, configured to generate a global binary image corresponding to the contour region based on the ray grayscale image if the confidence level is greater than the fourth threshold and less than the third threshold.
[0078] Regarding the material identification device in the above embodiments, the specific way in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated here.
[0079] Based on the same inventive concept, this disclosure also provides a material sorting device. For example... Figure 11 As shown, the material sorting equipment may include: a material identification device 200 provided in this disclosure, used to determine the type of the material to be identified; and a sorting mechanism 300, used to sort the material to be identified based on the type.
[0080] Based on the same inventive concept, this disclosure also provides a computer-readable storage medium storing a program for performing the material sorting method of any of the foregoing embodiments.
[0081] This disclosure uses specific terms to describe embodiments of the present disclosure. Terms such as "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of the present disclosure. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Furthermore, certain features, structures, or characteristics in one or more embodiments of the present disclosure can be appropriately combined.
[0082] In the context of this disclosure, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.
[0083] Similarly, it should be noted that, in order to simplify the description of this disclosure and thus aid in the understanding of one or more embodiments, the foregoing description of embodiments of this disclosure may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this disclosure method does not imply that the subject matter of this disclosure requires more features than the features claimed. In fact, the embodiments contain fewer features than all the features of the single embodiments disclosed above.
[0084] The basic concepts have been described above. It is obvious that the above disclosure is merely illustrative and does not constitute a limitation of this disclosure. Although not explicitly stated herein, various modifications, improvements, and corrections may be made to this disclosure by those skilled in the art. Such modifications, improvements, and corrections are suggested in this disclosure and therefore remain within the spirit and scope of the embodiments of this disclosure.
Claims
1. A material identification method, comprising: Obtain the grayscale image of the material to be identified; Identify and extract the contour region corresponding to the material to be identified in the grayscale image; Based on the ray grayscale image, a global binary image corresponding to the contour region is generated; Based on the global binary image, at least one contour to be identified is determined within the contour region; Based on the contour information of the contour to be identified, the type of the material to be identified is determined to be either a symbiotic mixture type or a first type, wherein the symbiotic mixture type is a type formed by the interpenetration of the first type and the second type.
2. The material identification method according to claim 1, wherein, The process of identifying and extracting the contour region corresponding to the material to be identified in the grayscale image includes: The material outline of the material to be identified is determined based on the adaptive median filter window and the gray values of each pixel in the ray grayscale image. Generate a mask corresponding to the material outline; The image region corresponding to the material contour is extracted using the mask to obtain the contour region.
3. The material identification method according to claim 1, wherein, The step of generating a global binary map corresponding to the contour region based on the ray grayscale image includes: Based on the ray grayscale image, the contour region is divided into grid blocks to obtain multiple grid regions; Based on the grayscale values within each of the grid regions, a local binary image is generated within each of the grid regions; Based on all the local binary maps, the global binary map corresponding to the contour region is obtained.
4. The material identification method according to claim 3, wherein, The step of generating a local binary map for each of the grid regions based on the grayscale values within each grid region includes: The grayscale threshold of the grid region is determined based on the grayscale value of each pixel within the grid region. Based on the grayscale threshold and the grayscale values of each pixel within the grid area, a local binary map within the grid area is generated.
5. The material identification method according to any one of claims 1-4, wherein, The step of determining the type of the material to be identified as a symbiotic mixture or the first type based on the contour information of the contour to be identified includes: Based on the contour information of the contour to be identified, determine the aspect ratio of the minimum bounding rectangle of the contour to be identified; If the aspect ratio is greater than or equal to the first threshold, then the type corresponding to the contour to be identified is determined to be the second type, and the type of the material to be identified is the symbiotic mixture type; If the aspect ratio is less than the first threshold, then the type of the material to be identified is determined to be the first type.
6. The material identification method according to claim 5, wherein, The step of determining the type of the material to be identified as a symbiotic mixture or the first type based on the contour information of the contour to be identified further includes: Based on the contour information of the contour to be identified, determine the maximum indentation depth of the contour to be identified; If the maximum indentation depth is greater than or equal to the second threshold, then the type corresponding to the contour to be identified is determined to be the second type, and the type of the material to be identified is the symbiotic mixture type; If the maximum indentation depth is less than the second threshold, then the type of the material to be identified is determined to be the first type.
7. The material identification method according to claim 1, wherein, The method further includes: Based on the ray grayscale image, determine the confidence level of the first type of the material to be identified; If the confidence level is greater than or equal to the third threshold, then the material is determined to be of the first type; If the confidence level is less than or equal to the fourth threshold, then the material is determined to be of the second type, wherein the fourth threshold is less than the third threshold; If the confidence level is greater than the fourth threshold and less than the third threshold, then the process of generating a global binary map corresponding to the contour region based on the ray grayscale image is performed.
8. A material identification device, comprising: The acquisition module is used to acquire the grayscale image of the material to be identified; The extraction module is used to identify and extract the contour region corresponding to the material to be identified in the grayscale image; The processing module is used to generate a global binary map corresponding to the contour region based on the ray grayscale map, and to determine at least one contour to be identified within the contour region based on the global binary map. The identification module is used to determine, based on the contour information of the contour to be identified, whether the material to be identified is a symbiotic mixture type or a first type, wherein the symbiotic mixture type is a type formed by the first type and the second type interpenetrating each other.
9. A material sorting device, comprising: The material identification device according to claim 8 is used to determine the type of the material to be identified; A sorting mechanism for sorting the materials to be identified based on the type.
10. A computer-readable storage medium storing computer instructions for causing a computer to perform the material identification method according to any one of claims 1 to 7.