Digital characterization method of hypoeutectic aluminum-silicon alloy and related device

Abstract

The invention discloses a digital characterization method of a hypoeutectic aluminum-silicon alloy and a related device, and the method comprises the steps: obtaining a metallographic structure microphotograph of the hypoeutectic aluminum-silicon alloy, carrying out the preprocessing of the metallographic structure microphotograph, and obtaining a target metallographic structure microphotograph; the target metallographic structure microphotograph is transmitted to a preset eutectic aluminum-silicon extraction model, an eutectic silicon mask is obtained, and the eutectic silicon mask and the target metallographic structure microphotograph are the same in size; determining a silicon segregation region based on the eutectic silicon mask; and determining a dendritic crystal diagram based on the eutectic silicon mask, and calculating secondary dendritic crystal spacing and / or microscopic grain size based on the dendritic crystal diagram. According to the process, the eutectic silicon mask is obtained based on the preset eutectic aluminum-silicon extraction model, the silicon segregation area, the secondary dendritic crystal spacing and the microscopic grain size can be determined based on the eutectic aluminum-silicon mask, all processes are automatically completed, manual operation is not needed any more, and the efficiency is high. The problems that in the prior art, artificial experience is high in subjectivity, analysis time is long, local statistics is limited, and precision is not high are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of quantitative metallographic analysis, and in particular to a digital characterization method and related device for hypoeutectic aluminum-silicon alloy. Background Art

[0002] With the rapid development of metal material science and technology, metal research and development has gradually been established on the basis of the quantitative relationship between composition, structure, organization and performance. For metal materials, through the generation and various subsequent processing processes, their phase structure and microstructure can be controlled to achieve the desired performance.

[0003] Quantitative metallographic analysis is an important method for studying the relationship between the composition, organization, process and performance of metal materials. Through the quantitative analysis of the metallographic organization of various materials, the quantitative relationship between microstructure and macroscopic performance is constructed. In engineering practice, this work is mainly completed by engineering and technical personnel. However, this method has problems such as strong subjectivity of manual experience, long analysis time, limited local statistics and low accuracy. Summary of the invention

[0004] In view of this, the present invention provides a digital characterization method and related devices for hypoeutectic aluminum-silicon alloys, which are used to solve the problem that the existing quantitative metallographic analysis is mainly completed by engineering technicians. However, this method has the problems of strong subjectivity of manual experience, long analysis time, local statistical limitations and low accuracy. The specific scheme is as follows:

[0005] A digital characterization method for hypoeutectic aluminum-silicon alloy, comprising:

[0006] Obtaining a metallographic micrograph of a hypoeutectic aluminum-silicon alloy, and preprocessing the metallographic micrograph to obtain a target metallographic micrograph;

[0007] Passing the target metallographic structure micrograph to a preset eutectic aluminum-silicon extraction model to obtain a eutectic silicon mask, wherein the eutectic silicon mask has the same size as the target metallographic structure micrograph;

[0008] determining silicon segregation regions based on the eutectic silicon mask, or;

[0009] A dendrite diagram is determined based on the eutectic silicon mask, and a secondary dendrite spacing and / or a microscopic grain size is calculated based on the dendrite diagram.

[0010] The digital characterization method of the hypoeutectic aluminum-silicon alloy described above may optionally include preprocessing the metallographic micrograph to obtain a target metallographic micrograph, including:

[0011] Converting the pixel length in the metallographic structure microphotograph into an actual length to obtain a first metallographic structure microphotograph;

[0012] Performing data enhancement processing on the first metallographic structure micrograph to obtain a second metallographic structure micrograph;

[0013] The second metallographic structure microphotograph is converted into a single-channel image, and the single-channel image is normalized to obtain a target metallographic structure microphotograph.

[0014] The above-mentioned digital characterization method of hypoeutectic aluminum-silicon alloy, optionally, determines the silicon segregation area based on the eutectic silicon mask, including:

[0015] Down-sampling the eutectic silicon mask to obtain a sampled eutectic silicon mask;

[0016] performing closing operation and opening operation on the sampled eutectic silicon mask in sequence to obtain an initial silicon segregation region;

[0017] The initial silicon segregation region is upsampled to obtain a silicon segregation region.

[0018] The above-mentioned digital characterization method of hypoeutectic aluminum-silicon alloy may optionally further include:

[0019] The initial silicon segregation region is identified, holes in the initial silicon segregation region whose area is smaller than a first preset number of pixels are filled, and / or regions in the initial silicon segregation region whose area is smaller than a second preset number of pixels are eliminated.

[0020] The above-mentioned digital characterization method of hypoeutectic aluminum-silicon alloy, optionally, determines the dendrite diagram based on the eutectic silicon mask, including:

[0021] Performing distance transformation on the eutectic silicon mask to obtain a distance transformation map;

[0022] The distance transformation graph is segmented based on a watershed algorithm to obtain a dendrite graph.

[0023] The digital characterization method of the hypoeutectic aluminum-silicon alloy described above may optionally calculate the microscopic grain size based on the dendrite diagram, including:

[0024] Based on a preset magnification, a circular area with a preset perimeter is drawn, and the number of intersections between the circular area and the connected domains in the dendrite graph is counted;

[0025] The quantity is passed to a preset microscopic grain size calculation formula to obtain the microscopic grain size.

[0026] A digital characterization device for hypoeutectic aluminum-silicon alloy, comprising:

[0027] A preprocessing module, used to obtain a metallographic micrograph of the hypoeutectic aluminum-silicon alloy, and preprocess the metallographic micrograph to obtain a target metallographic micrograph;

[0028] An extraction module, used for transferring the target metallographic structure microphotograph to a preset eutectic aluminum silicon extraction model to obtain a eutectic silicon mask, wherein the eutectic silicon mask has the same size as the target metallographic structure microphotograph;

[0029] a first determination module, configured to determine a silicon segregation region based on the eutectic silicon mask, or;

[0030] The second determination module is used to determine a dendrite diagram based on the eutectic silicon mask, and calculate a secondary dendrite spacing and / or a microscopic grain size based on the dendrite diagram.

[0031] The digital characterization device for the hypoeutectic aluminum-silicon alloy, optionally, the pre-processing module comprises:

[0032] A conversion unit, used for converting the pixel length in the metallographic structure microphotograph into an actual length to obtain a first metallographic structure microphotograph;

[0033] An enhancement unit, used for performing data enhancement processing on the first metallographic structure micrograph to obtain a second metallographic structure micrograph;

[0034] The conversion and normalization unit is used to convert the second metallographic structure microphotograph into a single-channel image and perform normalization processing on the single-channel image to obtain a target metallographic structure microphotograph.

[0035] The digital characterization device for the hypoeutectic aluminum-silicon alloy, optionally, the first determination module includes:

[0036] A down-sampling unit, used for down-sampling the eutectic silicon mask to obtain a sampled eutectic silicon mask;

[0037] A calculation unit, used for sequentially performing a closing operation and an opening operation on the sampled eutectic silicon mask to obtain an initial silicon segregation region;

[0038] An up-sampling unit is used to up-sample the initial silicon segregation region to obtain a silicon segregation region.

[0039] The digital characterization device for the hypoeutectic aluminum-silicon alloy, optionally, the second determination module includes:

[0040] A transformation unit, used for performing distance transformation on the eutectic silicon mask to obtain a distance transformation map;

[0041] The segmentation unit is used to segment the distance transformation graph based on a watershed algorithm to obtain a dendrite graph.

[0042] A storage medium comprises a stored program, wherein the program executes the steps of the above-mentioned method for digital characterization of hypoeutectic aluminum-silicon alloy.

[0043] A computer program product comprises a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for digital characterization of hypoeutectic aluminum-silicon alloy.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] The present invention discloses a digital characterization method and related device for hypoeutectic aluminum-silicon alloy, including: obtaining a metallographic micrograph of hypoeutectic aluminum-silicon alloy, preprocessing the metallographic micrograph to obtain a target metallographic micrograph; passing the target metallographic micrograph to a preset eutectic aluminum-silicon extraction model to obtain a eutectic silicon mask, wherein the eutectic silicon mask is the same size as the target metallographic micrograph; determining the silicon segregation region based on the eutectic silicon mask, or; determining a dendrite diagram based on the eutectic silicon mask, and calculating the secondary dendrite spacing and / or microcrystalline grain size based on the dendrite diagram. The above process obtains a eutectic silicon mask based on a preset eutectic aluminum-silicon extraction model, and the silicon segregation region, secondary dendrite spacing and microcrystalline grain size can be determined based on the eutectic aluminum-silicon mask, all of which are completed automatically and no longer based on manual work, thus avoiding the problems of strong subjectivity of manual experience, long analysis time, limited local statistics and low precision in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A flow chart of a digital characterization method for a hypoeutectic aluminum-silicon alloy disclosed in an embodiment of the present invention;

[0048] Figure 2 A structural block diagram of a digital characterization device for a hypoeutectic aluminum-silicon alloy disclosed in an embodiment of the present invention;

[0049] Figure 3 The present invention is a schematic diagram of a device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] The present invention discloses a digital characterization method and related devices for hypoeutectic aluminum-silicon alloy, which are used for automated characterization of eutectic silicon segregation, secondary dendrite spacing and micro grain size of hypoeutectic aluminum-silicon alloy, avoiding the existing quantitative metallographic analysis which is mainly completed by engineering technicians. However, this method has the problems of strong subjectivity of manual experience, long analysis time, local statistical limitations and low accuracy. The execution flow of the method is as follows: Figure 1 As shown, the steps include:

[0052] S101, obtaining a metallographic micrograph of a hypoeutectic aluminum-silicon alloy, and preprocessing the metallographic micrograph to obtain a target metallographic micrograph;

[0053] In an embodiment of the present invention, a metallographic micrograph of a hypoeutectic aluminum-silicon alloy is obtained. The metallographic micrograph can be pre-stored in a designated location and can be obtained at the designated location. The designated location can be set based on a specific scenario. In the embodiment of the present invention, the designated location is not specifically limited. The hypoeutectic aluminum-silicon alloy can also be directly placed on a microscope to obtain the alloy. In the embodiment of the present invention, the specific acquisition method is not limited.

[0054] After obtaining the metallographic micrograph, the metallographic micrograph is preprocessed. The specific preprocessing process is as follows:

[0055] In the metallographic micrograph, silicon particles are usually dark particles, and the size of the particles can range from a few microns to tens of microns, depending on the composition of the alloy and the processing conditions. The morphology of unmodified eutectic silicon is coarse flakes or needles, and the modified eutectic silicon particles are round or oval. The silicon particle template of the metallographic micrograph is marked by using image processing software. Since the size in the metallographic microphotograph is directly measured to obtain the pixel length, in order to obtain the corresponding actual length, it is necessary to add scale information to the collected image, and take the product of the pixel length and the scale information as the corresponding actual length to obtain the first metallographic microphotograph. Then, the first metallographic microphotograph is subjected to data enhancement processing to obtain the second metallographic microphotograph, wherein the data enhancement processing includes two types, the first data enhancement method includes: common means such as cropping, flipping, rotation, translation and mirroring; the second data enhancement method includes: adjusting the brightness value and contrast of the image to simulate the microphotographs under different polishing, corrosive agents and lighting conditions, so that the model has better robust performance, further, the second metallographic microphotograph is converted into a single-channel image, preferably, the single-channel image is an 8-bit single-channel image, and the single-channel image is normalized to obtain the target metallographic microphotograph, wherein the specific processing process of converting the second metallographic microphotograph into a single-channel image is as follows:

[0056] In computer science and image processing, images are usually represented as two-dimensional matrices or three-dimensional matrices (for color images). Each element (or pixel) in the matrix represents a point in the image and contains the color information or grayscale value of the point.

[0057] For grayscale images, a two-dimensional matrix is ​​usually used to represent them, where each element contains a grayscale value, indicating the brightness of the corresponding position in the image. The rows and columns of this matrix correspond to the height and width of the image, respectively.

[0058] For color images, a three-dimensional matrix is ​​usually used to represent them, where the first dimension represents the color channel (such as red, green, blue), and the second two dimensions represent the height and width of the image. Such a matrix is ​​called an RGB image because it consists of three color channels: red (R), green (G), and blue (B).

[0059] By performing various mathematical operations on these matrices, such as matrix multiplication and convolution, various image processing methods can be performed, such as filtering, edge detection, image enhancement, etc.

[0060] If the original image is a 16-bit image, you need to linearly map the 16-bit pixel values ​​into the 8-bit range. This can be done by simply dividing the 16-bit pixel value by 256 and then rounding to the nearest integer value.

[0061] In the embodiment of the present invention, the second metallographic structure micrograph is an RGB image. The three-channel RGB image is converted into a single-channel grayscale image usually using a weighted average method, in which the R, G, and B channel values ​​of each pixel are weighted averaged according to a certain weight to obtain a single grayscale value.

[0062] A common weighted average method is to use the following formula:

[0063] Single channel image gray value = 0.299×R+0.587×G+0.114×B

[0064] This formula is determined based on the sensitivity of the human eye to different colors. Preferably, the weights of the three channels R, G, and B are 0.299, 0.587, and 0.114, respectively.

[0065] The single-channel image is normalized to obtain a target metallographic micrograph as follows: the single-channel image matrix is ​​divided by 255, the pixel values ​​of the image matrix are normalized to the interval [0, 1], and the target metallographic micrograph is obtained.

[0066] S102, passing the target metallographic structure micrograph to a preset eutectic aluminum-silicon extraction model to obtain a eutectic silicon mask, wherein the eutectic silicon mask has the same size as the target metallographic structure micrograph;

[0067] In an embodiment of the present invention, the preset eutectic aluminum silicon extraction model selects Unet as the segmentation model network, adopts the cross entropy / Dice coefficient linear combination as the loss function, selects Adamw as the optimization algorithm, and sets the learning rate to 5e-5. The deep learning model is trained on the GPU until convergence is reached. In an embodiment of the present invention, the specific training process is not limited, the input of the preset eutectic aluminum silicon extraction model is the target microscopic photograph of the structure, and the output of the preset eutectic aluminum silicon extraction model is the eutectic silicon mask, wherein the eutectic silicon mask is the same size as the target metallographic microscopic photograph.

[0068] S103, determining a silicon segregation region based on the eutectic silicon mask, or;

[0069] In the embodiment of the present invention, during the solidification process of the die-cast aluminum alloy, due to the structure of the casting, hot nodes may be formed in certain parts, causing silicon segregation. In the metallographic micrograph, the silicon content of the eutectic silicon segregation area is significantly higher than that of other areas. Through morphological processing, the area with dense eutectic silicon can be located, and the steps are as follows:

[0070] In order to improve the efficiency of subsequent calculations, the eutectic silicon mask is downsampled to 1 / 3 of the original size by the nearest neighbor method to obtain the sampled eutectic silicon mask. The nearest neighbor downsampling is an image processing technology used to reduce the resolution of an image. The principle is very simple, that is, for an image to be reduced in resolution, a pixel is selected from each pixel block as its representative to construct a new image with lower resolution. The main steps of the nearest neighbor downsampling are as follows:

[0071] For each pixel in the eutectic silicon mask, determine its position in the downsampled image.

[0072] According to the determined position, the nearest pixel is selected from the eutectic silicon mask as the downsampled pixel value.

[0073] Assign the selected pixel value to the pixel at the corresponding position in the downsampled image.

[0074] The sampled eutectic silicon mask is closed and then opened. The operation kernel uses an 11*11 elliptical kernel (suitable for images magnified 100X) to obtain the initial silicon segregation area.

[0075] In the embodiment of the present invention, the closing operation is a process of first performing an expansion operation on the image and then performing an erosion operation. Its formula is as follows:

[0076] (1)

[0077] in, is the input image (eutectic silicon mask), and B is the structure element used to define the shape and size of dilation and erosion.

[0078] The principle of closing operation is to dilate the image with a structural element, fill small holes in the image and remove small protrusions, and then use the same structural element to erode the dilated image, thereby retaining large objects and eliminating small objects.

[0079] Opening Operation: It is the process of first performing an erosion operation on the image and then performing an expansion operation. The formula is as follows:

[0080] (2)

[0081] in, is the input image (eutectic silicon mask after closing operation), B is the structure element, and the structure element refers to the operation kernel.

[0082] The principle of the opening operation is opposite to that of the closing operation. It first performs an erosion operation to eliminate small objects and protrusions in the image, and then performs an expansion operation to retain large objects and smooth the edges of objects.

[0083] Furthermore, dilation and erosion are basic operations in morphological image processing. Closing and opening are two composite morphological operations. They are usually used for image denoising, edge detection, shape analysis and other tasks.

[0084] Dilation expands the object area in the image. The basic principle is to slide the structural element (also called the kernel) on the image. If any pixel in the area covered by the structural element is the foreground (white), the center position of the structural element is set as the foreground. The formula is:

[0085] (3)

[0086] Where f is the input image, B is the structure element (also called kernel), represents the translation of the structure element B relative to the position x in the image.

[0087] The dilation operation makes objects larger, fills small holes, connects separated objects, and accentuates the boundaries of objects.

[0088] Erosion shrinks the object area in the image. The basic principle is to slide the structure element on the image. Only when all the pixels in the area covered by the structure element are foreground, the center position of the structure element will be set as foreground. The formula is:

[0089] (4)

[0090] Where f is the input image and B is the structure element.

[0091] The erosion operation will make objects smaller, eliminate small objects, fill small depressions and smooth the boundaries of objects.

[0092] In the embodiment of the present invention, the initial silicon segregation region is upsampled to the original size using the nearest neighbor method to obtain the silicon segregation region.

[0093] Further, the initial silicon segregation region is identified, holes in the initial silicon segregation region whose area is smaller than a first preset number of pixels are filled, and / or regions in the initial silicon segregation region whose area is smaller than a second preset number of pixels are eliminated, wherein the first preset number of pixels and the second preset number of pixels can be set based on experience or specific circumstances, and are not specifically limited in the embodiments of the present invention. Preferably, the first preset number of pixels is 800, and the second preset number of pixels is 3000.

[0094] Furthermore, in digital image processing, a connected domain refers to a set of pixels with the same pixel value and adjacent to each other. In layman's terms, a connected domain can be regarded as a connected area or block in an image, where these pixels are adjacent in space and have the same properties, such as grayscale value.

[0095] Connected region labeling means finding and marking all connected regions in an image. This process usually uses a pixel-based scanning algorithm, of which the two most commonly used methods are the scanning line algorithm and the row-by-row and column-by-column scanning algorithm.

[0096] By identifying the connected domain of the background of the digital image, the holes in the digital image can be obtained and post-processed according to the size of the holes.

[0097] By identifying the connected domain of the foreground of the digital image, the object in the foreground of the digital image can be obtained and post-processed according to the size of the object.

[0098] S104, determining a dendrite diagram based on the eutectic silicon mask, and calculating a secondary dendrite spacing and / or a microscopic grain size based on the dendrite diagram.

[0099] In the embodiment of the present invention, the traditional secondary dendrite spacing measurement method mainly relies on manually measuring the total length d of multiple secondary dendrites, and then calculating SDAS (statistical value of secondary dendrite spacing) by dividing it by the number of dendrites n, that is, SDAS = d / n. However, this method has a series of problems, including being subject to manual experience, resulting in greater subjectivity, longer analysis time, limited local statistics, and low accuracy.

[0100] In order to overcome these limitations, a more efficient secondary dendrite spacing measurement method is proposed based on digital image processing in the embodiment of the present invention, which realizes automatic and rapid evaluation and improves the measurement accuracy. The method adopts a segmentation method based on topological theory and mathematical morphology, and directly obtains the closed dendrite boundary and dendrite spacing through distance transformation and watershed algorithm. The implementation steps are generally divided into the following key steps:

[0101] Taking the eutectic silicon mask as the foreground, distance transformation calculation is performed to obtain a distance transformation map (an image of the same size as the original image, in which the gray value of each pixel represents the distance from the pixel to the target);

[0102] In the embodiment of the present invention, the distance transform in digital image processing is a commonly used image processing technology, which is used to calculate the distance from each pixel in the image to the specified target. Distance transform is widely used in the fields of morphological processing, image segmentation, edge detection, etc. The basic principle of distance transform is to calculate the distance from each pixel in the image to the target area (usually the target boundary). The distance can be calculated according to different metrics, such as Euclidean distance, Manhattan distance, etc. The result of the distance transform is a distance transform image with the same size as the original image.

[0103] In the embodiment of the present invention, taking the Euclidean distance transform as an example, for pixels in a two-dimensional image , the Euclidean distance can be calculated by the following formula:

[0104] (5)

[0105] in, is a pixel in the target area, is the Euclidean distance from pixel P to the target area.

[0106] Furthermore, the distance transformation graph is segmented based on the watershed algorithm to obtain a dendrite graph. The specific processing process is as follows:

[0107] The watershed algorithm is an algorithm commonly used for image segmentation. Its basic principle is to regard the image as terrain and the grayscale value in the image as height, and then find the dividing line between different areas in the image by simulating the flow of water.

[0108] Image gradient calculation: First, the gradient of the distance transform map is calculated, and operators such as Sobel and Prewitt are usually used to calculate the gradient amplitude and direction of the image.

[0109] Marking regions: Based on the gradient information, the pixels in the image are divided into different regions and each pixel is marked. Usually, the local minimum point (for example, the local minimum value below the threshold) is used as the starting point of the watershed, and these points are marked as different regions.

[0110] Generate marking lines: By simulating the process of water flow, starting from the local minimum point and gradually flowing to the local maximum point, dividing lines are formed. These dividing lines are called watershed lines.

[0111] Watershed transformation: According to the generated watershed line, the pixels in the image are divided into different areas to form the final segmentation result.

[0112] Merge regions: For regions that are too small or unreasonable in the segmentation results, post-processing can be performed, such as merging regions to further optimize the segmentation results.

[0113] Furthermore, for the dendrite diagram, which contains multiple dendrites, the width of each dendrite in the dendrite diagram is calculated, and the average width of all dendrites is used as the secondary dendrite spacing, wherein the width of each dendrite can be calculated by an external rectangular frame, multiple calculations and averaging, or other preferred calculation methods. In the embodiment of the present invention, the specific calculation method is not limited.

[0114] In addition, for the dendrite diagram, a circular area with a preset circumference is drawn based on a preset magnification, and the number of intersections between the circular area and the connected domain in the dendrite diagram is counted, where the number is N, wherein the preset magnification and the preset circumference are set based on the standard of Section 14 of ASTM E112, the preset magnification is 100, and the preset circumference is 2500 μm.

[0115] The quantity is passed to a preset microscopic grain size calculation formula to obtain the microscopic grain size G, wherein the preset microscopic grain size calculation formula is as follows:

[0116] (6)

[0117] Furthermore, with regard to the calculation of the silicon segregation area, the secondary dendrite spacing and the microscopic grain size, one, two or all of them are selected for calculation based on a specific scenario.

[0118] The present invention discloses a digital characterization method for hypoeutectic aluminum-silicon alloys, including: obtaining a metallographic micrograph of a hypoeutectic aluminum-silicon alloy, preprocessing the metallographic micrograph to obtain a target metallographic micrograph; passing the target metallographic micrograph to a preset eutectic aluminum-silicon extraction model to obtain a eutectic silicon mask, wherein the eutectic silicon mask is the same size as the target metallographic micrograph; determining the silicon segregation region based on the eutectic silicon mask, or; determining a dendrite diagram based on the eutectic silicon mask, and calculating the secondary dendrite spacing and / or microcrystalline grain size based on the dendrite diagram. The above process obtains a eutectic silicon mask based on a preset eutectic aluminum-silicon extraction model, and the silicon segregation region, secondary dendrite spacing and microcrystalline grain size can be determined based on the eutectic aluminum-silicon mask, all of which are completed automatically and no longer based on manual work, thus avoiding the problems of strong subjectivity of manual experience, long analysis time, limited local statistics and low precision in the prior art.

[0119] In the embodiment of the present invention, a deep learning model (Unet) is used to automatically extract and identify eutectic silicon segregation, ensuring efficient processing of metallographic structures. Through morphological processing, eutectic silicon segregation identification is achieved, and through the use of distance transformation and watershed algorithm, the automatic measurement of secondary dendrite spacing and microscopic grain size is achieved, which greatly reduces the subjectivity of manual experience and improves analysis efficiency.

[0120] Based on the above-mentioned digital characterization method of a hypoeutectic aluminum-silicon alloy, an embodiment of the present invention further provides a digital characterization device for a hypoeutectic aluminum-silicon alloy, and the structural block diagram of the device is as follows: Figure 2 As shown, including:

[0121] A preprocessing module 201 , an extraction module 202 , a first determination module 203 and a second determination module 204 .

[0122] in,

[0123] The preprocessing module 201 is used to obtain a metallographic micrograph of the hypoeutectic aluminum-silicon alloy, and preprocess the metallographic micrograph to obtain a target metallographic micrograph;

[0124] The extraction module 202 is used to transfer the target metallographic structure microphotograph to a preset eutectic aluminum-silicon extraction model to obtain a eutectic silicon mask, wherein the eutectic silicon mask has the same size as the target metallographic structure microphotograph;

[0125] The first determination module 203 is used to determine the silicon segregation area based on the eutectic silicon mask, or;

[0126] The second determination module 204 is configured to determine a dendrite diagram based on the eutectic silicon mask, and calculate a secondary dendrite spacing and / or a microscopic grain size based on the dendrite diagram.

[0127] The present invention discloses a digital characterization device for hypoeutectic aluminum-silicon alloy, including: obtaining a metallographic micrograph of hypoeutectic aluminum-silicon alloy, preprocessing the metallographic micrograph to obtain a target metallographic micrograph; passing the target metallographic micrograph to a preset eutectic aluminum-silicon extraction model to obtain a eutectic silicon mask, wherein the eutectic silicon mask is the same size as the target metallographic micrograph; determining the silicon segregation region based on the eutectic silicon mask, or; determining a dendrite diagram based on the eutectic silicon mask, and calculating the secondary dendrite spacing and / or microcrystalline grain size based on the dendrite diagram. The above process obtains a eutectic silicon mask based on a preset eutectic aluminum-silicon extraction model, and the silicon segregation region, secondary dendrite spacing and microcrystalline grain size can be determined based on the eutectic aluminum-silicon mask, all of which are completed automatically and no longer based on manual work, thus avoiding the problems of strong subjectivity of manual experience, long analysis time, limited local statistics and low precision in the prior art.

[0128] In the embodiment of the present invention, the preprocessing module 201 includes:

[0129] Transformation unit, Augmentation unit, and Transformation and Normalization unit.

[0130] in,

[0131] The conversion unit is used to convert the pixel length in the metallographic structure microphoto into an actual length to obtain a first metallographic structure microphoto;

[0132] The enhancement unit is used to perform data enhancement processing on the first metallographic structure micrograph to obtain a second metallographic structure micrograph;

[0133] The conversion and normalization unit is used to convert the second metallographic structure microphotograph into a single-channel image, and perform normalization processing on the single-channel image to obtain a target metallographic structure microphotograph.

[0134] In the embodiment of the present invention, the first determining module 203 includes:

[0135] Downsampling unit, computing unit and upsampling unit.

[0136] in,

[0137] The down-sampling unit is used to down-sample the eutectic silicon mask to obtain a sampled eutectic silicon mask;

[0138] The operation unit is used to sequentially perform closing operation and opening operation on the sampled eutectic silicon mask to obtain an initial silicon segregation region;

[0139] The up-sampling unit is used to up-sample the initial silicon segregation region to obtain a silicon segregation region.

[0140] In the embodiment of the present invention, the second determining module 204 includes:

[0141] Transformation unit and segmentation unit.

[0142] in,

[0143] The transformation unit is used to perform distance transformation on the eutectic silicon mask to obtain a distance transformation map;

[0144] The segmentation unit is used to segment the distance transformation graph based on a watershed algorithm to obtain a dendrite graph.

[0145] The digital representation device includes a processor and a memory. The preprocessing module, extraction module, first determination module and second determination module are all stored in the memory as program units. The processor executes the program units stored in the memory to implement corresponding functions.

[0146] The processor contains a kernel, which calls the corresponding program unit from the memory. One or more kernels can be set, and the kernel parameters can be adjusted to solve problems such as strong subjectivity of manual experience, long analysis time, limited local statistics, and low accuracy.

[0147] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0148] An embodiment of the present invention provides a computer storage medium having a program stored thereon, which, when executed by a processor, implements the steps of the digital characterization method of the hypoeutectic aluminum-silicon alloy.

[0149] An embodiment of the present invention provides a processor, which is used to run a program, wherein the program executes the steps of the digital characterization method of the hypoeutectic aluminum-silicon alloy when it is run.

[0150] An embodiment of the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of a digital characterization method for a hypoeutectic aluminum-silicon alloy provided in an embodiment of the present application.

[0151] An embodiment of the present invention provides a device, the structural block diagram of which is as follows Figure 3 As shown, the device includes: a processor 301, a storage medium 302, and a program stored in the storage medium 302 and executable on the processor 302. When the processor 301 executes the program, the following steps are implemented:

[0152] Obtaining a metallographic micrograph of a hypoeutectic aluminum-silicon alloy, and preprocessing the metallographic micrograph to obtain a target metallographic micrograph;

[0153] Passing the target metallographic structure micrograph to a preset eutectic aluminum-silicon extraction model to obtain a eutectic silicon mask, wherein the eutectic silicon mask has the same size as the target metallographic structure micrograph;

[0154] determining silicon segregation regions based on the eutectic silicon mask, or;

[0155] A dendrite diagram is determined based on the eutectic silicon mask, and a secondary dendrite spacing and / or a microscopic grain size is calculated based on the dendrite diagram.

[0156] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0157] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0158] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0159] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0160] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0161] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0162] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0163] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0164] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0165] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A digital characterization method for hypoeutectic aluminum-silicon alloy, characterized in that: include: Obtaining a metallographic micrograph of a hypoeutectic aluminum-silicon alloy, and preprocessing the metallographic micrograph to obtain a target metallographic micrograph; Passing the target metallographic structure micrograph to a preset eutectic aluminum-silicon extraction model to obtain a eutectic silicon mask, wherein the eutectic silicon mask has the same size as the target metallographic structure micrograph; determining silicon segregation regions based on the eutectic silicon mask, or; A dendrite diagram is determined based on the eutectic silicon mask, and a secondary dendrite spacing and / or a microscopic grain size is calculated based on the dendrite diagram.

2. The digital characterization method of hypoeutectic aluminum-silicon alloy according to claim 1, characterized in that: Preprocessing the metallographic structure micrograph to obtain a target metallographic structure micrograph includes: Converting the pixel length in the metallographic structure microphotograph into an actual length to obtain a first metallographic structure microphotograph; Performing data enhancement processing on the first metallographic structure micrograph to obtain a second metallographic structure micrograph; The second metallographic structure microphotograph is converted into a single-channel image, and the single-channel image is normalized to obtain a target metallographic structure microphotograph.

3. The digital characterization method of hypoeutectic aluminum-silicon alloy according to claim 1, characterized in that: Determining a silicon segregation region based on the eutectic silicon mask includes: Down-sampling the eutectic silicon mask to obtain a sampled eutectic silicon mask; performing closing operation and opening operation on the sampled eutectic silicon mask in sequence to obtain an initial silicon segregation region; The initial silicon segregation region is upsampled to obtain a silicon segregation region.

4. The digital characterization method of hypoeutectic aluminum-silicon alloy according to claim 3, characterized in that: Also includes: The initial silicon segregation region is identified, holes in the initial silicon segregation region whose area is smaller than a first preset number of pixels are filled, and / or regions in the initial silicon segregation region whose area is smaller than a second preset number of pixels are eliminated.

5. The digital characterization method of hypoeutectic aluminum-silicon alloy according to claim 1, characterized in that: Determining a dendrite map based on the eutectic silicon mask includes: Performing distance transformation on the eutectic silicon mask to obtain a distance transformation map; The distance transformation graph is segmented based on a watershed algorithm to obtain a dendrite graph.

6. The digital characterization method of hypoeutectic aluminum-silicon alloy according to claim 5, characterized in that: Calculating the microscopic grain size based on the dendrite diagram includes: Based on a preset magnification, a circular area with a preset perimeter is drawn, and the number of intersections between the circular area and the connected domains in the dendrite graph is counted; The quantity is passed to a preset microscopic grain size calculation formula to obtain the microscopic grain size.

7. A digital characterization device for hypoeutectic aluminum-silicon alloy, characterized in that: include: A preprocessing module, used to obtain a metallographic micrograph of the hypoeutectic aluminum-silicon alloy, and preprocess the metallographic micrograph to obtain a target metallographic micrograph; An extraction module, used for transferring the target metallographic structure microphotograph to a preset eutectic aluminum silicon extraction model to obtain a eutectic silicon mask, wherein the eutectic silicon mask has the same size as the target metallographic structure microphotograph; a first determination module, configured to determine a silicon segregation region based on the eutectic silicon mask, or; The second determination module is used to determine a dendrite diagram based on the eutectic silicon mask, and calculate a secondary dendrite spacing and / or a microscopic grain size based on the dendrite diagram.

8. The digital characterization device for hypoeutectic aluminum-silicon alloy according to claim 7, characterized in that: The preprocessing module comprises: A conversion unit, used for converting the pixel length in the metallographic structure microphotograph into an actual length to obtain a first metallographic structure microphotograph; An enhancement unit, used for performing data enhancement processing on the first metallographic structure micrograph to obtain a second metallographic structure micrograph; The conversion and normalization unit is used to convert the second metallographic structure microphotograph into a single-channel image and perform normalization processing on the single-channel image to obtain a target metallographic structure microphotograph.

9. The digital characterization device for hypoeutectic aluminum-silicon alloy according to claim 7, characterized in that: The first determining module comprises: A down-sampling unit, used for down-sampling the eutectic silicon mask to obtain a sampled eutectic silicon mask; A calculation unit, used for sequentially performing a closing operation and an opening operation on the sampled eutectic silicon mask to obtain an initial silicon segregation region; An up-sampling unit is used to up-sample the initial silicon segregation region to obtain a silicon segregation region.

10. The digital characterization device for hypoeutectic aluminum-silicon alloy according to claim 7, characterized in that: The second determining module comprises: A transformation unit, used for performing distance transformation on the eutectic silicon mask to obtain a distance transformation map; The segmentation unit is used to segment the distance transformation graph based on a watershed algorithm to obtain a dendrite graph.

11. A storage medium, characterized in that: The storage medium includes a stored program, wherein the program executes the steps of the digital characterization method for the hypoeutectic aluminum-silicon alloy according to any one of claims 1 to 6.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the digital characterization method of the hypoeutectic aluminum-silicon alloy described in any one of claims 1 to 6 are implemented.

Images