Foam concrete image processing method based on sub-pixel edge reconstruction

Abstract

The invention discloses a foam concrete image processing method based on sub-pixel edge reconstruction, and particularly relates to the field of foam concrete image processing.The method comprises the steps that an image of foam concrete is obtained, and a boundary response feature set containing edge direction information, edge gradient change information and local intensity comparison information is extracted; performing edge structure classification on the image region based on boundary response feature set combined threshold judgment; and aiming at the image region which is judged to be the local direction convergence mutation region, identifying a cross mutation point set by fitting a direction tensor convergence trend, and executing directionally-guided corrosion and expansion image processing. By constructing a structure judgment variable based on image boundary response characteristics and guiding a corrosion expansion path and a vein backtracking path to execute sub-pixel-level aperture boundary separation, image structure processing and contour reconstruction of a porous adhesion area are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of foamed concrete image processing. More specifically, the present invention relates to a foamed concrete image processing method based on sub-pixel edge reconstruction. Background Art

[0002] In the image processing of the surface of foamed concrete, the conventional pore size recognition method based on pixel-level binarization and connected component extraction is difficult to deal with the problem of void adhesion caused by fuzzy pore boundaries, close fitting of pore walls, or local structural collapse; Especially at the microscale, multiple actual independent pores often present as a continuous high-gray-scale area, resulting in the traditional algorithm misidentifying them as a single large pore, destroying the true statistical structure of the pore size distribution, and further affecting the accurate evaluation of porosity, mechanical properties, and material looseness; Therefore, the current problem in the image processing of the surface of foamed concrete is: how to separate the porous boundary regions that are mismerged due to structural blurring at the sub-pixel scale. Summary of the Invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a foamed concrete image processing method based on sub-pixel edge reconstruction. By constructing a structure determination variable based on the image boundary response characteristics, and guiding the dual strategies of the erosion and dilation path and the vein backtracking path to perform sub-pixel level pore size boundary separation, so as to realize the image structure processing and contour reconstruction of the porous adhesion region.

[0004] To achieve the above object, the present invention provides the following technical solution: A foamed concrete image processing method based on sub-pixel edge reconstruction, including: Obtain an image of foamed concrete, and extract a set of boundary response characteristics including edge direction information, edge gradient change information, and local intensity contrast information; perform edge structure classification on the image region based on the combined threshold judgment of the set of boundary response characteristics; For the image region determined to be a local direction convergence mutation region, identify the set of cross mutation points by fitting the convergence trend of the direction tensor, and perform image processing of erosion and dilation guided by directionality; For the image region determined to be a continuous closed high-consistency region, construct a structure potential field and perform a combined judgment of path tension and boundary coincidence degree, identify the true fracture path, and perform topological division accordingly; For the pore size structure maps respectively output by the image processing of the erosion and dilation path and the vein backtracking path, perform cross matching judgment on the three consistency scoring items in the boundary matching scoring matrix, and selectively integrate or retain the structure difference regions according to the fusion conditions; Construct a structure determination mechanism by fusing paths and scoring results, perform path update and callback correction on the boundaries of the regions marked as blurred in the image, and achieve the recognition of the aperture structure and the optimization of the image processing process.

[0005] In a preferred embodiment, obtain the original image data of the foamed concrete image, perform standardized grayscale processing and multi-scale edge enhancement operations on the original image data to obtain a set of boundary response features including edge direction information, edge gradient change information, and local intensity contrast information; Based on the set of boundary response features, extract three structure determination variables, where the structure determination variables include the boundary direction consistency metric B 1 , the boundary strength gradient variation value B 2 and the regional edge closure ratio B 3 ; If B 1 is less than the boundary direction consistency metric threshold T 1 and B 2 is greater than the boundary strength gradient variation threshold T 2 , then the image region is marked as a locally directionally convergent mutation region; if B 1 is greater than or equal to T 1 and B 3 is greater than the regional edge closure ratio threshold T 3 , then the image region is marked as a continuously closed high-consistency region; if the three are in a critical crossing state, then the image region is marked as a boundary feature determination blurred region; Temporarily store the image blocks marked as boundary feature determination blurred regions as blurred region masks, and jointly include them in the fusion judgment process with the image blocks marked as locally directionally convergent mutation regions and the image blocks marked as continuously closed high-consistency regions to perform unified determination of the path attribution of the three types of regions.

[0006] In a preferred embodiment, input all locally directionally convergent mutation regions into the structure direction tensor extraction process, construct a tensor atlas of the corresponding main axis through the set of boundary response features, perform a convergence trend fitting of the main axis direction on the tensor atlas to generate a direction tensor convergence map, and identify the aggregated boundary region; Based on image analysis, identify the set of direction gradient mutation points in the aggregated boundary region as the fracture candidate subset, and perform tensor cross-angle analysis on each candidate point in the fracture candidate subset; if the cross angle is greater than the preset fracture criterion, then the point is identified as a valid fracture point and used to generate a fracture-induced layer; Perform a directional tensor weighted erosion operation based on the fracture-induced layer, and then perform a limited dilation operation that maintains the main axis direction to preserve the boundary continuity and spatial consistency formed by the fracture; perform a structural contour extraction operation on the processed region, form a contour set by combining all the extracted boundary contours, and record the processing path label corresponding to each contour at the same time; Output the obtained contour set as the initial pore size structure map I under the erosion and dilation path.

[0007] In a preferred embodiment, input all continuous closed high-consistency regions into the structural potential construction process, generate a sub-pixel level structural potential field based on the edge strength and closed stability, perform a guided path search in the structural potential field to obtain a set of potential fracture paths, and calculate the path tension value and the structural boundary overlap degree for each path in turn; If the path tension value is greater than the path tension threshold T 4 and the structural boundary overlap degree is less than the structural boundary overlap degree threshold T 5 , then this path is confirmed as a real fracture path and used as a boundary cleavage trigger path; Perform a topological structure reconstruction operation on the original image using the confirmed real fracture path, and divide it into multiple logically porous sub-blocks; mark the pore structures of the divided porous sub-blocks with path labels and output them as the initial pore size structure map II under the vein backtracking path.

[0008] In a preferred embodiment, merge the sets of the initial pore size structure map I and the initial pore size structure map II of the erosion and dilation path and the vein backtracking path, and construct a candidate boundary set for the boundary pairs in the two maps; for each pair of candidate boundaries in the candidate boundary set, calculate the boundary coincidence degree R 1 , the direction alignment degree R 2 and the structural center offset amount R 3 , and form a boundary matching score matrix; In the boundary matching score matrix, if in a certain boundary pair R 1 is greater than the boundary coincidence degree threshold T 6 , and R 2 and R 3 are less than the direction alignment degree threshold T 7 and the structural center offset amount threshold T 8 , then this boundary pair is determined as a fusion boundary, perform layer merging and assign a unified number; If the boundary matching score matrix does not meet the fusion boundary determination criteria, then this image region is retained as a boundary feature determination fuzzy region; perform sub-pixel curve fitting on the boundaries of all determined image regions, and output a vector boundary set and the corresponding path label information.

[0009] In a preferred embodiment, path tags, scoring results of a boundary matching scoring matrix, and a structure status table are established with each boundary structure in the contour set as the main index. The structure status table is expanded based on the scoring results of the boundary matching scoring matrix by introducing a path execution status and a region fusion determination result. In the structure status table, the original policy path and the scoring matching value corresponding to each boundary structure are recorded; For all image regions marked as regions with ambiguous boundary feature determination, evaluate the path determination confidence and the scoring consistency difference, and determine whether there is aperture recognition uncertainty; If the path confidence is less than the rejudgment threshold T 9 and the scoring consistency difference is greater than the conflict tolerance threshold T 10 , then re - perform boundary structure determination based on three structure determination variables to complement the missing or misclassified aperture boundaries; According to the new round of structure determination results, re - specify the image structure path, and execute the corresponding image processing method to obtain the updated aperture boundary extraction result; the image structure path includes an erosion - dilation path or a vein backtracking path; replace the corresponding region in the original boundary layer with the updated image structure boundary, and synchronously update its corresponding path identification information.

[0010] In a preferred embodiment, define to represent the joint classification result of the three structure determination variables corresponding to the pixel point in the image; define to represent the boundary direction consistency measurement value B of the pixel point 1 ; define to represent the boundary strength gradient variation value B of the pixel point 2 ; define to represent the regional edge closure ratio B to which the point 3 in the image belongs, which is used to measure whether the contour is closed and complete; ; ; ; ; ; where represents the neighborhood set of the pixel point ; represents the main edge direction angle of the pixel point in the neighborhood; represents the current pixel point The edge principal direction; is a point The direction divergence suppression factor calculated in the local tensor field; is a multi-scale image pyramid set; represents at the scale image, the gradient intensity modulus value of the pixel point ; is the anisotropy adjustment factor; represents all the pixel point sets on the boundary paths extracted starting from the pixel point through the boundary tracking algorithm; is the local connection score of the point on the path; is a point The set of multiple closed boundary candidate paths formed by the region where it is located; is the closed cost function; is a very small positive number; is the joint classification function.

[0011] In a preferred embodiment, for the image region structure in the local direction convergence mutation region, by constructing a non-linear path mutation metric value that fuses the tensor direction torsion rate, local tension gradient variation, and perturbation diffusion response , construct a fracture induction function to determine potential boundary fracture points in the image, and accordingly guide the subsequent erosion and dilation processes to form a structural contour path; ; ; Among them, the fracture induction function represents whether the pixel point in the image is marked as a structure fracture induction point, and the output is a boolean value; respectively represent the principal direction angles extracted from the current pixel point and its neighboring pixel points in the tensor map; represents the directed angle difference between the principal directions; is the tension function; are respectively the gradients of the structural tension in the horizontal and vertical directions; is the direction consistency function; is the mixed second-order partial derivative of the direction consistency; is the direction mutation intensity factor; is the local response rate of the direction diffusion tensor; additionally, in the formula represents the non-linear expansion term for constructing the structural perturbation propagation intensity; is the structural mutation joint determination threshold; Input all continuous closed high - consistency regions into the structural potential construction process to generate the structural potential trajectory function and the reference guiding path function , and combine with the closed perturbation function to construct the cleavage scoring function ; if the score is higher than the preset threshold , then this point is marked as a real fracture path point, and the cleavage path set is output for topological division and pore reconstruction; ; ; In in the formula, represents the sub - pixel coordinates of the currently processed pixel in the image; the structural potential trajectory function represents the potential - driven path within the closed region; , , and represent the first - order derivative along ; , is the second - order partial derivative of , used to calculate the local structural curvature change; the reference guiding path function is used as the comparison reference path for the potential path ; , is the derivative of in two directions; the closed perturbation function represents the sub - pixel perturbation tension residual function of the path points in the closed region; the cleavage scoring function is a scoring function that fuses three factors and is used as the basis for judging whether a fracture path is formed; the preset threshold is the path cleavage scoring determination threshold; is the binary determination function for the fracture path, and the output of

[0012] In a preferred embodiment, candidate boundaries are constructed for the initial pore size structure map one and the initial pore size structure map two extracted from the corrosion expansion path and the vein backtracking path respectively, and a boundary matching scoring matrix is constructed based on three types of scoring quantities: the boundary coincidence degree R 1 , the direction alignment degree R 2 and the structural center offset amount R 3 to judge the boundary fusion condition; The boundary coincidence degree R is defined by1 Score of: ; Define the degree of direction alignment R through 2 Score of: ; Define the structural center offset R through 3 Score of: ; In 、 、 、 in the formula represents the pair of candidate boundary pairs; is the pair of candidate boundary pairs The overlapping part length of the initial aperture structure map one and the initial aperture structure map two; is the pair of candidate boundary pairs The total length of the union of the initial aperture structure map one and the initial aperture structure map two; is the pair of candidate boundary pairs The difference in the image edge gradient change in the overlapping area; is the pair of candidate boundary pairs The corresponding topological fracture complexity index; , is the pair of candidate boundary pairs The main direction angle in the erosion-dilation path map and the vein backtracking path map; is the pair of candidate boundary pairs The average curvature of; , is the pair of candidate boundary pairs The horizontal and vertical offset of the geometric center of; is the pair of candidate boundary pairs The structural symmetry coefficient of.

[0013] In a preferred embodiment, for the image block marked as the boundary feature determination fuzzy area, establish a path confidence index and a scoring consistency difference index indexed by the boundary structure, and build a dynamic callback mechanism based on the two to update the image path and boundary judgment; define the path confidence as ; define the scoring consistency difference as ; ; ; Where is the boundary layer tension response extracted at the pixel point ; is a pixel point The difference value of the image processing strategies under two paths; is a pixel point The score under the erosion and dilation path; is a pixel point The score under the vein backtracking path.

[0014] The technical effects and advantages of the present invention: Through three types of structure determination variables, namely boundary direction consistency measurement, boundary strength gradient variation, and regional edge closure ratio, it is possible to classify local direction convergence mutation regions and closed high-consistency regions, achieve structural correction of misjudgments of adjacent pore adhesions, thereby breaking through the problem of mis-merging multiple pore diameters in traditional pixel-level processing, and ensuring the granularity and authenticity of pore diameter recognition; Adopt structure tensor fitting to construct the direction convergence trend, form a fracture-induced layer, and perform directional erosion and dilation operations under the guidance of the main axis; at the same time, guide path search through the structure potential energy field, calculate the path tension value and the structure boundary overlap degree to identify the true cleavage path, realize the dual-path extraction mechanism of the image structure contour, and enhance the physical rationality and direction adaptability of path construction; Fuse the pore diameter maps output by the erosion and dilation path and the vein backtracking path, form a matching score matrix by constructing three scoring factors: boundary coincidence degree, direction alignment degree, and structural center offset amount, and systematically judge the fusion conditions of candidate boundaries to solve the problems of boundary inconsistency or redundant overlap under traditional multi-path processing; For the structurally uncertain regions, by recording the boundary path scores and execution status, combining the path confidence and score consistency difference indicators to construct a dynamic callback mechanism, re-specify the processing path according to the structure determination variables, and finally realize the path callback and boundary reconstruction of the fuzzy regions, improving the flexibility of the overall image processing; By integrating multi-scale boundary response features, tensor main axis trend, structure potential energy function, and score matrix mechanism, realize multi-dimensional cognitive modeling of the image structure from pixel level to sub-pixel level, from geometric features to path topology, improve the characterization ability of complex pore morphology, and is applicable to image structure analysis and pore diameter evaluation of various types of foamed concrete materials. Brief Description of the Drawings

[0015] Figure 1 is the method flow framework diagram of the present invention. Detailed Embodiments

[0016] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0017] Referring to the attached Figure 1 drawings, a method for processing images of foamed concrete based on sub-pixel edge reconstruction according to an embodiment of the present invention includes: Obtaining an image of foamed concrete, and extracting a boundary response feature set including edge direction information, edge gradient change information, and local intensity contrast information; performing edge structure classification on the image region based on the joint threshold judgment of the boundary response feature set; For the image region determined to be a locally directionally convergent mutation region, identifying a set of cross-mutation points by fitting the convergence trend of the direction tensor, and performing image processing of directional guided erosion and dilation, so as to achieve fracture splitting and contour extraction while maintaining structural continuity; For the image region determined to be a continuous closed high-consistency region, constructing a structural potential field and performing a joint judgment of path tension and boundary coincidence degree, identifying the true fracture path and performing topological division accordingly, and finally completing the contour reconstruction of the veined pores; For the pore size structure maps respectively output by the image processing of the erosion and dilation path and the vein backtracking path, performing cross-matching judgment on the three consistency scoring items in the boundary matching scoring matrix, and selectively integrating or retaining the structural difference regions according to the fusion conditions; By fusing the path and the scoring result to construct a structural determination mechanism, performing path update and callback correction on the boundary marked as a fuzzy region in the image, and realizing the identification of the pore size structure and the optimization of the image processing flow.

[0018] Obtaining the original image data of the foamed concrete image, performing standardized gray processing and multi-scale edge enhancement operations on the original image data, and obtaining a boundary response feature set including edge direction information, edge gradient change information, and local intensity contrast information; Based on the boundary response feature set, extracting three structural determination variables, and the structural determination variables include the boundary direction consistency measurement value B 1 , the boundary strength gradient variation value B 2 and the regional edge closure ratio B 3 ; If B 1 is less than the boundary direction consistency measurement threshold T 1 and B 2 is greater than the boundary strength gradient variation threshold T 2, the image region is marked as a locally directionally convergent mutation region; if B 1 is greater than or equal to T 1 and B 3 is greater than the regional edge closure ratio threshold T 3 , the image region is marked as a continuously closed high-consistency region; if the three are in a critical crossing state, the image region is marked as a boundary feature determination fuzzy region; The image blocks marked as boundary feature determination fuzzy regions are temporarily stored as fuzzy region masks, and are jointly incorporated into the fusion judgment process with the image blocks marked as locally directionally convergent mutation regions and the image blocks marked as continuously closed high-consistency regions to perform unified determination of the path attribution of the three types of regions.

[0019] All locally directionally convergent mutation regions are input into the structural direction tensor extraction process. Through the boundary response feature set, a tensor map of the corresponding principal axis is constructed. The convergence trend fitting of the principal axis direction of the tensor map is performed to generate a direction tensor convergence map, and the aggregated boundary region is identified; Based on image analysis, the set of direction gradient mutation points in the aggregated boundary region is identified as the fracture candidate subset, and tensor cross-angle analysis is performed on each candidate point in the fracture candidate subset; if the cross angle is greater than the preset fracture criterion, the point is identified as a valid fracture point and is used to generate a fracture-induced layer; Based on the fracture-induced layer, a direction tensor weighted erosion operation is performed, and then a limited dilation operation that maintains the principal axis direction is performed to retain the boundary continuity and spatial consistency formed by the fracture; a structural contour extraction operation is performed on the processed region, and all the extracted boundary contours are combined into a contour set, and the processing path label corresponding to each contour is recorded at the same time; The obtained contour set is output as the initial aperture structure map one under the erosion and dilation path.

[0020] All continuously closed high-consistency regions are input into the structural potential energy construction process. According to the edge strength and closure stability, a sub-pixel level structural potential energy field is generated. A guided path search is performed in the structural potential energy field to obtain a set of potential fracture paths, and the path tension value and the structural boundary overlap degree are calculated for each path in turn; If the path tension value is greater than the path tension threshold T 4 and the structural boundary overlap degree is less than the structural boundary overlap degree threshold T 5 , then the path is confirmed as a real fracture path and serves as a boundary cleavage trigger path; Perform topological structure reconstruction operations on the original image using the confirmed true fracture path, and divide it into multiple logically porous sub-blocks; mark the pore structures of the obtained porous sub-blocks with path labels and output them as the initial pore size structure map II under the vein backtracking path; in addition, the calculation of the path tension value is based on the consistency between the pixel point tension response intensity on each potential fracture path in the structural potential field and its gradient direction along the path direction, and is obtained by weighted accumulation in combination with the continuity of the guiding directions of each point on the path; the calculation of the structural boundary overlap degree is based on the ratio between the length of the geometric coincidence area between the potential fracture path and the current image boundary contour and the total length of the path, specifically evaluating whether the path effectively covers the original boundary structure, thus serving as an important basis for determining the true fracture path.

[0021] Merge the set of the initial pore size structure map I of the erosion and dilation path and the initial pore size structure map II of the vein backtracking path, and construct a candidate boundary set for the boundary pairs in the two maps; for each pair of candidate boundaries in the candidate boundary set, calculate the boundary coincidence degree R 1 , the direction alignment degree R 2 and the structural center offset R 3 , and form a boundary matching score matrix; In the boundary matching score matrix, if R 1 in a certain boundary pair is greater than the boundary coincidence degree threshold T 6 , and R 2 and R 3 are less than the direction alignment degree threshold T 7 and the structural center offset threshold T 8 , then this boundary pair is determined as a fused boundary, perform layer merging and assign a unified number; If the boundary matching score matrix does not meet the fused boundary determination criteria, then this image region is retained as a boundary feature determination fuzzy region; perform sub-pixel curve fitting on the boundaries of all determined image regions, and output a vector boundary set and corresponding path label information.

[0022] Establish a path label, the scoring result of the boundary matching score matrix, and a structure status table with each boundary structure in the contour set as the main index. The structure status table is extended based on the scoring result of the boundary matching score matrix by introducing the path execution status and the region fusion determination result, and record the original strategy path and scoring matching value corresponding to each boundary structure in the structure status table; For all image regions marked as boundary feature determination fuzzy regions, evaluate their path determination confidence and scoring consistency difference, and judge whether there is pore size recognition uncertainty; If the path confidence is less than the rejudgment threshold T 9 and the scoring consistency difference is greater than the conflict tolerance threshold T 10, then re - perform the boundary structure determination based on the three structure determination variables to complete the missing or mis - classified aperture boundaries; According to the results of the new round of structure determination, re - specify the image structure path and execute the corresponding image processing method to obtain the updated aperture boundary extraction result; the image structure path includes the erosion - dilation path or the vein back - tracing path; replace the corresponding area in the original boundary layer with the updated image structure boundary and synchronously update its corresponding path identification information, and finally implement the adaptive closed - loop aperture recognition method composed of five stages: image region structure determination, strategy path judgment, path execution, boundary fusion, and callback correction.

[0023] It should be noted that in the formula structure involved in this solution, the dimensionless term can be used as a proportional or structural adjustment factor. When combined with the quantity with units, it only plays a role in numerical scaling and does not introduce new physical dimensions. Therefore, it will not change or confuse the overall unit system of the expression; the combination of such "dimensionless term and quantity with units" can be understood as the composite structure expression form commonly used in mathematical - physical modeling, which conforms to the principle of dimensional consistency and has a clear physical interpretation basis; Secondly, in the formula structure of this solution, if it involves multiple variable terms with different physical units, including but not limited to time - type, mass - type, or energy - type variables, their joint appearance is to express the collaborative modeling relationship of multiple physical mechanisms. Each variable can form a unified structure through function mapping, ratio combination, or normalization adjustment, with clear units and clear meanings. The overall expression conforms to the principle of dimensional consistency and the common norms of engineering modeling; In this solution, if constants, weights, adjustment factors, threshold parameters, proportional coefficients, etc. are designed, they all belong to adjustable control parameters for different application environments. Their values depend on the target device configuration, data input characteristics, and performance optimization goals, and converge and are set within a reasonable range through methods such as model verification, performance constraints, or engineering calibration during the implementation stage; although these parameters do not preset unique values, they have clear adjustment logics and calculation paths, belonging to the deterministic setting process in engineering implementation. The purpose of such setting is to ensure that the solution has both general adaptability and reproducibility and operability, without affecting its technical clarity and feasibility; Define to represent the pixel point in the image The joint classification result of the three structure determination variables corresponding to it is used to determine its structural region category in the image; define to represent the pixel point The boundary direction consistency metric value B of 1 , which is used to measure whether the edge directions in its local neighborhood have the characteristic of main - direction aggregation, The higher the value of represents the stronger the boundary direction consistency; define The boundary strength gradient variation value B 2 , indicating whether there is an obvious intensity jump at this point in the multi-scale image, and used to reflect the severity of the edge structure; define indicating the point in the image The closed ratio B of the region edge to which it belongs 3 , used to measure whether the contour is closed and complete; ; ; ; ; ; where indicating the neighborhood set of the pixel point , and in practical applications includes a square region window for to calculate the direction consistency; indicating the edge main direction angle of the pixel point in the neighborhood. In practical applications the value of is calculated from the image edge gradient direction, and the unit is radians; indicating the edge main direction of the current pixel point , and comparing it with its neighborhood direction is used to judge the direction consistency; is the direction divergence suppression factor calculated for the point in the local tensor field, used to enhance the weighting of the direction difference in the high direction fluctuation region, and reflect the local instability of the direction; is the multi-scale image pyramid set. The multi-scale image pyramid set includes multiple resolution versions such as the original image, the reduced image, and the enlarged image, used to extract the gradient jump behavior consistent across scales; indicating the gradient intensity modulus value of the pixel point in the image at the scale; is the anisotropy adjustment factor. The anisotropy adjustment factor is constructed according to the gradient ratio of the main direction and its orthogonal direction at the scale , and its expression is: ; where indicating the orthogonal direction of the gradient, is added to prevent the denominator from being a minimum positive number of zero. The anisotropy adjustment factor is used to enhance the response of those regions where the gradient direction jumps violently; indicating from the pixel point through the boundary tracking algorithm Starting from the set of pixel points on all boundary paths extracted, the boundary tracking algorithm includes edge connection, Canny edge growth, etc.; For the points on the path The local connection score, the local connection score includes three weighted dimensions: boundary direction consistency, local gray contrast, local gradient jump amplitude, and is fused to form a single-point connection quality value; For the point The set of multiple closed boundary candidate paths formed by the region where it is located, and the closed boundary candidate path set is used to analyze its boundary closure degree; The closed cost function, the closed cost function represents the minimum structural cost value among multiple candidate closed paths, and includes various sub-items such as closed side length, contour curvature, direction jump penalty, etc. The closed cost function is also used to measure the cost of the "easiest closed path"; Is a very small positive number, used to avoid mathematical calculation instability caused by the closed cost being zero; Is the joint classification function. After inputting three structural determination variables in the joint classification function, the structural category of the region is judged according to the threshold combination logic and the structure label is output.

[0024] For the image region structure in the local direction convergence mutation region, by constructing a non-linear path mutation metric value that fuses the tensor direction twist rate, local tension gradient variation, and perturbation diffusion response , construct a fracture induction function Determine the potential boundary fracture points in the image, and accordingly guide the subsequent erosion and dilation processes to form a structural contour path; ; ; Among them, the fracture induction function Indicates whether the pixel point in the image is marked as a structure fracture induction point, and the output is a Boolean value of, 0 indicates that the pixel point is not judged as a structural fracture point, and 1 indicates that it is judged as a fracture point; respectively represent the main direction angles extracted from the current pixel point and its neighboring pixel points in the tensor atlas, used to capture the sharp rotation in the direction field; Represents the directed angular difference between the main directions (that is, the angular twist rate), and non-linearly describes the rotation direction and amplitude of the main axis change; Is the tension function, and the tension function represents the contraction or expansion trend of the image structure boundary, generated by the density change in the structural path; Are the gradients of the structural tension in the horizontal and vertical directions respectively, and the two are used to reflect the local structural tension perturbation; is the direction consistency function, which is used to characterize the continuity of the principal axis direction of the tensor in the image boundary structure; is the mixed second-order partial derivative of the direction consistency, and the mixed second-order partial derivative of the direction consistency represents the asymmetric perturbation of the consistency jump in the two-way structure change; is the direction mutation intensity factor, which is used to extract the gradient jump amplitude information from the boundary response set and measure the drastic turning behavior of the edge; is the local response rate of the direction diffusion tensor, and the local response rate of the direction diffusion tensor is used to characterize the expansion or contraction effect of the boundary information on the principal axis of the direction tensor; in addition, in the formula represents the non-linear expansion term for constructing the propagation intensity of the structural perturbation, which is used to enhance the selective discrimination ability of the boundary mutation points; is the joint determination threshold of the structural mutation, and the joint determination threshold of the structural mutation is used to comprehensively evaluate the response lower limit of the fracture sensitivity; Input all continuous closed high-consistency regions into the structural potential construction process to generate the structural potential trajectory function and the reference guiding path function , and combine the closed perturbation function to construct the cleavage scoring function ; if the score is higher than the preset threshold , then this point is marked as a true fracture path point, and the cleavage path set is output for topological division and pore reconstruction; ; ; In in the formula represents the sub-pixel coordinates of the currently processed pixel in the image; the structural potential trajectory function represents the potential-driven path within the closed region; , , and represent along the first-order derivative; , is the second-order partial derivative of , , which is used to calculate the local structural curvature change; the reference guiding path function is used as the comparison reference path for the potential path ; , is the derivative of The sub-pixel perturbation tension residual function of path points in a closed region, used to model the torsional tension of direction; the cleavage scoring function The scoring function that fuses three factors, used as the basis for judging whether a fracture path is formed; the preset threshold It is the threshold for path cleavage scoring determination; It is the binary determination function for the fracture path, If the output of is 1, it means that it is recognized as a pore cleavage point, otherwise it is 0; in addition, in the formula It is expressed as the structural curvature mutation term in the cleavage scoring function; It is expressed as the path guiding offset term in the cleavage scoring function; It is expressed as the closed perturbation tension term in the cleavage scoring function; among them, the sub-pixel coordinates do not refer to the original pixel points in the image, but are the continuous positions estimated inside the pixel grid through interpolation or fitting, etc., used for more accurate positioning of image structure features such as edges, corners or paths, so as to realize the fine-grained extraction and determination of key structures in image analysis; In addition, in this solution 、 is a further solution of the fracture path determination logic constructed based on the two structural determination variables of "path tension value" and "structural boundary overlap degree". The constructed unified cleavage scoring function integrates the above two structural determination variables and the structural potential response variable, and transforms them into a single scoring mechanism for path recognition, so as to realize the structural optimization expression of variable integration and continuous determination process on the basis of keeping the path validity judgment logic in the original solution unchanged.

[0025] Construct candidate boundaries for the initial pore size structure map one and the initial pore size structure map two extracted from the corrosion expansion path and the vein backtracking path respectively, and construct a boundary matching scoring matrix based on the three scoring quantities of the boundary coincidence degree R 1 、the direction alignment degree R 2 and the structural center offset amount R 3 to judge the boundary fusion condition; Through to define the scoring of the boundary coincidence degree R 1 : ; Through to define the scoring of the direction alignment degree R 2 : ; Through to define the scoring of the structural center offset amount R 3 : ; In , , in the formula, represents the th candidate boundary pair; is the candidate boundary pair in the overlapping part length of the initial aperture structure map one and the initial aperture structure map two; is the candidate boundary pair in the total length of the union of the initial aperture structure map one and the initial aperture structure map two; is the candidate boundary pair the difference in the image edge gradient change in the overlapping area; is the candidate boundary pair corresponding to the topological fracture complexity index; , is the candidate boundary pair the main direction angle in the erosion-dilation path map and the vein backtracking path map; is the candidate boundary pair average curvature of; , is the candidate boundary pair the horizontal and vertical offset of the geometric center of; is the candidate boundary pair structural symmetry coefficient of.

[0026] For the image block marked as the boundary feature determination fuzzy area, establish a path confidence index and a scoring consistency difference index indexed by the boundary structure, and build a dynamic callback mechanism based on the two to update the image path and boundary judgment; define the path confidence as ; define the scoring consistency difference as ; ; ; where is the boundary layer tension response extracted at the pixel point ; is the difference value of the image processing strategy at the pixel point under two paths; is the score of the pixel point under the erosion-dilation path; is the score of the pixel point under the vein backtracking path; where the score under the path includes three dimensions: boundary coincidence degree, direction alignment degree, and structural center offset amount, corresponding to R 1 , R 2 , R 3, these three indicators together constitute the matching evaluation basis for each candidate boundary pair. Therefore, there are three scores for each path; the difference in score consistency is used to dynamically judge path conflicts; the path confidence is used to judge whether the path update is reasonable.

[0027] Generally speaking, this solution aims to solve the problem of pore adhesion caused by blurred boundaries, tightly fitting pore walls, or local collapse in the image processing of foamed concrete surfaces; traditional pixel-level pore size recognition methods are prone to misidentifying multiple actually independent pores as a whole when dealing with high-density continuous gray areas, resulting in distorted judgments of the material's porosity and structural looseness; for this reason, the present invention proposes a fine pore size recognition method with structural recognition ability and adaptive correction ability by establishing a multi-stage image processing process centered on sub-pixel edge structures. The entire solution starts with the "structural determination of boundary response characteristics" and constructs a region classification mechanism driven by structural determination variables such as boundary direction consistency, intensity jump variation, and edge closure ratio. Three types of structural regions are identified from the image boundary level: locally directionally convergent mutation regions, continuously closed high-consistency regions, and boundary feature determination fuzzy regions; then, for the first two types of regions, boundary contour extraction paths driven by structure tensors and potential energy are designed respectively, and two sets of initial pore size maps are formed. On this basis, to cope with the differences between the two maps, a fusion judgment mechanism is further constructed. A boundary matching score matrix is constructed through three scoring indicators: boundary coincidence degree, direction alignment degree, and structural center offset amount, to achieve disambiguation and integration of redundant or fuzzy boundaries in the maps; finally, for the remaining determination fuzzy regions, a structure state table and a path score callback mechanism are constructed, and the paths and boundaries are corrected again according to the results of the determination variables to achieve the closed-loop optimization of structure recognition. The core of the solution lies in the coordinated use of multi-path joint judgment and structural closed-loop correction mechanisms; first, the image processing extracts three structural determination variables through boundary response characteristics. These three variables analyze the essence of the image structure from three dimensions: direction consistency, intensity mutation, and geometric closure, and establish a preliminary structural classification of the image region. In the structurally clear region, for the locally directionally convergent mutation region, the main axis change trend is fitted through the direction tensor field, the local direction torsion rate and the abnormal points of the structural tension gradient are extracted, and erosion and limited expansion operations in the main axis direction are performed. Finally, a contour set guided by structural fracture is obtained. For the continuously closed high-consistency region, by constructing a structural potential energy field and guiding path search, path tension evaluation and boundary overlap degree judgment are performed on each potential path, so as to screen out the real fracture paths, and then the pore structure is divided through topological reconstruction. Subsequently, the above two types of graphs are input into the fusion determination module to construct a boundary matching scoring matrix, and based on three scoring indicators - coincidence degree, alignment degree, and offset, operations of fusion, retention, or re - determination are performed on the boundaries; For all regions with ambiguous determination, the scoring path and execution status corresponding to the boundary structure are recorded through the established structure status table. When there is a scoring consistency conflict or the path confidence is lower than the threshold, the system will automatically call the determination variable to re - identify the structure type of the boundary and specify a new image processing path, ultimately achieving path update, layer replacement, and closed - loop correction; The design of this solution not only breaks the dependence of traditional aperture recognition on pixel - level edges, but also proposes an image processing logic oriented to the "synergistic expression of structural continuity and fracture mutation". Its core parts include: Using structural determination variables to explicitly classify image regions to enhance the cognitive ability of image structures; Guiding the boundary extraction paths of the two types of regions through structure tensors and potential energy functions respectively to form a highly targeted heterogeneous graph construction mechanism; Introducing a matching scoring matrix to fuse the two types of graphs to enhance the collaborative and structural consistency judgment ability between paths; Constructing a structure status table and a path scoring callback mechanism to form an image processing flow with closed - loop feedback ability, which can dynamically adapt to the uncertainty of fuzzy region recognition.

[0028] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A foam concrete image processing method based on sub-pixel edge reconstruction, characterized in that: include: Acquire an image of foamed concrete and extract a boundary response feature set including edge direction information, edge gradient change information and local intensity contrast information; Perform edge structure classification on the image region based on the boundary response feature set combined with threshold judgment; For the image area determined as the local direction convergence mutation area, the crossover mutation point set is identified by fitting the direction tensor convergence trend, and the direction-guided erosion and dilation image processing is performed; For image regions that are judged to be continuous, closed, and highly consistent regions, a structural potential field is constructed and a joint judgment of path tension and boundary coincidence is performed to identify the true fracture path and perform topological division accordingly. For the aperture structure maps output by the image processing of the corrosion expansion path and the vein tracing path, a cross-matching judgment of the three consistency scoring items in the boundary matching scoring matrix is ​​performed, and the structural difference areas are selectively integrated or retained according to the fusion conditions; By fusing the path and the scoring results, a structure determination mechanism is constructed, and path updating and callback correction are performed on the boundaries of the blurred areas in the image to achieve the recognition of the aperture structure and the optimization of the image processing flow.

2. The foam concrete image processing method based on sub-pixel edge reconstruction according to claim 1 is characterized in that: The original image data of the foam concrete image is obtained, and the grayscale standardization and multi-scale edge enhancement operations are performed on the original image data to obtain a boundary response feature set including edge direction information, edge gradient change information and local intensity contrast information; Based on the boundary response feature set, three structural determination variables are extracted, including the boundary direction consistency measure value B1, the boundary intensity gradient variation value B2 and the regional edge closure ratio B3; If B1 is less than the boundary direction consistency measurement threshold T1 and B2 is greater than the boundary intensity gradient variation threshold T2, the image area is marked as a local direction convergence mutation area; if B1 is greater than or equal to T1 and B3 is greater than the regional edge closure ratio threshold T3, the image area is marked as a continuous closed high consistency area; if the three are in a critical intersection state, the image area is marked as a fuzzy area for boundary feature determination; The image blocks marked as fuzzy areas for boundary feature judgment are temporarily stored as fuzzy area masks and included in the fusion judgment process together with the image blocks marked as local direction convergence mutation areas and the image blocks marked as continuous closed high consistency areas, so as to perform unified judgment on the path ownership of the three types of areas.

3. The foam concrete image processing method based on sub-pixel edge reconstruction according to claim 2 is characterized in that: All local directional convergence mutation areas are input into the structural directional tensor extraction process, and the tensor map corresponding to the principal axis is constructed through the boundary response feature set. The convergence trend of the principal axis direction is fitted on the tensor map to generate the directional tensor convergence map and identify the aggregation boundary area. Based on image analysis, the set of directional gradient mutation points in the aggregated boundary area is identified as a fracture candidate subset, and tensor cross angle analysis is performed on each candidate point in the fracture candidate subset; if the cross angle is greater than the preset fracture criterion, the point is identified as a valid fracture point and used to generate a fracture induction layer; Based on the fracture-induced layer, a weighted erosion operation of the direction tensor is performed, and then a limited expansion operation is performed to maintain the direction of the principal axis to preserve the boundary continuity and spatial consistency formed by the fracture; a structural contour extraction operation is performed on the processed area, and all the extracted boundary contours are combined into a contour set, and the processing path label corresponding to each contour is recorded at the same time; The obtained contour set is output as the initial aperture structure map 1 under the corrosion and expansion path.

4. The foam concrete image processing method based on sub-pixel edge reconstruction according to claim 3 is characterized in that: All continuous closed high-consistency regions are input into the structural potential energy construction process, and a sub-pixel structural potential energy field is generated based on edge strength and closure stability. A guided path search is performed in the structural potential energy field to obtain a set of potential fracture paths, and the path tension value and structural boundary overlap are calculated for each path in turn. If the path tension value is greater than the path tension threshold T4 and the structural boundary overlap is less than the structural boundary overlap threshold T5, the path is confirmed as a true fracture path and serves as a boundary cracking trigger path; Using the confirmed real fracture paths, a topological structure reconstruction operation is performed on the original image, and multiple logical pore sub-blocks are divided; The pore structures of the divided pore sub-blocks are marked with path labels and output as the initial pore structure atlas 2 under the context tracing path.

5. The foam concrete image processing method based on sub-pixel edge reconstruction according to claim 4 is characterized in that: The sets of initial aperture structure atlas 1 and initial aperture structure atlas 2 of the corrosion expansion path and the vein backtracking path are merged, and a candidate boundary set is constructed for the boundary pairs in the two atlases; for each pair of candidate boundaries in the candidate boundary set, the boundary coincidence R1, the direction alignment degree R2 and the structure center offset R3 are calculated, and a boundary matching score matrix is ​​formed; In the boundary matching score matrix, if R1 in a boundary pair is greater than the boundary coincidence threshold T6, and R2 and R3 are less than the direction alignment threshold T7 and the structure center offset threshold T8, then the boundary pair is determined to be a fused boundary, and the layer is merged and assigned a unified number; If the boundary matching score matrix does not meet the fusion boundary judgment criteria, the image area is retained as a fuzzy area for boundary feature judgment; sub-pixel curve fitting is performed on the boundaries of all determined image areas, and a vector boundary set and corresponding path label information are output.

6. The foam concrete image processing method based on sub-pixel edge reconstruction according to claim 5 is characterized in that: Establish a path label with each boundary structure in the contour set as the main index, a scoring result of the boundary matching scoring matrix, and a structural state table. The structural state table is formed by introducing the path execution state and the regional fusion judgment result on the basis of the scoring result of the boundary matching scoring matrix. The original strategy path and the scoring matching value corresponding to each boundary structure are recorded in the structural state table; For all image regions marked as fuzzy regions for boundary feature determination, evaluate the difference between their path determination confidence and score consistency to determine whether there is aperture identification uncertainty; If the path confidence is less than the re-judgment threshold T9 and the score consistency difference is greater than the conflict tolerance threshold T 10 , then the boundary structure determination is performed again based on the three structure determination variables to complete the missed or misclassified aperture boundaries; According to the new round of structure determination results, the image structure path is re-specified, and the corresponding image processing method is executed to obtain the updated aperture boundary extraction result; the image structure path includes an erosion and dilation path or a vein tracing path; the updated image structure boundary replaces the corresponding area in the original boundary layer, and its corresponding path identification information is synchronously updated.

7. The foam concrete image processing method based on sub-pixel edge reconstruction according to claim 6 is characterized in that: definition Represents the pixel points in the image The joint classification result of the three corresponding structure judgment variables; definition Represents pixel The boundary direction consistency measure B1 of Represents pixel The boundary intensity gradient variation value B2 is defined as Indicates the midpoint of the image The edge closure ratio of the region to which it belongs is B3, Used to measure whether the contour is closed and complete; ； ； ； ； ； in Represents pixel The neighborhood set of ; Represents the pixel in the neighborhood The main direction angle of the edge; Indicates the current pixel The main direction of the edge; For point Directional divergence suppression factor calculated in the local tensor field; is a set of multi-scale image pyramids; Indicated in Pixels in the scaled image The gradient intensity modulus of is the anisotropy adjustment factor; Indicates that the boundary tracing algorithm is used to trace from pixel points Starting from, extract the pixel set on all boundary paths; Point on the path The local connectivity score of For point A set of multiple closed boundary candidate paths formed by the area; is a closed cost function; is a very small positive number; is the joint classification function.

8. The foam concrete image processing method based on sub-pixel edge reconstruction according to claim 7 is characterized in that: Aiming at the image region structure in the local direction convergence mutation region, a nonlinear path mutation metric is constructed by integrating the tensor direction torsion rate, local tension gradient variation and disturbance diffusion response. , construct the fracture induction function Determine potential boundary breakpoints in the image and use them to guide subsequent erosion and dilation processes to form a structural contour path; ； ； The fracture induction function Represents the pixel points in the image Whether it is marked as a structural fracture induction point, the output is Boolean value of Respectively represent the main direction angles extracted from the tensor map between the current pixel and its neighboring pixels; Represents the directed angle difference between the main directions; is the tension function; are the gradients of structural tension in the horizontal and vertical directions, respectively; is the direction consistency function; is the mixed second-order partial derivative with direction consistency; is the directional mutation intensity factor; is the local response rate of the directional diffusion tensor; in addition, The nonlinear expansion term representing the propagation intensity of disturbances in the constructed structure; is the threshold for joint determination of structural mutations; Input all continuous closed high consistency areas into the structural potential energy construction process to generate the structural potential energy trajectory function With reference to the guided path function , and combined with the closed perturbation function Constructing the cracking scoring function ; If the score is higher than the preset threshold , then the point is marked as a true fracture path point, and the output cracking path set is used for topological partitioning and pore reconstruction; ； ； exist In the formula Represents the sub-pixel coordinates of the currently processed pixel in the image; structural potential energy trajectory function represents the potential energy driven path within the enclosed area; , , and express along The first derivative of ; , for The second-order partial derivative of , Used to calculate the local structure curvature change; refer to the guided path function Used as a potential energy path The comparison reference path; , for Derivatives in both directions; closed perturbation functions Sub-pixel perturbation tension residual function representing path points in the closed region; cracking score function The scoring function integrating the three factors is used as the basis for judging whether a fracture path is formed; the preset threshold is the threshold for determining the pathway cleavage score; is the binary decision function of the fracture path, The output of is 1 if it is identified as a pore rupture point, otherwise it is 0.

9. The foam concrete image processing method based on sub-pixel edge reconstruction according to claim 8 is characterized in that: The candidate boundaries are constructed for the initial aperture structure map 1 and the initial aperture structure map 2 extracted from the corrosion expansion path and the vein tracing path respectively, and the boundary matching score matrix is ​​constructed based on three types of score quantities: boundary coincidence R1, direction alignment R2 and structure center offset R3, to determine the boundary fusion conditions. pass To define the score of boundary overlap R1: ； pass To define the score of directional alignment R2: ； pass To define the score of the structure center offset R3: ； exist , , In the formula Indicates for candidate boundary pairs; Candidate boundary pairs The length of the overlapped portion between the initial aperture structure spectrum 1 and the initial aperture structure spectrum 2; Candidate boundary pairs The total length of the union of the initial aperture structure map 1 and the initial aperture structure map 2; Candidate boundary pairs Differences in image edge gradient changes in overlapping areas; Candidate boundary pairs Corresponding topological fracture complexity index; , Candidate boundary pairs The main direction angle in the corrosion expansion path map and the vein tracing path map; Candidate boundary pairs The average curvature of , Candidate boundary pairs The geometric center is offset horizontally and vertically; Candidate boundary pairs The structural symmetry coefficient.

10. The foam concrete image processing method based on sub-pixel edge reconstruction according to claim 9, characterized in that: For the image blocks marked as fuzzy areas for boundary feature judgment, a path confidence index and a score consistency difference index indexed by the boundary structure are established, and a dynamic callback mechanism is constructed based on the two to update the image path and boundary judgment; the path confidence is defined as ; Define the score consistency difference as ; ； ； in Pixel The tension response of the boundary layer extracted at ; Pixel The difference value of image processing strategies under the two paths; Pixel Scoring under the corrosion expansion path; Pixel Scoring under the contextual tracing path.

Images