A method and system for recovering multiple peaks of micro-holes based on DOG focusing response

Abstract

This invention discloses a method and system for multi-peak depth recovery of micropores based on DOG focus response. The method involves acquiring a multi-focal plane image sequence of the micropore under different focal plane positions, performing differential Gaussian processing on each frame to obtain the DOG response, and generating a DOG focus evaluation value. An axial focus response curve is established for each pixel position, and multiple candidate focus peaks are extracted. Sub-step peak position correction is performed on the candidate focus peaks, and candidate depth values ​​are calculated. A comprehensive confidence score is calculated based on peak intensity, peak prominence, peak sharpness, and neighborhood support. Valid candidate peaks are selected based on confidence level and pore wall structure constraints. A micropore depth map or 3D point cloud is generated based on the valid candidate peaks. This invention can retain the weak response peaks of the true pore wall, suppress noise peaks and spurious peak interference, and is suitable for micropore wall depth recovery.

Description

Technical Field

[0001] This invention belongs to the field of micropore three-dimensional measurement technology, specifically relating to a method and system for recovering multi-peak depth of micropores based on the difference of Gaussian (DOG) focusing response. Background Technology

[0002] Micropore structures are widely found in components such as hot-end parts of aero-engines and injection disks of liquid rocket engines. Their aperture, axial orientation, and wall morphology affect cooling efficiency, injection uniformity, flow distribution, and service reliability. Therefore, high-precision three-dimensional measurement of micropore structures is an important step in evaluating the quality of micropore fabrication and analyzing its service performance.

[0003] Existing micro-hole measurement methods mainly include contact measurement, industrial CT measurement, and optical vision measurement. Contact measurement is easily limited by probe size, measurement rigidity, and accessibility within the hole, making it difficult to reliably acquire complete 3D topography of small-diameter, high-angle hole wall regions. While industrial CT can acquire 3D information within the hole, it is costly, has long scanning and reconstruction times, and is susceptible to metal artifacts, voxel resolution, and measurement uncertainty. Ordinary optical vision measurement is relatively efficient, but it is mainly suitable for detecting hole opening contours and external surface features, and has limited ability to recover 3D information of hole wall depth and high-angle hole wall regions.

[0004] Zoom microscopy involves acquiring a sequence of multi-focal images along the optical axis and determining the optimal focusing layer based on the variation of image sharpness with scanning position, thereby recovering the depth information of the measured surface. This method is suitable for non-contact 3D measurement of the aperture and wall regions of micropores. However, micropore structures often exhibit abrupt changes in aperture boundaries, tilted walls, local occlusion, weak texture, and reflections, which can cause multiple local peaks to appear in the axial focusing response curve at the same pixel location or local area.

[0005] Existing zoom microscopy depth recovery methods typically employ a single-peak maximum strategy, selecting the position of the maximum response in the focus evaluation curve as the sole depth value. When the focus response at the aperture edge or in a locally strong reflection region is higher than the weak response at the aperture wall, this strategy prioritizes the strong response peak while ignoring the weaker but real aperture wall focus peak, resulting in the loss of depth information in the lower part of the aperture wall. The resulting depth map or point cloud is prone to problems such as aperture wall discontinuities, local missing values, height jumps, and spurious peak interference, which in turn affects the extraction stability of geometric parameters such as aperture diameter, aperture axis orientation, and aperture wall morphology. Therefore, a multi-peak depth recovery method for micropores is needed that can identify and retain the real weak aperture wall response peaks while suppressing noise peaks and spurious peak interference. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a method and system for micro-aperture multi-peak depth recovery based on DOG focusing response, which addresses the shortcomings of the prior art. This method solves the technical problems that are easily caused by the loss of weak response of the real aperture wall, noise pseudo-peak interference, discontinuity of aperture wall depth information and local missing point cloud when using the single peak maximum value strategy for depth recovery in existing micro-aperture zoom microscopy.

[0007] The present invention adopts the following technical solution: A method for recovering the multi-peak depth of micropores based on DOG focusing response includes the following steps: S1. Obtain a multi-focal plane image sequence of the micropore under different focal plane positions; perform DOG processing on each frame of the multi-focal plane image sequence to obtain the DOG response, and obtain the DOG focus evaluation value based on the DOG response. S2. For pixel positions, extract DOG focus evaluation values ​​at different focal plane positions along the axial scanning direction, establish pixel-level focus response curves, and extract multiple candidate focus peaks from the pixel-level focus response curves based on local maximum criteria. S3. Based on the DOG focusing evaluation value of each candidate focusing peak and its adjacent scanning layers, perform sub-step peak position correction on the candidate focusing peak to obtain the corresponding candidate depth value, and calculate the comprehensive confidence score based on the peak intensity, peak prominence, peak sharpness and neighborhood support of the candidate focusing peak. S4. The candidate focused peaks are classified according to the comprehensive confidence score and neighborhood support, and the candidate focused peaks after confidence classification are screened according to the hole wall structure constraints to obtain effective candidate peaks; wherein, the hole wall structure constraints include hole wall radius zone constraints, hole axis compatibility constraints and local continuity constraints. S5. Generate a micropore depth map or a three-dimensional point cloud based on the pixel coordinates, object pixel size, axial scanning step distance and candidate depth value corresponding to the effective candidate peak.

[0008] Further, in S1, each frame of the multi-focal plane image sequence is filtered using a first Gaussian kernel and a second Gaussian kernel, respectively. The first Gaussian kernel and the second Gaussian kernel have different standard deviations, and the difference between the two filtering results is used to obtain the DOG response. The amplitude or square value of the DOG response is used as the DOG focus evaluation value.

[0009] Furthermore, in S2, before extracting multiple candidate focus peaks, the pixel-level focus response curve is smoothed, and the scan layer position that satisfies the local maximum condition in the smoothed pixel-level focus response curve is determined as the candidate focus peak.

[0010] Furthermore, in S2, when extracting the candidate focused peaks, the scanning layer positions that satisfy the local maximum condition are also screened based on the peak intensity, peak prominence, and peak spacing.

[0011] Further, in S3, the substep peak position correction includes: performing quadratic curve interpolation based on the DOG focus evaluation values ​​of the scanning layer where the candidate focus peak is located and its adjacent scanning layers before and after it to obtain the substep offset of the candidate focus peak relative to the integer scanning layer, and calculating the candidate depth value based on the scanning start position, axial scanning step distance, integer scanning layer position and the substep offset.

[0012] Furthermore, in S3, the neighborhood support is determined based on the number of supporting peaks in the spatial neighborhood of the candidate focus peak, wherein the supporting peak is a neighboring candidate focus peak whose scan layer position difference from the current candidate focus peak does not exceed a preset tolerance.

[0013] Further, in S4, classifying the candidate focus peaks according to the comprehensive confidence score and neighborhood support includes: classifying candidate focus peaks that meet the high confidence score threshold and high confidence neighborhood support conditions as high confidence peaks; classifying candidate focus peaks that are not classified as high confidence peaks but meet the medium confidence score threshold and medium confidence neighborhood support conditions as medium confidence peaks; and classifying candidate focus peaks that are not classified as high confidence peaks or medium confidence peaks as low confidence peaks.

[0014] Further, in S4, the hole wall structure constraint screening includes: determining whether the candidate focused peak satisfies the hole wall radius constraint based on the radial distance from the pixel position to the hole axis projection center; determining whether the candidate focused peak satisfies the hole axis compatibility constraint based on the deviation between the radial distance and the target hole wall radius; and determining whether the candidate focused peak satisfies the local continuity constraint based on whether there is an effective neighborhood of similar depth supporting the candidate focused peak in the spatial neighborhood.

[0015] Furthermore, in S5, the three-dimensional spatial coordinates are calculated based on the pixel coordinates corresponding to the effective candidate peaks, the image reference center, the object pixel size, and the candidate depth value; when there are multiple effective candidate peaks at the same pixel location, multiple three-dimensional depth points are generated based on the candidate depth values ​​of the multiple effective candidate peaks respectively.

[0016] Secondly, embodiments of the present invention provide a micropore multi-peak depth recovery system based on the Difference of Gaussian (DOG) focusing response, comprising: The image acquisition and response calculation module is used to acquire a multi-focal plane image sequence of the micro-hole under different focal plane positions, perform differential Gaussian processing on each frame of the multi-focal plane image sequence to obtain the DOG response, and obtain the DOG focus evaluation value based on the DOG response. The candidate peak extraction module is used to extract DOG focus evaluation values ​​at different focal plane positions along the axial scanning direction for pixel positions, establish pixel-level focus response curves, and extract multiple candidate focus peaks from the pixel-level focus response curves according to the local maximum criterion. The correction and evaluation module is used to perform sub-step peak position correction on each candidate focus peak and its adjacent scan layer based on the DOG focus evaluation value, to obtain the corresponding candidate depth value, and to calculate the comprehensive confidence score based on the peak intensity, peak prominence, peak sharpness and neighborhood support of the candidate focus peak. The constraint filtering module is used to classify the candidate focused peaks according to the comprehensive confidence score and neighborhood support, and to filter the candidate focused peaks after confidence classification according to the hole wall structure constraints to obtain effective candidate peaks; wherein, the hole wall structure constraints include hole wall radius zone constraints, hole axis compatibility constraints and local continuity constraints; The depth recovery module is used to generate a micropore depth map or a three-dimensional point cloud based on the pixel coordinates, object pixel size, axial scan step size and candidate depth value corresponding to the effective candidate peak.

[0017] Thirdly, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the above-described micropore multi-peak depth recovery method based on DOG focusing response.

[0018] Fourthly, embodiments of the present invention provide a computer-readable storage medium including a computer program, which, when executed by a processor, implements the steps of the above-described micropore multi-peak depth recovery method based on DOG focusing response.

[0019] Fifthly, a chip includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the above-described micropore multi-peak depth recovery method based on DOG focusing response.

[0020] In a sixth aspect, embodiments of the present invention provide an electronic device, including a computer program, which, when executed by the electronic device, implements the steps of the above-described micropore multi-peak depth recovery method based on DOG focusing response.

[0021] Compared with the prior art, the present invention has at least the following beneficial effects: A multi-peak depth recovery method for micropores based on DOG focusing response is proposed. By extracting multiple candidate focusing peaks from the DOG focusing response curve, it can retain multi-height response information generated by the micropore opening, tilted pore wall, and observable area inside the pore, reducing the loss of true pore wall depth information caused by the traditional single-peak maximum value strategy. By constructing a candidate peak confidence evaluation model based on peak intensity, peak prominence, peak sharpness, and neighborhood support, it can improve the ability to distinguish between true pore wall peaks and noise peaks and spurious peaks. By introducing structural constraints such as pore wall radius band, pore axis compatibility, and local continuity, it can suppress isolated spurious peaks while retaining true weak pore wall peaks, thereby improving the integrity, continuity, and point cloud stability of micropore wall depth recovery.

[0022] Furthermore, by using Gaussian kernel filtering with two different standard deviations and subtracting them, the local structural responses related to the focusing state, such as the micropore opening edge, pore wall texture, and gray-scale abrupt changes, can be highlighted. Using the DOG response amplitude or square value as the focusing evaluation value can avoid the cancellation of positive and negative responses, making the subsequent focusing curve more suitable for peak detection.

[0023] Furthermore, smoothing the pixel-level focus response curve before candidate peak extraction can reduce the impact of image noise and local high-frequency fluctuations on local maximum determination; by extracting candidate peaks through local maximum conditions, multiple possible depth positions can be retained for the same pixel location.

[0024] Furthermore, by introducing peak intensity, peak prominence, and peak spacing as filters based on local maxima, isolated peaks formed by weak noise fluctuations can be initially eliminated, avoiding the interference of false peaks in subsequent confidence evaluation and point cloud generation due to an excessive set of candidate peaks.

[0025] Furthermore, by performing quadratic curve interpolation using the DOG focus evaluation values ​​of the candidate peak and its adjacent scanning layers, the position of the candidate peak can be refined from integer scanning layers to between adjacent focal planes. The advantage of this is that it reduces axial quantization error and improves the positioning accuracy of candidate depth values ​​without increasing the number of acquisition frames.

[0026] Furthermore, neighborhood support utilizes support peaks with similar scanning layer positions within the spatial neighborhood to determine the spatial continuity of candidate peaks. True aperture wall responses typically exhibit a continuous distribution within the neighborhood, while noise peaks and reflective spurious peaks often display isolated responses. Therefore, this feature helps distinguish between true aperture wall peaks and isolated spurious peaks.

[0027] Furthermore, by classifying candidate peaks into high, medium, and low confidence levels based on a comprehensive confidence score and neighborhood support, the peaks can be processed in a stratified manner: high-confidence peaks are directly retained, low-confidence peaks are removed, and medium-confidence peaks are then subjected to structural constraint judgment. This setting avoids the mistaken deletion of genuine weak pore wall peaks due to overly strict single-threshold screening.

[0028] Furthermore, pore wall radius constraints, pore axis compatibility constraints, and local continuity constraints screen candidate peaks from three aspects: spatial location, geometric distribution, and local depth continuity, respectively. Medium-confidence peaks are only retained if they conform to the geometric laws of the micropore wall, thus balancing weak response retention and spurious peak suppression.

[0029] Furthermore, three-dimensional spatial coordinates are calculated based on pixel coordinates, image reference center, object pixel size, and candidate depth values, so that effective candidate peaks can be mapped to depth maps or point clouds; when multiple effective candidate peaks exist at the same pixel location, multiple three-dimensional depth points are generated, which can express the multi-height structure of the micropore opening, pore wall, and pore interior region.

[0030] It is understood that the beneficial effects of the second to sixth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0031] In summary, this invention constructs axial response curves based on DOG focusing evaluation, extracts and retains multiple candidate focusing peaks, improves depth positioning accuracy through sub-step correction, and combines confidence grading and hole wall structure constraints to screen true hole wall peaks. This can reduce the loss of weak hole wall response, missing point cloud, and height jumps caused by single-peak strategies, and improve the integrity, continuity, and stability of micropore 3D depth recovery.

[0032] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0033] Figure 1 This is a flowchart of the micropore multi-peak depth recovery method based on DOG focusing response of the present invention; Figure 2 This is a schematic diagram of the multi-height focusing response of a single pixel in micro-aperture zoom microscopy. Figure 3 Flowchart for candidate peak confidence evaluation and classification; Figure 4 This is a schematic diagram of the distribution of candidate peaks and the rejection area before and after the pore wall structure constraint screening in this invention; Figure 5 This is a comparison image of the micropore point clouds obtained by the method of the present invention and the traditional single-peak depth recovery method; Figure 6 A schematic diagram of a computer device provided in an embodiment of the present invention; Figure 7 This is a block diagram of a chip provided in one embodiment of the present invention.

[0034] Among them, 60. Computer equipment; 61. Processor; 62. Memory; 63. Computer program; 600. Electronic device; 610. Processing unit; 620. Storage unit; 6201. Random access memory unit; 6202. Cache memory unit; 6203. Read-only memory unit; 6204. Program / utility; 6205. Program module; 630. Bus; 640. Display unit; 650. Input / output interface; 660. Network adapter; 700. External device. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0037] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0038] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this invention generally indicates that the preceding and following objects have an "or" relationship.

[0039] It should be understood that although terms such as first, second, third, etc., may be used in the embodiments of the present invention to describe the preset range, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from one another. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0040] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0041] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0042] This invention provides a micropore multi-peak depth recovery method based on DOG focusing response. Instead of using the global maximum value in the focusing evaluation curve as the sole criterion for depth determination, it extracts and retains multiple candidate focusing peaks based on the DOG focusing response. Furthermore, it combines the peak intensity, peak prominence, peak sharpness, and neighborhood support of the candidate peaks for confidence evaluation, and introduces pore wall structure constraints to screen the candidate peaks. This method not only suppresses noise peaks and spurious peak interference, but also retains the depth information corresponding to the weak response of the real pore wall, improving the integrity, continuity, and stability of the depth recovery of the micropore wall region.

[0043] Please see Figure 1 The present invention provides a method for recovering the multi-peak depth of micropores based on DOG focusing response, comprising the following steps: S1. Obtain the microaperture multifocal image sequence and calculate the DOG focusing response. A multi-focal-plane image sequence of the micropore under test at different focal plane positions is acquired. This multi-focal-plane image sequence is obtained by scanning along the optical axis using a zoom microscopy system. Let the acquired image sequence be:

[0044] in, Indicates the first Pixel position in frame image grayscale value at that location This represents the total number of frames in the image sequence.

[0045] For each frame of the image, two Gaussian kernels with different standard deviations (i.e., different spatial smoothing scales) are applied for filtering, and the difference between the two sets of filtering results is used to obtain the DOG response:

[0046] in, G σ1 and G σ2 The standard deviations are respectively represented as σ 1 and σ A Gaussian kernel function of 2, preferably satisfying the following conditions: σ 2> σ 1. Symbols This represents the convolution operation.

[0047] The spatial smoothing scale is determined by the Gaussian kernel standard deviation: smaller scale σ 1. Used to preserve the detailed structural response of micropore opening edges, pore wall textures, etc., at a larger scale. σ 2. Used to obtain a wider range of background smoothness response; the difference between the two can approximately highlight local high-frequency structures related to the focus state and weaken slowly changing illumination background. Preferably, σ 2 / σ 1. Take 1.4 to 2.0, or select according to the magnification of the microscope system, the size of the object pixel and the width of the micro-aperture edge.

[0048] To avoid the cancellation of positive and negative responses and to enhance the local structural response, the amplitude or square of the DOG response is used as the focusing evaluation value. The focusing evaluation function is constructed using the square value:

[0049] in, Indicates pixel position In the DOG focus rating in a frame image.

[0050] S2. Establish pixel-level focus response curves and extract multiple candidate peaks. For each pixel location or local region, extract the DOG focus evaluation value at different focal plane positions along the axial scanning direction to establish a pixel-level focus response curve:

[0051] in, Indicates pixel position The corresponding axial focusing response curve.

[0052] To reduce the impact of image noise and local high-frequency fluctuations on peak detection, the focusing response curve is smoothed. A three-point mean smoothing method is used:

[0053] in, This represents the smoothed focus response value.

[0054] Subsequently, the local maximum criterion was used to extract multiple candidate focused peaks. If the first The layer satisfies:

[0055] Then The corresponding positions are used as candidate focus peaks. Thus, multiple candidate peaks can be retained at the same pixel location to express the multi-height response information that may be generated by the micro-aperture opening, the inclined aperture wall, and the observable area inside the aperture.

[0056] S3. Perform substep size correction and confidence evaluation on candidate peaks. For the extracted candidate focused peaks, sub-step offset is performed by combining the focused response values ​​of their adjacent scan layers. The sub-step offset is not a pre-fixed empirical step size, but rather a value based on the scan step distance. The value is calculated using quadratic curve interpolation based on the focused response values ​​of the candidate peak and its adjacent layers, with the value as the unit. If the integer frame number corresponding to the candidate peak is... Its substep offset is ,but Represented as:

[0057] Pixel coordinates are omitted for simplification. , This represents the smoothed DOG focus evaluation value. To normalize the inter-frame offset, it is typically limited to 0.5≤ ≤0.5; If the absolute value of the denominator is less than the preset threshold or the calculation result exceeds this range, the peak shape is considered unstable, and an integer layer depth can be used for the candidate peak or... i The amplitude is limited to the above range. Therefore, the axial depth value corresponding to the candidate peak is:

[0058] in, z i This represents the candidate depth value corresponding to the i-th candidate peak. z 0 is the scan start position. Δz This refers to the axial scan step size; Indicates the integer scan layer depth. This indicates the substep correction amount.

[0059] Furthermore, for each candidate peak, its peak intensity, peak prominence, peak sharpness, and neighborhood support are calculated, and a comprehensive confidence score is constructed. The comprehensive confidence score of the candidate peak is expressed as:

[0060] in:

[0061] in, Indicates the first The overall confidence score of each candidate peak; , , and These represent the normalized results for peak intensity, peak prominence, peak sharpness, and neighborhood support, respectively. , , , These are the weight coefficients for the corresponding features.

[0062] S4. Screening effective peaks based on confidence level and pore wall structure constraints. Based on the overall confidence score and neighborhood support of the candidate peaks, they are classified into high-confidence peaks, medium-confidence peaks, and low-confidence peaks. The confidence level of a candidate peak is represented as follows:

[0063] in, and These are the score thresholds for high confidence and medium confidence, respectively. ; The neighborhood support number of the candidate peak; and These represent the minimum number of neighborhood support required for high confidence and medium confidence levels, respectively. .

[0064] For high-confidence peaks, they are directly retained as valid depth peaks; for low-confidence peaks, they are directly discarded; for medium-confidence peaks, further screening is performed using hole wall structure constraints. These hole wall structure constraints include hole wall radius zone constraints, hole-axis compatibility constraints, and local continuity constraints.

[0065] Let the projection center of the micro-pore axis on the image plane be... pixel position The radial distance to the center of the hole shaft projection is:

[0066] If the pixel location of the candidate peak satisfies:

[0067] Therefore, the candidate peak is considered to fall within the suspected pore wall radius zone. and These represent the minimum and maximum radial distances corresponding to the suspected borehole wall region, respectively.

[0068] Furthermore, let the radius of the target hole wall be... The permissible radial deviation is When the following conditions are met:

[0069] The candidate peak is considered to be compatible with the expected pore wall geometry.

[0070] For the local continuity constraint, it is determined whether a candidate peak has a supporting peak of similar depth and sufficient confidence level within its spatial neighborhood. A four-neighborhood direction criterion is used: a candidate peak is considered to satisfy the local continuity constraint if it has effective neighborhood support in at least two of the four directions (up, down, left, and right).

[0071] Based on the above constraints, the intermediate confidence peak is only retained as an effective hole wall depth peak when it simultaneously satisfies the hole wall radius zone constraint, the hole-axis compatibility constraint, and the local continuity constraint.

[0072] S5. Generate micropore depth maps or 3D point clouds based on valid candidate peaks. High-confidence peaks and medium-confidence peaks that satisfy the hole wall structure constraints are used as valid candidate peaks. Based on the pixel coordinates, object pixel size, axial scan step distance and substep correction amount of the candidate peaks, their corresponding spatial coordinates are calculated to generate microhole depth maps or 3D point clouds.

[0073] Let the column coordinates of the pixels be... The row coordinates are The image reference center is The object pixel sizes are respectively and Then the coordinates of a three-dimensional point are represented as:

[0074]

[0075]

[0076] The Z coordinates mentioned above z i Substep length correction item already included Δk i Δz Therefore, the candidate peak positions can be refined to the distance between adjacent focal planes, which is used to preserve the true depth corresponding to the weak response peaks of the aperture wall.

[0077] When multiple valid candidate peaks satisfying the conditions exist at the same pixel location, multiple depth points are retained to express the multi-height structural information of the micropore region:

[0078] in, This indicates the number of valid candidate peaks retained at that pixel location.

[0079] Example 1 In another embodiment of the present invention, a micropore multi-peak depth recovery system based on DOG focusing response is provided. This system can be used to implement the above-mentioned micropore multi-peak depth recovery method based on DOG focusing response. Specifically, the micropore multi-peak depth recovery system based on DOG focusing response includes an image acquisition and response calculation module, a candidate peak extraction module, a correction and evaluation module, a constraint screening module, and a depth recovery module.

[0080] The image acquisition and response calculation module is used to acquire a multi-focal plane image sequence of the micro-hole under different focal plane positions, perform differential Gaussian processing on each frame of the multi-focal plane image sequence to obtain the DOG response, and obtain the DOG focus evaluation value based on the DOG response. The candidate peak extraction module is used to extract DOG focus evaluation values ​​at different focal plane positions along the axial scanning direction for pixel positions, establish pixel-level focus response curves, and extract multiple candidate focus peaks from the pixel-level focus response curves according to the local maximum criterion. The correction and evaluation module is used to perform sub-step peak position correction on each candidate focus peak and its adjacent scan layer based on the DOG focus evaluation value, to obtain the corresponding candidate depth value, and to calculate the comprehensive confidence score based on the peak intensity, peak prominence, peak sharpness and neighborhood support of the candidate focus peak. The constraint filtering module is used to classify the candidate focused peaks according to the comprehensive confidence score and neighborhood support, and to filter the candidate focused peaks after confidence classification according to the hole wall structure constraints to obtain effective candidate peaks; wherein, the hole wall structure constraints include hole wall radius zone constraints, hole axis compatibility constraints and local continuity constraints; The depth recovery module is used to generate a micropore depth map or a three-dimensional point cloud based on the pixel coordinates, object pixel size, axial scan step size and candidate depth value corresponding to the effective candidate peak.

[0081] Example 2 This invention provides a terminal device comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve a corresponding method flow or corresponding function. The processor described in this embodiment can be used for the operation of a micro-orifice multi-peak depth recovery method based on DOG focusing response, including: A multi-focal plane image sequence of the micropore under test at different focal plane positions is acquired; each frame in the multi-focal plane image sequence is subjected to DOG processing to obtain the DOG response, and the DOG focus evaluation value is obtained based on the DOG response; for pixel positions, the DOG focus evaluation values ​​at different focal plane positions are extracted along the axial scanning direction to establish a pixel-level focus response curve, and multiple candidate focus peaks are extracted from the pixel-level focus response curve based on the local maximum criterion; based on each candidate focus peak and the DOG focus evaluation value of its adjacent scanning layers, the candidate focus peaks are corrected by sub-stepping peak position correction to obtain the DOG focus evaluation value of the micropore under test. The corresponding candidate depth values ​​are obtained, and a comprehensive confidence score is calculated based on the peak intensity, peak prominence, peak sharpness, and neighborhood support of the candidate focused peaks. The candidate focused peaks are then classified according to their confidence level based on the comprehensive confidence score and neighborhood support. The classified candidate focused peaks are then filtered according to the hole wall structure constraints to obtain valid candidate peaks. The hole wall structure constraints include hole wall radius constraints, hole axis compatibility constraints, and local continuity constraints. Based on the pixel coordinates, object-side pixel size, axial scan step distance, and candidate depth values ​​corresponding to the valid candidate peaks, a micro-hole depth map or a 3D point cloud is generated.

[0082] Please see Figure 6The terminal device is a computer device. In this embodiment, the computer device 60 includes a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When executed by the processor 61, the computer program 63 implements the micropore multi-peak depth recovery method based on DOG focusing response in this embodiment. To avoid repetition, details are omitted here. Alternatively, when executed by the processor 61, the computer program 63 implements the functions of each model / unit in the micropore multi-peak depth recovery system based on DOG focusing response in this embodiment. To avoid repetition, details are omitted here.

[0083] Computer device 60 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. Computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art will understand that... Figure 6 This is merely an example of computer device 60 and does not constitute a limitation on computer device 60. It may include more or fewer components than shown, or combine certain components, or different components. For example, computer device may also include input / output devices, network access devices, buses, etc.

[0084] The processor 61 may be a Central Processing Unit (CPU), or other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0085] The memory 62 can be an internal storage unit of the computer device 60, such as a hard disk or memory of the computer device 60. The memory 62 can also be an external storage device of the computer device 60, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on the computer device 60.

[0086] Furthermore, the memory 62 may include both internal storage units of the computer device 60 and external storage devices. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 can also be used to temporarily store data that has been output or will be output.

[0087] Please see Figure 7 The terminal device is an electronic device 600, which is manifested in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including storage unit 620 and processing unit 610), a display unit 640, etc.

[0088] The storage unit stores program code, which can be executed by the processing unit 610 to perform the steps described in the method section of this specification according to various exemplary embodiments of the present invention. For example, the processing unit 610 can perform actions such as... Figure 1 The steps are shown in the figure.

[0089] Storage unit 620 may include readable media in the form of volatile storage units, such as random access memory (RAM) 6201 and / or cache memory 6202, and may further include read-only memory (ROM) 6203.

[0090] Storage unit 620 may also include a program / utility 6204 having a set (at least one) program module 6205, such program module 6205 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0091] Bus 630 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the multiple bus structures.

[0092] Electronic device 600 can also communicate with one or more external devices 700 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 600, and / or with any device that enables electronic device 600 to communicate with one or more other computing devices (e.g., router, modem). This communication can be performed via input / output interface 650. Furthermore, electronic device 600 can also communicate with one or more networks (e.g., local area network, wide area network, and / or public network, such as the Internet) via network adapter 660. Network adapter 660 can communicate with other modules of electronic device 600 via bus 630. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms.

[0093] Example 3 This invention also provides a storage medium, specifically a computer-readable storage medium, which is a memory device in a terminal device for storing programs and data. It is understood that the computer-readable storage medium here can include both built-in storage media in the terminal device and extended storage media supported by the terminal device; it can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor, which can be one or more computer programs (including program code). More specific examples of the computer-readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical fiber, portable compact disk read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0094] Computer-readable storage media also include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium can also be any readable medium other than a readable storage medium that can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium can be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, radio frequency, etc., or any suitable combination thereof.

[0095] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0096] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the micro-orifice multi-peak depth recovery method based on DOG focusing response in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps: A multi-focal plane image sequence of the micropore under test at different focal plane positions is acquired; each frame in the multi-focal plane image sequence is subjected to DOG processing to obtain the DOG response, and the DOG focus evaluation value is obtained based on the DOG response; for pixel positions, the DOG focus evaluation values ​​at different focal plane positions are extracted along the axial scanning direction to establish a pixel-level focus response curve, and multiple candidate focus peaks are extracted from the pixel-level focus response curve based on the local maximum criterion; based on each candidate focus peak and the DOG focus evaluation value of its adjacent scanning layers, the candidate focus peaks are corrected by sub-stepping peak position correction to obtain the DOG focus evaluation value of the micropore under test. The corresponding candidate depth values ​​are obtained, and a comprehensive confidence score is calculated based on the peak intensity, peak prominence, peak sharpness, and neighborhood support of the candidate focused peaks. The candidate focused peaks are then classified according to their confidence level based on the comprehensive confidence score and neighborhood support. The classified candidate focused peaks are then filtered according to the hole wall structure constraints to obtain valid candidate peaks. The hole wall structure constraints include hole wall radius constraints, hole axis compatibility constraints, and local continuity constraints. Based on the pixel coordinates, object-side pixel size, axial scan step distance, and candidate depth values ​​corresponding to the valid candidate peaks, a micro-hole depth map or a 3D point cloud is generated.

[0097] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0098] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0099] Example 4 First, the micropore to be measured is placed in a zoom microscopy system, and the microscopic imaging module or the sample is controlled to perform axial scanning along the optical axis, acquiring multi-focal plane image sequences at different focal plane positions. Let the acquired image sequence be:

[0100] in, Indicates the first Pixel position in frame image grayscale value at that location This represents the total number of frames in the image sequence. If the scanning process uses a constant step method, then the [number of frames] is [number of frames]. The axial position corresponding to the frame image is represented as:

[0101] in, This is the starting position for the scan. This is the axial scan step distance between two adjacent images.

[0102] Then, the DOG focus response is calculated for each frame of the image. Specifically, the k-th frame is filtered using two Gaussian kernels with different standard deviations (i.e., different spatial smoothing scales) to obtain two sets of smoothed images. The DOG response is expressed as:

[0103] in, G σ1 and G σ2 The standard deviations are respectively represented as σ 1 and σ The Gaussian kernel function of 2, and σ 2> σ 1. Symbols This represents the convolution operation. The spatial smoothing scale here is determined by the standard deviation of the Gaussian kernel. σ 1 corresponds to the response of a finer-scale structure. σ 2 corresponds to a wider scale background response; the DOG differential response is used to enhance edge, texture, and gray-level abrupt change information related to the focus state. To avoid positive and negative responses canceling each other out and to enhance the local structural response of the image, the squared value of the DOG response is used as the focus evaluation value in this embodiment:

[0104] in, Indicates pixel position In the The DOG focus evaluation value in the frame image. Through the above processing, the focus response of the micropore opening edge, the transition region of the pore wall, and the local gray-scale abrupt change region can be enhanced, providing a foundation for subsequent candidate peak extraction.

[0105] Then, for each pixel position Extract the DOG focus evaluation value in all scanned layers and establish a focus response curve that varies along the axial scan direction:

[0106] like Figure 2 As shown, for microporous structures, since the aperture edge, inclined aperture wall, and observable area within the aperture may correspond to different height positions, the focus response curve of the same pixel or local area may exhibit multiple local peaks. Therefore, this embodiment no longer selects only the global maximum peak as the unique depth position, but extracts multiple candidate focus peaks from the focus response curve to retain the multi-height response information that may exist in the microporous region.

[0107] To reduce the impact of image noise and local high-frequency fluctuations on candidate peak extraction, the focus response curve is smoothed. In this embodiment, a three-point mean smoothing method is used:

[0108] in, This represents the smoothed focus response value.

[0109] Candidate focused peaks are extracted using the local maximum criterion. If the... The layer satisfies:

[0110] And it satisfies:

[0111] Then The corresponding positions are used as candidate focused peaks. To avoid local maxima formed by weak noise fluctuations being misjudged as candidate peaks, further conditions such as minimum peak intensity, minimum peak prominence, and minimum peak spacing are set. Let the first... The peak intensity of each candidate peak is Peak prominence is Then we can set:

[0112]

[0113] in, The background response value in the neighborhood of the candidate peak can be determined by the local valleys or local mean values ​​on both sides of the candidate peak. When the candidate peak satisfies:

[0114]

[0115] At that time, it was used as a preliminary valid candidate peak. Among them, Peak intensity threshold, This is the peak prominence threshold. This method allows for the initial removal of some noise peaks while retaining multiple potential true peaks.

[0116] To improve the axial depth positioning accuracy, this embodiment further performs sub-step peak position correction on the candidate peaks. Specifically, the sub-step offset is not manually specified, but rather selected from the candidate peaks. k i and its adjacent positions k i 1. k i The focus response value of +1 is used for quadratic curve fitting to calculate the offset of the candidate peak relative to the integer frame number. Δk i This is represented as:

[0117] For simplicity, pixel coordinates (x, y) are omitted. Δk i The dimensionless offset in units of the scan layer interval typically satisfies 0.5≤ Δk i ≤0.5. If the absolute value of the denominator is less than the preset threshold, or the calculation result exceeds the above range, it indicates that the peak shape of the candidate peak is unstable, and the integer layer position or the value of the candidate peak can be used. Δk i Amplitude limiting is performed. After substep correction, the candidate depth value corresponding to the i-th candidate peak is:

[0118] This allows for the expansion of depth localization from discrete integer focal planes to positions between adjacent focal planes without increasing the number of image acquisition layers, reducing quantization errors caused by axial scanning step size and improving the calculation accuracy of depth positions corresponding to candidate peaks.

[0119] After obtaining multiple candidate peaks, it is necessary to evaluate their reliability. If all candidate peaks are used directly for 3D reconstruction, noise peaks and spurious peaks are easily introduced into the point cloud; if only the largest peak is retained, the true weak hole wall peaks may be lost. Therefore, this embodiment constructs a comprehensive confidence evaluation model for candidate peaks from four aspects: peak intensity, peak prominence, peak sharpness, and neighborhood support.

[0120] The sharpness of the candidate peak can be expressed in the form of a second-order difference:

[0121] in, Indicates the first The sharpness of each candidate peak. The larger the value, the more concentrated the candidate peaks are near the focal point, and the sharper the peak shape.

[0122] To eliminate differences in the dimensions of different characteristics, peak intensity, peak prominence, and peak sharpness are normalized:

[0123]

[0124]

[0125] in, , and These represent the normalized results for peak intensity, peak prominence, and peak sharpness, respectively. This serves as a reference threshold for peak sharpness. This represents a truncation function used to restrict input values ​​to a specific range. Within the range.

[0126] Furthermore, neighborhood support is introduced to evaluate the spatial continuity of candidate peaks. Let candidate peaks be... Located at pixel position The corresponding frame number is , Indicates Center, radius The neighborhood window. If a candidate peak exists in the neighboring pixels. And its frame number With the current candidate peak frame number The difference does not exceed the allowable tolerance. If a candidate peak in the neighborhood is found to provide valid support for the current candidate peak, then that candidate peak is considered to provide effective support. The neighborhood support number is represented as:

[0127] in, Indicates candidate peak The number of neighborhood support, This is an indicator function; it takes the value 1 when the condition within the parentheses is true, and 0 otherwise.

[0128] If the maximum possible number of supports in the neighborhood is Then the neighborhood support normalization result is:

[0129] Based on this, the overall confidence score of the candidate peak is expressed as:

[0130] in:

[0131] in, express The overall confidence score of each candidate peak; , , , These represent the weighting coefficients corresponding to peak intensity, peak prominence, peak sharpness, and neighborhood support, respectively.

[0132] like Figure 3 As shown, based on the comprehensive confidence score and neighborhood support of the candidate peaks, they are classified into high-confidence peaks, medium-confidence peaks, and low-confidence peaks. The confidence level of the candidate peaks is expressed as follows:

[0133] in, and These are the score thresholds for high confidence and medium confidence, respectively. ; and These represent the minimum number of neighborhood support required for high confidence and medium confidence levels, respectively. .

[0134] For high-confidence peaks, they are directly retained as valid depth peaks; for low-confidence peaks, they are directly discarded; for medium-confidence peaks, further screening based on pore wall structure constraints is performed. This avoids the accidental deletion of genuine weak pore wall peaks due to simple single-threshold screening, and also avoids interference from spurious peaks caused by directly retaining all candidate peaks.

[0135] like Figure 4 As shown, hole wall structure constraints are used to screen peaks with medium confidence levels. These constraints include hole wall radius constraints, hole-axis compatibility constraints, and local continuity constraints.

[0136] Let the projection center of the micro-pore axis on the image plane be... pixel position The radial distance to the center of the hole shaft projection is:

[0137] When the pixel position of the candidate peak satisfies:

[0138] At that time, it was considered that the candidate peak fell within the suspected pore wall radius zone. Among them, and These represent the minimum and maximum radial distances corresponding to the suspected borehole wall region, respectively.

[0139] Furthermore, let the radius of the target hole wall be... The permissible radial deviation is When satisfied At that time, it was considered that the candidate peak was compatible with the expected pore wall geometry.

[0140] For local continuity constraints, it is determined whether a candidate peak has supporting peaks of similar depth and sufficient confidence within its spatial neighborhood. This embodiment uses a four-neighborhood direction criterion: a candidate peak is considered to satisfy the local continuity constraint when it has effective neighborhood support in at least two of the four directions (up, down, left, and right).

[0141] in, Indicates candidate peak The number of effective support directions in the four neighboring directions.

[0142] Taking into account the above constraints, the final retention condition for the intermediate confidence peak is expressed as:

[0143] in, This indicates that the candidate peak has been preserved. This indicates that the candidate peak has been removed; This represents the set of candidate peaks that satisfy the radius constraint of the hole wall. This represents the set of candidate peaks that satisfy the hole-shaft compatibility constraint. This represents the set of candidate peaks that satisfy the local continuity constraint.

[0144] Finally, high-confidence peaks and medium-confidence peaks satisfying the hole wall structure constraints are selected as valid candidate peaks. Based on the pixel coordinates, object-space pixel size, scan step size, and sub-step size correction amount corresponding to the candidate peaks, they are mapped to three-dimensional spatial points. Let the pixel column coordinates be... The row coordinates are The image reference center is The object pixel sizes are respectively and Then the coordinates of a three-dimensional point are represented as:

[0145]

[0146] Z= z i = z 0+( k i 1+ Δk i ) Δz Among them, the Z coordinate z i The results already include substep correction. Δki Δz This represents the axial refinement of the candidate peak relative to the integer scan layer; therefore, the depth coordinates of the 3D point cloud can reflect the peak position between adjacent focal planes, rather than just the position of discrete scan layers.

[0147] When multiple valid candidate peaks satisfying the conditions exist at the same pixel location, multiple depth points can be retained:

[0148] in, This indicates the number of valid candidate peaks retained at that pixel location. Ultimately, a micropore depth map or 3D point cloud containing multi-height information is generated, which is used for subsequent extraction of geometric parameters such as pore diameter, pore axis orientation, and pore wall morphology.

[0149] like Figure 5 As shown, compared with the traditional single-peak depth recovery method, the method of the present invention can retain more continuous depth points in the weak response region of the borehole wall, reduce the missing points and height jumps of the point cloud on the side of the borehole wall, and make the obtained point cloud more suitable for subsequent evaluation of geometric parameters such as borehole diameter, borehole axis direction and borehole wall morphology.

[0150] This embodiment calculates the DOG focus response of a multi-focal-plane image sequence of micropores, establishing a pixel-level axial focus response curve. Multiple candidate focus peaks are extracted from this curve, thus preserving the multi-height response information that may exist in the micropore structure. Based on this, the calculation accuracy of candidate depth positions is improved through sub-step peak position correction, and the confidence of candidate peaks is comprehensively evaluated by combining peak intensity, peak prominence, peak sharpness, and neighborhood support. Furthermore, medium-confidence candidate peaks are screened using pore wall radius constraints, pore axis compatibility constraints, and local continuity constraints. Finally, effective candidate peaks are mapped to depth maps or 3D point clouds. This method can suppress noise peaks and spurious peaks while preserving the depth information corresponding to the weak responses of the real pore walls, thereby improving the integrity, continuity, and stability of depth recovery in the micropore wall region.

[0151] In summary, this invention provides a method and system for multi-peak depth recovery of micro-apertures based on DOG focus response. Addressing the problem of easily losing the weak response of the true aperture wall in "single-peak maximum depth determination" during micro-aperture zoom microscopy, a multi-peak depth recovery mechanism based on DOG focus response is established. By performing differential Gaussian processing on multi-focal image sequences, the structural responses related to the focus state, such as aperture edges, aperture wall texture, and local gray-scale abrupt changes, can be highlighted, while the influence of illumination background and low-frequency interference on focus evaluation is reduced.

[0152] Furthermore, this invention extracts multiple candidate focusing peaks from the pixel-level axial focusing response curve, and no longer uses the global maximum peak as the only depth position. This allows the preservation of multi-height response information corresponding to the orifice, inclined orifice wall, and observable area inside the orifice, reducing the risk that the true orifice wall depth is covered by strong reflection peaks or orifice edge peaks.

[0153] Meanwhile, this invention refines the depth position of candidate peaks through sub-step peak position correction, so that the depth value is not limited to discrete scanning layers, thereby improving the axial depth positioning accuracy. By constructing a comprehensive confidence evaluation based on peak intensity, peak prominence, peak sharpness, and neighborhood support, it can effectively distinguish between real hole wall peaks, noise peaks, and reflective pseudo-peaks.

[0154] Furthermore, this invention introduces constraints on the borehole wall radius, borehole axis compatibility, and local continuity to perform secondary screening on medium-confidence peaks. This allows weak response peaks that conform to the geometry of the micropore wall to be retained, while isolated spurious peaks are eliminated. The resulting depth map or 3D point cloud exhibits better borehole wall continuity, integrity, and stability, which is beneficial for the accurate evaluation of subsequent geometric parameters such as borehole diameter, borehole axis direction, and borehole wall morphology.

[0155] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0156] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0157] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0158] In the embodiments provided by this invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0159] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0160] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0161] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random-access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0162] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0163] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0164] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0165] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method for recovering the multi-peak depth of micropores based on DOG focusing response, characterized in that, Includes the following steps: S1. Obtain a multi-focal plane image sequence of the micropore under different focal plane positions; perform differential Gaussian processing on each frame of the multi-focal plane image sequence to obtain the DOG response, and obtain the DOG focus evaluation value based on the DOG response. S2. For pixel positions, extract DOG focus evaluation values ​​at different focal plane positions along the axial scanning direction, establish pixel-level focus response curves, and extract multiple candidate focus peaks from the pixel-level focus response curves based on local maximum criteria. S3. Based on the DOG focusing evaluation value of each candidate focusing peak and its adjacent scanning layers, perform sub-step peak position correction on the candidate focusing peak to obtain the corresponding candidate depth value, and calculate the comprehensive confidence score based on the peak intensity, peak prominence, peak sharpness and neighborhood support of the candidate focusing peak. S4. The candidate focused peaks are classified according to the comprehensive confidence score and neighborhood support, and the candidate focused peaks after confidence classification are screened according to the hole wall structure constraints to obtain effective candidate peaks; wherein, the hole wall structure constraints include hole wall radius zone constraints, hole axis compatibility constraints and local continuity constraints. S5. Generate a micropore depth map or a three-dimensional point cloud based on the pixel coordinates, object pixel size, axial scanning step distance and candidate depth value corresponding to the effective candidate peak.

2. The micropore multi-peak depth recovery method based on DOG focusing response according to claim 1, characterized in that, In S1, each frame of the multi-focal plane image sequence is filtered using a first Gaussian kernel and a second Gaussian kernel. The first Gaussian kernel and the second Gaussian kernel have different standard deviations. The difference between the two filtering results is used to obtain the DOG response. The amplitude or square value of the DOG response is used as the DOG focus evaluation value.

3. The micropore multi-peak depth recovery method based on DOG focusing response according to claim 1, characterized in that, In S2, before extracting multiple candidate focus peaks, the pixel-level focus response curve is smoothed, and the scan layer position that satisfies the local maximum condition in the smoothed pixel-level focus response curve is determined as the candidate focus peak.

4. The micropore multi-peak depth recovery method based on DOG focusing response according to claim 1, characterized in that, In S2, when extracting the candidate focused peaks, the scanning layer positions that satisfy the local maximum condition are also screened based on the peak intensity, peak prominence, and peak spacing.

5. The micropore multi-peak depth recovery method based on DOG focusing response according to claim 1, characterized in that, In S3, the substep peak position correction includes: performing quadratic curve interpolation based on the DOG focus evaluation values ​​of the scanning layer where the candidate focus peak is located and its adjacent scanning layers before and after it to obtain the substep offset of the candidate focus peak relative to the integer scanning layer, and calculating the candidate depth value based on the scanning start position, axial scanning step distance, integer scanning layer position and the substep offset.

6. The method for recovering the depth of micropores based on DOG focusing response according to claim 1, characterized in that, In S3, the neighborhood support is determined based on the number of supporting peaks in the spatial neighborhood of the candidate focus peak. The supporting peak is a neighboring candidate focus peak whose scan layer position difference from the current candidate focus peak does not exceed a preset tolerance.

7. The micropore multi-peak depth recovery method based on DOG focusing response according to claim 1, characterized in that, In S4, classifying the candidate focus peaks according to the comprehensive confidence score and neighborhood support includes: classifying candidate focus peaks that meet the high confidence score threshold and high confidence neighborhood support conditions as high confidence peaks; classifying candidate focus peaks that are not classified as high confidence peaks but meet the medium confidence score threshold and medium confidence neighborhood support conditions as medium confidence peaks; and classifying candidate focus peaks that are not classified as high confidence peaks or medium confidence peaks as low confidence peaks.

8. The method for recovering the multi-peak depth of micropores based on DOG focusing response according to claim 1, characterized in that, In S4, the hole wall structure constraint screening includes: determining whether the candidate focused peak satisfies the hole wall radius constraint based on the radial distance from the pixel position to the hole axis projection center; determining whether the candidate focused peak satisfies the hole axis compatibility constraint based on the deviation between the radial distance and the target hole wall radius; and determining whether the candidate focused peak satisfies the local continuity constraint based on whether there is an effective neighborhood of similar depth supporting the candidate focused peak in the spatial neighborhood.

9. The method for recovering the depth of micropores based on DOG focusing response according to claim 1, characterized in that, In S5, the three-dimensional spatial coordinates are calculated based on the pixel coordinates corresponding to the effective candidate peaks, the image reference center, the object pixel size, and the candidate depth value. When there are multiple effective candidate peaks at the same pixel location, multiple three-dimensional depth points are generated based on the candidate depth values ​​of the multiple effective candidate peaks.

10. A microporous multi-peak depth recovery system based on DOG focusing response, characterized in that, include: The image acquisition and response calculation module is used to acquire a multi-focal plane image sequence of the micro-hole under different focal plane positions, perform differential Gaussian processing on each frame of the multi-focal plane image sequence to obtain the DOG response, and obtain the DOG focus evaluation value based on the DOG response. The candidate peak extraction module is used to extract DOG focus evaluation values ​​at different focal plane positions along the axial scanning direction for pixel positions, establish pixel-level focus response curves, and extract multiple candidate focus peaks from the pixel-level focus response curves according to the local maximum criterion. The correction and evaluation module is used to perform sub-step peak position correction on each candidate focus peak and its adjacent scan layer based on the DOG focus evaluation value, to obtain the corresponding candidate depth value, and to calculate the comprehensive confidence score based on the peak intensity, peak prominence, peak sharpness and neighborhood support of the candidate focus peak. The constraint filtering module is used to classify the candidate focused peaks according to the comprehensive confidence score and neighborhood support, and to filter the candidate focused peaks after confidence classification according to the hole wall structure constraints to obtain effective candidate peaks; wherein, the hole wall structure constraints include hole wall radius zone constraints, hole axis compatibility constraints and local continuity constraints; The depth recovery module is used to generate a micropore depth map or a three-dimensional point cloud based on the pixel coordinates, object pixel size, axial scan step size and candidate depth value corresponding to the effective candidate peak.

Images