A method and system for monitoring wear of a cemented carbide tool based on grating projection image processing

Abstract

The application relates to the technical field of cutter wear image monitoring, and discloses a cemented carbide cutter wear monitoring method and system based on grating projection image processing, wherein the method comprises the following steps: acquiring cutter rotation synchronization information of a cemented carbide cutter; performing grating projection image acquisition; performing residual dynamic blur suppression processing; performing cutter three-dimensional profile reconstruction by adopting a grating phase reliable decoding mechanism; constructing a wear candidate stable region; and outputting a cemented carbide cutter wear monitoring result. Compared with the method based on static grating projection or single-frame image processing in the prior art, especially under the condition of high-speed rotating cutter online processing, the dynamic blur caused by rotation makes it difficult to realize stable identification of the wear region. By introducing cutter rotation synchronization control, rotation domain dynamic blur suppression and rotation period consistency constraint, the stability of the cutter wear monitoring result is improved.

Description

Technical Field

[0001] This invention relates to the field of tool wear image monitoring technology, and in particular to a method and system for monitoring the wear of cemented carbide tools based on grating projection image processing. Background Technology

[0002] Currently, the wear condition of cemented carbide cutting tools is a crucial factor affecting machining accuracy, efficiency, and tool life. In machining scenarios such as CNC milling, tool wear typically requires online or near-online monitoring. However, existing tool wear monitoring methods largely rely on indirect signals such as cutting force, vibration, and acoustic emission, or employ optical imaging and three-dimensional measurement under static conditions, which suffer from insufficient applicability in real-world high-speed machining environments.

[0003] For example, in tool wear detection methods based on grating projection or structured light imaging, existing technologies typically assume that the tool under test is stationary or rotating at low speed, obtaining the tool surface morphology by performing phase decoding and 3D reconstruction on single frames or a small number of images. However, under high-speed rotational machining conditions, the continuous rotation of the tool causes significant dynamic blur, stripe trailing, and phase distortion in the grating projection images. This makes it difficult for image processing and phase decoding methods based on static models to stably extract the true tool contour information, thus affecting the accurate identification of wear areas and the reliable calculation of wear. Furthermore, although some existing technologies attempt to alleviate the dynamic blur problem by increasing the sampling frequency or the number of images, these methods often do not fully consider the periodic characteristics and rotational consistency constraints during tool rotation. They are easily affected by instantaneous noise, random reflections, or local reconstruction errors, leading to unstable wear detection results and making it difficult to operate reliably for a long time under actual high-speed rotation and continuous machining conditions.

[0004] Therefore, there is an urgent need for a method that can reliably acquire clear raster projection images and reliably identify and quantitatively assess tool wear areas under high-speed rotation conditions without reducing tool speed or affecting normal machining processes, so as to improve the stability, repeatability and engineering application value of cemented carbide tool wear monitoring results. Summary of the Invention

[0005] To address the aforementioned technical shortcomings, the present invention aims to propose a method for monitoring the wear of cemented carbide cutting tools based on grating projection image processing. This method addresses the technical problem that existing methods, which are mostly based on static grating projection or single-frame image processing, struggle to achieve stable identification of wear areas, especially under high-speed rotating tool online machining conditions due to dynamic blurring caused by rotation.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a method for monitoring the wear of cemented carbide cutting tools based on grating projection image processing.

[0007] The method for monitoring wear of cemented carbide cutting tools based on grating projection image processing includes:

[0008] Step S10: Obtain the tool rotation synchronization information of the cemented carbide tool. Based on the tool rotation synchronization information, use a synchronous stroboscopic acquisition method based on phase-gated consistency verification to acquire the grating projection image and obtain the original grating projection image. ;

[0009] Step S20: For the original raster projection image An image optimization task is performed using a residual dynamic blur suppression mechanism based on rotation domain constraints, outputting an optimized raster projection image. ;

[0010] Step S30: Based on the optimized raster projection image A reliable grating phase decoding mechanism is used to perform the tool 3D contour reconstruction task, and outputs a 3D contour data set P.

[0011] Step S40: Based on the 3D contour data set P, a 3D topography mutation screening mechanism based on rotation period consistency constraints is used to construct wear candidate stable regions, and a set of wear candidate stable regions is output. ;

[0012] Step S50: Based on the set of candidate stable regions for wear A wear region quantitative evaluation mechanism based on spatial weighted constraints is adopted to perform wear region locking and quantitative wear parameter calculation tasks, and finally output the wear monitoring results of cemented carbide tools.

[0013] Preferably, in step S10, the tool rotation synchronization information of the cemented carbide tool is obtained, and based on the tool rotation synchronization information, a synchronous stroboscopic acquisition method based on phase-gated consistency verification is used to acquire the grating projection image to obtain the original grating projection image. The steps specifically include:

[0014] Step S101: Obtain the tool rotation synchronization information of the carbide tool at time t. The tool rotation synchronization information includes the spindle encoder angle synchronization signal. With angular velocity Set the target phase-locked angle. Lock-in tolerance window According to the spindle encoder angle synchronization signal Target phase-locked angle Lock-in tolerance window Define phase-locked gate function , ;in, An angle wrap-around operator is used to map the angle difference to the interval (−π,π) to eliminate the phase jump effect caused when the principal axis angle crosses a 2π period.

[0015] Step S102: In the phase-locked gate function The phase-locked window emits A short pulse flash, in The grating projection sub-exposure image is acquired within a short pulse strobe, and the target phase-locked angle corresponding to the m-th short pulse strobe is obtained. Angle synchronization signal with spindle encoder Instantaneous phase deviation between ;

[0016] Step S103: When the instantaneous phase deviation If the instantaneous phase deviation does not exceed the preset threshold, the m-th short pulse strobe is determined to have phase consistency. The grating projection sub-exposure images acquired within the short pulse strobe with phase consistency are fused using an accumulation integration method based on phase consistency constraints, and the original grating projection image is output. .

[0017] Preferably, in step S20, the original raster projection image is... An image optimization task is performed using a residual dynamic blur suppression mechanism based on rotation domain constraints, outputting an optimized raster projection image. The steps specifically include:

[0018] Step S201: First, using the rotation center point of the carbide tool as the origin of the polar coordinates, project the original raster image... The rotating domain grating projection image is obtained by mapping from the Cartesian coordinate system to the polar coordinate system. ;

[0019] Step S202: Based on the physical characteristics of the movement of the carbide tool's rotation center point, including the rotational angular characteristics of the tool's angular rotation around a fixed rotation center; based on the rotational angular characteristics, a one-dimensional deconvolution method is used to project the rotation domain grating image. The residual dynamic blur in the image is constructed as a one-dimensional convolutional rotation domain sharp raster projection image. ;

[0020] Step S203: Finally, project the one-dimensional convolutional rotation domain sharpened raster image. The polar coordinate system is inversely transformed to the Cartesian coordinate system to output an optimized raster projection image. .

[0021] Preferably, in step S30, the optimized raster projection image is used... The steps for performing tool 3D contour reconstruction using a grating phase reliable decoding mechanism and outputting a 3D contour data set P specifically include:

[0022] Step S301: Obtain the optimized raster projection image The projected fringe data is used to estimate parameters by local least squares fitting, outputting the gray-level distribution relationship of the grating fringes. Based on the gray-level distribution relationship, an arctangent phase solver is used to calculate and optimize the grating projection image. The initial phase value corresponding to the pixel with x-coordinate and y-coordinate. ;

[0023] Step S302: Obtain the position of the pixel with x-coordinate and y-coordinate. The corresponding tool rotation angle is used to filter out outliers based on a threshold-based method, and the initial phase value is then... Group the data and calculate the phase change between adjacent groups. When the phase change amount When the position is less than the preset rotation phase consistency threshold, the position is determined. The phase of the corresponding pixel is the reliable phase; when the phase change amount When the position is greater than or equal to the preset rotation phase consistency threshold, the position is determined. The phase of the corresponding pixel is considered an unreliable phase; for pixels with unreliable phases, phase correction or removal is performed to obtain reliable grating phase values. ;

[0024] Step S303: Based on the reliable grating phase value Combined with preset system calibration parameters, perform tool 3D contour reconstruction processing and output a 3D contour data set P.

[0025] Preferably, step S303, which involves performing tool 3D contour reconstruction based on the reliable grating phase value and preset system calibration parameters, and outputting a 3D contour data set P, specifically includes:

[0026] Step S3031: Obtain preset system calibration parameters, including the projected baseline distance. And the geometric relationship parameters of projection imaging; based on the system calibration parameters, the reliable grating phase value is converted into the tool surface height value using the triangulation principle. ;

[0027] Step S3032: Based on the geometric relationship parameters of projection imaging and the tool surface height value The projective inversion mapping method is used to perform the mapping process from two-dimensional pixel coordinates to three-dimensional spatial coordinates, and outputs a reliable phase point in three-dimensional space.

[0028] Step S3033: Represent the credible phase points in three-dimensional space as a set, and construct and output the three-dimensional contour data set P.

[0029] Preferably, in step S40, a three-dimensional morphology mutation screening mechanism based on rotation period consistency constraints is used to construct wear candidate stable regions based on the three-dimensional contour data set P, and a set of wear candidate stable regions is output. The steps specifically include:

[0030] Step S401: Local Topography Gradient Construction Stage: Based on the 3D contour data set P, a topography extraction method based on local gradient calculation is used to extract the local topography gradient in the 3D height field. ;

[0031] Step S402: Morphological mutation threshold screening stage: setting morphological mutation threshold Based on morphological abrupt change threshold and local topography gradient Constructing candidate masks , ,in This is an indicator function used to represent regions that meet a threshold condition; for candidate masks Connectivity analysis is performed using a connectivity labeling method based on the eight-neighbor connectivity rule, outputting connected regions, which are then used as the initial set of candidate wear regions. ;

[0032] Step S403: Stability Verification Stage: A complete tool rotation cycle k is preset. If a candidate mask continuously exists within the complete tool rotation cycle k... At the pixel position with x-coordinate and y-coordinate When the value at a location is 1, determine the position. The pixels at that location are stable pixels; the final set of candidate wear regions is merged. Output a set of candidate stable wear regions for all stable pixels within the range. .

[0033] Preferably, in step S50, based on the set of wear candidate stable regions... The steps involved in performing wear region locking and quantitative wear parameter calculation using a wear region quantification evaluation mechanism based on spatial weighted constraints, and finally outputting the wear monitoring results of carbide tools, specifically include:

[0034] Step S501: Obtain the set of candidate stable regions for wear For the corresponding contour height data, extract the weighted centroid of the contour height data, and use the weighted centroid as the center position of the wear area. Construct a two-dimensional Gaussian space weighted function with the center position of the wear area as the center. ;

[0035] ;in, The pixel with x-coordinate and y-coordinate is the position of the pixel. The two-dimensional Gaussian space weighted function at the location; This is a spatial expansion scale parameter for the wear region, used to adjust the decay rate of the contribution weight of the center and edge pixels in the wear amount calculation; The x-coordinate of the pixel at the center of the wear area; The ordinate of the pixel at the center of the wear area;

[0036] Step S502: Weighting function based on two-dimensional Gaussian space Set of candidate stable regions for wear The contour height data within the field is spatially weighted and locked to suppress abnormal height values ​​caused by rotation reconstruction errors or residual phase perturbations, and the spatially weighted contour height field is output.

[0037] Step S503: Calculate the wear parameters based on the weighted height difference statistical analysis method of neighborhood height difference according to the spatial weighted profile height field. The wear parameters include at least the weighted maximum wear depth, weighted wear area and weighted wear volume. Finally, output the wear monitoring results of carbide tools according to the wear parameters.

[0038] The present invention also provides a carbide tool wear monitoring system based on grating projection image processing, comprising:

[0039] The tool rotation synchronization and stroboscopic acquisition module is used to acquire the tool rotation synchronization information of cemented carbide tools. Based on the tool rotation synchronization information, a synchronous stroboscopic acquisition method based on phase-gated consistency verification is used to acquire grating projection images, resulting in the original grating projection image. ;

[0040] Rotation domain residual dynamic blur suppression module, used for original raster projection images An image optimization task is performed using a residual dynamic blur suppression mechanism based on rotation domain constraints, outputting an optimized raster projection image. ;

[0041] The grating phase reliable decoding and 3D reconstruction module is used for optimizing grating projection images. A reliable grating phase decoding mechanism is used to perform the tool 3D contour reconstruction task, and outputs a 3D contour data set P.

[0042] The morphological mutation screening module is used to construct wear candidate stable regions based on a 3D contour data set P using a 3D morphological mutation screening mechanism based on rotation period consistency constraints, and outputs a set of wear candidate stable regions. ;

[0043] The spatially weighted wear quantization assessment module is used to evaluate wear based on the set of candidate stable regions. A wear region quantitative evaluation mechanism based on spatial weighted constraints is adopted to perform wear region locking and quantitative wear parameter calculation tasks, and finally output the wear monitoring results of cemented carbide tools.

[0044] The present invention also provides a cemented carbide tool wear monitoring device based on raster projection image processing, comprising: a memory, a processor, and a cemented carbide tool wear monitoring program based on raster projection image processing stored in the memory and executable on the processor. When the cemented carbide tool wear monitoring program based on raster projection image processing is executed by the processor, a cemented carbide tool wear monitoring method based on raster projection image processing is implemented.

[0045] The present invention also provides a computer program product, including a carbide tool wear monitoring program based on raster projection image processing, wherein the carbide tool wear monitoring program based on raster projection image processing implements the carbide tool wear monitoring method based on raster projection image processing when executed by a processor.

[0046] The beneficial effects of this invention are as follows: By introducing a tool rotation synchronization control and a synchronous stroboscopic acquisition mechanism based on phase gating consistency verification, and combining it with residual dynamic blur suppression processing under rotation domain constraints, this invention effectively solves the problems of stripe trailing, phase distortion and contour blurring in grating projection images under high-speed rotation conditions, enabling the acquisition of grating projection images with clear structural information even under non-stop, high-speed operation.

[0047] Based on three-dimensional contour reconstruction, this invention introduces a morphological mutation screening mechanism with rotation cycle consistency constraints and a wear region quantitative evaluation mechanism based on spatial weighted constraints. By jointly constraining the temporal stability and spatial contribution of the wear region, it achieves reliable locking of the tool wear region and quantitative calculation of wear amount, thereby avoiding misjudgment caused by instantaneous noise, local reconstruction error or edge anomaly points, and improving the stability, repeatability and engineering practical value of wear monitoring results in high-speed rotational machining scenarios. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a flowchart illustrating the first embodiment of a method for monitoring wear of cemented carbide cutting tools based on grating projection image processing according to the present invention.

[0050] Figure 2 This is a schematic diagram showing the distribution of acquisition times for different image acquisition methods in the first embodiment of a cemented carbide tool wear monitoring method based on grating projection image processing according to the present invention.

[0051] Figure 3 This is a schematic diagram of grating projection under ideal static conditions, representing the first embodiment of a cemented carbide tool wear monitoring method based on grating projection image processing according to the present invention.

[0052] Figure 4 This is a magnified schematic diagram of a partial grating projection under ideal static conditions, representing a first embodiment of a cemented carbide tool wear monitoring method based on grating projection image processing according to the present invention.

[0053] Figure 5 This is a schematic diagram of grating projection under Cartesian domain processing conditions, representing a first embodiment of the cemented carbide tool wear monitoring method based on grating projection image processing according to the present invention.

[0054] Figure 6 This is a magnified schematic diagram of a partial grating projection under Cartesian domain processing conditions, representing a first embodiment of a cemented carbide tool wear monitoring method based on grating projection image processing according to the present invention.

[0055] Figure 7 This is a schematic diagram of grating projection under rotational domain constraint processing conditions, representing a first embodiment of a cemented carbide tool wear monitoring method based on grating projection image processing according to the present invention.

[0056] Figure 8 This is a magnified schematic diagram of a portion of the grating projection under rotational domain constraint processing conditions, representing a first embodiment of a cemented carbide tool wear monitoring method based on grating projection image processing according to the present invention.

[0057] Figure 9 This is a schematic diagram of a device for monitoring the wear of cemented carbide cutting tools based on grating projection image processing, according to the present invention. Detailed Implementation

[0058] 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 embodiments of the present invention, and not all embodiments. 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.

[0059] Example 1: As Figure 1 The diagram shown is a flowchart of the first embodiment of the cemented carbide tool wear monitoring method based on grating projection image processing of the present invention, which presents the first embodiment of the cemented carbide tool wear monitoring method based on grating projection image processing of the present invention.

[0060] In the first embodiment, the carbide tool wear monitoring method based on grating projection image processing includes:

[0061] Step S10: Obtain the tool rotation synchronization information of the cemented carbide tool. Based on the tool rotation synchronization information, use a synchronous stroboscopic acquisition method based on phase-gated consistency verification to acquire the grating projection image and obtain the original grating projection image. ;

[0062] The "synchronous stroboscopic acquisition method based on phase-gated consistency verification" in this step refers to the following: Before acquiring the grating projection image, tool rotation synchronization information that reflects the real-time rotation state of the cemented carbide tool is first obtained. Based on this synchronization information, rotation phase determination conditions are constructed. Only when the tool rotation phase corresponding to the current acquisition time meets the preset consistency gating conditions is the exposure and acquisition operation of the grating projection image triggered. Among them, the phase-gated consistency verification is not only used to limit the correspondence between the acquisition time and the tool rotation angle, but also to eliminate non-consistent acquisition times caused by small fluctuations in rotation speed, instantaneous load changes, or mechanical vibration, thereby ensuring the consistency and comparability of the acquired original grating projection image in the rotation phase dimension.

[0063] By introducing a phase-gated control mechanism based on tool rotation synchronization information during the grating projection image acquisition stage, this invention can transform the continuously changing tool motion state under high-speed rotation conditions into a set of image observation results that maintain a consistent rotation phase without changing the actual tool rotation speed or interfering with the normal machining process. This significantly reduces grating fringe displacement and exposure superposition caused by tool rotation, resulting in a marked improvement in the fringe clarity and structural stability of the acquired original grating projection image, providing a more reliable input basis for subsequent image optimization and 3D contour reconstruction.

[0064] Compared to traditional grating projection image acquisition methods that use fixed exposure periods or simple strobe control, this invention introduces a phase-gated consistency verification mechanism, transforming the image acquisition process from "time-driven" to "rotation-state-driven." This fundamentally avoids random blurring and phase drift problems caused by the mismatch between the acquisition timing and the actual rotation state of the tool under high-speed rotation conditions. At the same time, this method can effectively suppress the cumulative error caused by slight fluctuations in rotation speed, resulting in higher consistency in the spatial structure of continuously acquired multi-frame grating projection images. This is significantly superior to traditional technical solutions that rely solely on increasing the sampling frequency or shortening the exposure time.

[0065] For example, such as Figure 2 As shown in the figure, the continuously changing sawtooth curve represents the evolution of the tool rotation phase within one or more rotation cycles, while the vertical markers distributed along the horizontal axis represent the actual acquisition trigger moments of the raster projection image. In traditional fixed-time acquisition methods, image acquisition triggering mainly depends on a preset time interval, and there is no strict correspondence between this and the actual tool rotation phase. Therefore, the distribution of acquisition moments on the rotation phase axis is quite dispersed, and the rotation phases corresponding to different acquisition moments differ significantly. In this case, even with a high acquisition frequency, it is difficult to ensure that different acquisition frames are in the same rotation state, which easily introduces stripe trailing and phase inconsistency problems under high-speed rotation conditions. In contrast, this invention employs a synchronous stroboscopic acquisition method based on phase-gated consistency verification, whereby the image acquisition trigger moments are concentrated within a specific phase interval of the rotation phase curve. That is, in multiple consecutive rotation cycles, the acquisition operation occurs when the tool rotates to the same or approximately the same rotation phase position. In this way, the image acquisition process is no longer simply time-driven, but driven by the tool rotation state, ensuring consistency between acquisition frames from the rotation phase dimension and providing a stable phase reference basis for subsequent image processing.

[0066] Step S20: For the original raster projection image An image optimization task is performed using a residual dynamic blur suppression mechanism based on rotation domain constraints, outputting an optimized raster projection image. ;

[0067] The "residual dynamic blur suppression mechanism based on rotation domain constraint" in this step refers to the following: after completing the synchronous stroboscopic acquisition based on phase-gated consistency verification, a rotation domain modeling method that matches the characteristics of the tool rotation motion is introduced to suppress the blur components in the image, which still exists in the original raster projection image. The residual dynamic blur mainly comes from the small angular displacement during the exposure time, the slight fluctuation of the rotation speed, and the non-ideal imaging effect near the rotation center. Its blur direction has a clear correlation with the tool rotation direction.

[0068] Under high-speed rotation conditions, even with synchronous stroboscopic acquisition, a small amount of blurring components distributed along the rotation direction may still exist in the grating projection image. If traditional deblurring or smoothing methods are directly applied in the Cartesian coordinate system, the true grating fringe structure is easily mixed with the blurring components caused by rotation, thereby weakening the fringe edges or introducing additional artifacts. This invention introduces rotation domain constraints into the image processing process, ensuring that the modeling and suppression direction of residual dynamic blur is consistent with the actual rotational motion of the tool. This effectively reduces the blurring effect along the rotation direction while preserving the true structural information of the grating fringe.

[0069] For example, such as Figure 3 As shown, a grating projection reference image obtained under ideal static or equivalent static conditions is presented, serving as a reference for the fringe structure. It can be observed that the grating fringes exhibit a regular, equidistant distribution in space, with clear fringe edges, stable transitions between light and dark areas, and almost no fringe tilting, stretching, or blurring. Figure 4 As shown, the stripes in the figure have consistent periods and continuous phases, which can accurately reflect the true structural characteristics of the grating projection under conditions without rotational interference.

[0070] like Figure 5 The image shows the result of applying traditional image processing methods (such as conventional sharpening, deconvolution, or spatial domain filtering) directly to the original blurred image in the Cartesian coordinate system. Overall, the image contrast is enhanced, the differences in stripe brightness are amplified, and local areas appear "sharper." However, this processing method does not distinguish between the actual raster structure and the directional blur component caused by rotation. Figure 6 As shown in the figure, it can be clearly observed that although the edges of the stripes become steeper, problems such as stripe breakage, local ringing, and pseudo-high-frequency noise enhancement also appear. In some areas, even non-physical stripe ghosting occurs. This indicates that the traditional Cartesian domain processing method is prone to amplifying the structural distortion caused by rotational blur in high-speed rotating scenes.

[0071] like Figure 7 As shown, this paper presents a raster projection image processed using the residual dynamic blur suppression mechanism based on rotation domain constraints proposed in this invention. Unlike traditional methods, this scheme explicitly introduces the tool rotation characteristics during processing, mapping the image to a rotation domain (polar coordinate domain) consistent with the rotational motion, and performing directional modeling and suppression of residual blur along the rotational tangential direction. From the overall effect, the fringe structure is more uniform globally, the trailing shadow in the rotational direction is effectively weakened, while the radial and fringe normal structures are well preserved. Figure 8 As shown, in the magnified area, it can be clearly seen that the stripes maintain both high contrast and good structural continuity, most closely resembling the true form of a clear reference.

[0072] Step S30: Based on the optimized raster projection image A reliable grating phase decoding mechanism is used to perform the tool 3D contour reconstruction task, and outputs a 3D contour data set P.

[0073] The "raster phase reliability decoding mechanism" in this step refers to the following: after completing the residual dynamic blur suppression based on rotation domain constraints, instead of directly performing uniform phase decoding on all pixels in the optimized raster projection image, a phase reliability judgment process is introduced. This process assesses the reliability of the raster phase calculation result for each pixel, and only phase values ​​that meet preset reliability conditions are included in the subsequent 3D contour reconstruction process. The phase reliability judgment comprehensively considers factors such as raster fringe contrast, phase continuity, rotational phase consistency, and local grayscale stability. It is used to identify abnormal phase points caused by residual noise, local occlusion, fringe distortion, or reconstruction errors, thereby avoiding interference from unreliable phases in the 3D contour reconstruction results.

[0074] In the grating projection measurement scenario of high-speed rotating cemented carbide cutting tools, even after residual dynamic blur suppression processing based on rotation domain constraints, grating projection images may still exhibit fringe distortion, grayscale fluctuations, or decreased signal-to-noise ratio in local areas, especially at locations where the tool edge, wear edge, or surface reflection characteristics change abruptly. Phase calculation results in these areas are often unstable; if directly used in 3D contour reconstruction, they can easily manifest as local height abrupt changes, noise protrusions, or spurious wear features in the 3D model. This invention introduces a reliability constraint in the phase decoding stage, ensuring that 3D contour reconstruction is based solely on physically meaningful and computationally stable phase information, thereby improving the overall spatial continuity and geometric consistency of the 3D contour data set P.

[0075] Step S40: Based on the 3D contour data set P, a 3D topography mutation screening mechanism based on rotation period consistency constraints is used to construct wear candidate stable regions, and a set of wear candidate stable regions is output. ;

[0076] The "3D morphology mutation screening mechanism based on rotation cycle consistency constraints" in this step refers to the following: after obtaining the tool's 3D contour data set P, instead of judging the wear area solely based on morphology changes at a single moment or within a single rotation cycle, the tool rotation cycle is introduced as a consistency constraint dimension to jointly determine the repeatability and stability of local morphology mutation features in the 3D contour across multiple complete rotation cycles. Here, a 3D morphology mutation refers to a region that exhibits significant changes in morphology indicators such as height, slope, or curvature relative to adjacent regions. In real wear scenarios, this change should maintain spatial consistency throughout the tool rotation cycle, rather than appearing randomly or fluctuating briefly.

[0077] In the online inspection environment of high-speed rotating tools, even if the overall quality of the 3D contour reconstruction result is high, it may still be affected by factors such as measurement noise, uneven surface reflection, and residual phase error, resulting in instantaneous 3D morphology anomalies in local areas. If wear determination is based solely on the 3D morphology changes of a single frame or a single cycle, these random anomalies are easily misjudged as wear features.

[0078] Compared to common wear detection methods based on single 3D scan results or simple height thresholds, this invention does not rely solely on the magnitude of morphological abrupt changes. Instead, it further introduces consistency judgment in the time-rotation cycle dimension, upgrading the selection of wear candidate regions from "static geometric judgment" to "dynamic consistency judgment constrained by the physical constraints of rotational motion." This approach is particularly suitable for high-speed rotation and continuous processing environments. Without adding extra sensors or increasing the scanning frequency, it significantly reduces the false detection rate, improves the stability and reliability of wear candidate regions, and has stronger engineering applicability.

[0079] For example, in an experiment involving continuous online monitoring of a high-speed milling carbide tool, morphological abrupt changes in the tool's 3D contour data set P were analyzed over multiple continuously acquired rotation cycles. Some local height abrupt changes caused by phase noise or reflection anomalies only occurred within a single rotation cycle and did not repeat in the same spatial location in subsequent rotation cycles. In contrast, the 3D morphological abrupt changes corresponding to the actual wear area remained stable over multiple consecutive rotation cycles, and their spatial distribution was basically consistent. Through the rotation cycle consistency constraint screening mechanism in this step, only the morphological abrupt change areas that repeatedly occurred in multiple rotation cycles were retained as wear candidate stable areas, thereby effectively eliminating random outliers. This makes the output set of wear candidate stable areas more stable and reliable in terms of spatial location and morphological characteristics, providing a high-confidence input for subsequent quantitative assessment of wear areas.

[0080] Step S50: Based on the set of candidate stable regions for wear A wear region quantitative evaluation mechanism based on spatial weighted constraints is adopted to perform wear region locking and quantitative wear parameter calculation tasks, and finally output the wear monitoring results of cemented carbide tools.

[0081] The "spatial weighted constraint-based quantitative assessment mechanism for wear regions" in this step refers to the following: After obtaining the set of candidate stable wear regions, instead of calculating wear parameters with equal weights for all 3D contour data points within that region, a spatial weighting function related to the center position of the wear region is introduced, using the spatial distribution characteristics of the wear region as a constraint. Different weights are assigned to the 3D height data at different locations within the region, thereby achieving stable locking of the wear region and reliable calculation of quantitative wear parameters. Specifically, the spatial weighting function is used to highlight the contribution of the area near the center of the wear region to the assessment of wear degree, while suppressing abnormal height values ​​at the region edges that may be affected by reconstruction errors or phase residual disturbances.

[0082] Compared to wear calculation methods based on regional extreme values, regional averages, or simple integration, this invention does not simply pursue obtaining the "maximum value" or "most significant change" wear index in a single detection. Instead, it emphasizes the stability and repeatability of wear parameters during multiple detections and long-term operation. Through a spatial weighted constraint mechanism, the wear calculation results are no longer highly sensitive to single-point anomalies or local noise, but rather reflect the dominant changing trend of the overall morphology of the wear area, thus significantly reducing the risk of misjudgment. This characteristic is particularly important for applications such as high-speed rotation and continuous machining where frequent machine stoppages for calibration are not possible. It allows wear monitoring results to be directly used for process decisions or life assessments without additional manual intervention. For example, in a series of online monitoring experiments for long-term continuous machining of carbide end mills, wear parameters were periodically output at different machining stages. When using the traditional equal-weighted region statistical method, in some detection cycles, the appearance of individual abnormal height points at the edge of the wear area caused a short-term surge in the calculated maximum wear depth, thus affecting the judgment of the wear trend. By adopting the wear region quantification evaluation mechanism based on spatial weighted constraints of the present invention, the stable height change of the wear region center plays a dominant role in the calculation, and the influence of edge anomalies on the results is effectively suppressed, so that parameters such as wear depth, wear area and wear volume show a smoother, more monotonous change curve that conforms to the actual wear evolution law with processing time.

[0083] Example 2: Furthermore, the present invention provides a carbide tool wear monitoring system based on grating projection image processing, employing a carbide tool wear monitoring method based on grating projection image processing as described in the above embodiments, which can solve the technical problem of carbide tool wear monitoring based on grating projection image processing. The beneficial effects of the carbide tool wear monitoring system based on grating projection image processing provided by the present invention are the same as those of the carbide tool wear monitoring method based on grating projection image processing provided in the above embodiments, and other technical features of the carbide tool wear monitoring system based on grating projection image processing are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0084] Example 3: This invention provides a cemented carbide tool wear monitoring device based on grating projection image processing. Please refer to... Figure 9A carbide tool wear monitoring device based on raster projection image processing includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the carbide tool wear monitoring method based on raster projection image processing described in Embodiment 1 above. The carbide tool wear monitoring device based on raster projection image processing in this embodiment of the invention may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. This carbide tool wear monitoring device based on raster projection image processing is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the invention. A carbide tool wear monitoring device based on raster projection image processing may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the carbide tool wear monitoring device based on raster projection image processing. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An I / O interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. The communication device 1009 allows a carbide tool wear monitoring device based on raster projection image processing to wirelessly or wiredly communicate with other devices to exchange data. While the figures show a carbide tool wear monitoring device based on raster projection image processing with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0085] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for monitoring the wear of cemented carbide tools based on raster projection image processing. The computer program product provided by this invention can solve the technical problem of monitoring the wear of cemented carbide tools based on raster projection image processing. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as those of the above-described method for monitoring the wear of cemented carbide tools based on raster projection image processing, and will not be repeated here.

[0086] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0087] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0088] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for monitoring wear of cemented carbide cutting tools based on grating projection image processing, characterized in that, The methods include: Step S10: Obtain the tool rotation synchronization information of the cemented carbide cutting tool. Based on the tool rotation synchronization information, use a synchronous stroboscopic acquisition method based on phase-gated consistency verification to acquire the grating projection image and obtain the original grating projection image. ; Step S20: For the original raster projection image An image optimization task is performed using a residual dynamic blur suppression mechanism based on rotation domain constraints, outputting an optimized raster projection image. ; Step S30: Based on the optimized raster projection image A reliable grating phase decoding mechanism is used to perform the tool 3D contour reconstruction task, and outputs a 3D contour data set P. Step S40: Based on the 3D contour data set P, a 3D topography mutation screening mechanism based on rotation period consistency constraints is used to construct wear candidate stable regions, and a set of wear candidate stable regions is output. ; Step S50: Based on the set of candidate stable regions for wear A wear region quantitative evaluation mechanism based on spatial weighted constraints is adopted to perform wear region locking and quantitative wear parameter calculation tasks, and finally output the wear monitoring results of cemented carbide tools.

2. The method for monitoring wear of cemented carbide cutting tools based on grating projection image processing as described in claim 1, characterized in that, In step S10, the tool rotation synchronization information of the cemented carbide tool is obtained. Based on the tool rotation synchronization information, a synchronous stroboscopic acquisition method based on phase-gated consistency verification is used to acquire the grating projection image, thereby obtaining the original grating projection image. The steps specifically include: Step S101: Obtain the tool rotation synchronization information of the carbide tool at time t. The tool rotation synchronization information includes the spindle encoder angle synchronization signal. With angular velocity Set the target phase-locked angle. Lock-in tolerance window According to the spindle encoder angle synchronization signal Target phase-locked angle Lock-in tolerance window Define phase-locked gate function , ;in, An angle wrap-around operator is used to map the angle difference to the interval (−π,π) to eliminate the phase jump effect caused when the principal axis angle crosses a 2π period. Step S102: In the phase-locked gate function The phase-locked window emits A short pulse flash, in The grating projection sub-exposure image is acquired within a short pulse strobe, and the target phase-locked angle corresponding to the m-th short pulse strobe is obtained. Angle synchronization signal with spindle encoder Instantaneous phase deviation between ; Step S103: When the instantaneous phase deviation If the instantaneous phase deviation does not exceed the preset threshold, the m-th short pulse strobe is determined to have phase consistency. The grating projection sub-exposure images acquired within the short pulse strobe with phase consistency are fused using an accumulation integration method based on phase consistency constraints, and the original grating projection image is output. .

3. The method for monitoring wear of cemented carbide cutting tools based on grating projection image processing as described in claim 1, characterized in that, In step S20, the original raster projection image is... An image optimization task is performed using a residual dynamic blur suppression mechanism based on rotation domain constraints, outputting an optimized raster projection image. The steps specifically include: Step S201: First, using the rotation center point of the carbide tool as the origin of the polar coordinates, project the original raster image... The rotation domain grating projection image is obtained by mapping from the Cartesian coordinate system to the polar coordinate system. ; Step S202: Based on the physical characteristics of the movement of the rotation center point of the carbide tool, including the rotational angular characteristics of the tool rotating around a fixed rotation center; based on the rotational angular characteristics, a one-dimensional deconvolution method is used to project the rotation domain grating image. The residual dynamic blur in the image is constructed as a one-dimensional convolutional rotation domain sharp raster projection image. ; Step S203: Finally, project the one-dimensional convolutional rotation domain sharpened raster image. The polar coordinate system is inversely transformed to the Cartesian coordinate system to output an optimized raster projection image. .

4. The method for monitoring wear of cemented carbide cutting tools based on grating projection image processing as described in claim 1, characterized in that, In step S30, based on the optimized raster projection image The steps for performing tool 3D contour reconstruction using a grating phase reliable decoding mechanism and outputting a 3D contour data set P specifically include: Step S301: Obtain the optimized raster projection image The projected fringe data is used to estimate parameters by local least squares fitting, outputting the gray-level distribution relationship of the grating fringes. Based on the gray-level distribution relationship, an arctangent phase solver is used to calculate and optimize the grating projection image. The initial phase value corresponding to the pixel with x-coordinate and y-coordinate. ; Step S302: Obtain the position of the pixel with x-coordinate and y-coordinate. The corresponding tool rotation angle is used to filter out outliers based on a threshold-based method, and the initial phase value is then... Group the data and calculate the phase change between adjacent groups. When the phase change amount When the position is less than the preset rotation phase consistency threshold, the position is determined. The phase of the corresponding pixel is the reliable phase; when the phase change amount When the position is greater than or equal to the preset rotation phase consistency threshold, the position is determined. The phase of the corresponding pixel is considered an unreliable phase; for pixels with unreliable phases, phase correction or removal is performed to obtain reliable grating phase values. ; Step S303: Based on the reliable grating phase value Combined with preset system calibration parameters, perform tool 3D contour reconstruction processing and output a 3D contour data set P.

5. The method for monitoring wear of cemented carbide cutting tools based on grating projection image processing as described in claim 4, characterized in that, Step S303, which involves performing tool 3D contour reconstruction based on the reliable grating phase value and preset system calibration parameters, and outputting a 3D contour data set P, specifically includes: Step S3031: Obtain preset system calibration parameters, including the projected baseline distance. And the geometric relationship parameters of projection imaging; based on the system calibration parameters, the reliable grating phase value is converted into the tool surface height value using the triangulation principle. ; Step S3032: Based on the geometric relationship parameters of projection imaging and the tool surface height value The projective inversion mapping method is used to perform the mapping process from two-dimensional pixel coordinates to three-dimensional spatial coordinates, and outputs a reliable phase point in three-dimensional space. Step S3033: Represent the credible phase points in three-dimensional space as a set, and construct and output the three-dimensional contour data set P.

6. The method for monitoring wear of cemented carbide cutting tools based on grating projection image processing as described in claim 4, characterized in that, In step S40, based on the three-dimensional contour data set P, a three-dimensional topography mutation screening mechanism based on rotation period consistency constraints is used to perform the task of constructing wear candidate stable regions, and outputs a set of wear candidate stable regions. The steps specifically include: Step S401: Local Topography Gradient Construction Stage: Based on the 3D contour data set P, a topography extraction method based on local gradient calculation is used to extract the local topography gradient in the 3D height field. ; Step S402: Morphological mutation threshold screening stage: setting morphological mutation threshold Based on morphological abrupt change threshold and local topography gradient Constructing candidate masks , ,in This is an indicator function used to represent regions that meet a threshold condition; for candidate masks Connectivity analysis is performed using a connectivity labeling method based on the eight-neighbor connectivity rule, outputting connected regions, which are then used as the initial set of candidate wear regions. ; Step S403: Stability Verification Stage: A complete tool rotation cycle k is preset. If a candidate mask continuously exists within the complete tool rotation cycle k... At the pixel position with x-coordinate and y-coordinate When the value at a location is 1, determine the position. The pixels at that location are stable pixels; the final set of candidate wear regions is merged. Output a set of candidate stable wear regions for all stable pixels within the range. .

7. The method for monitoring wear of cemented carbide cutting tools based on grating projection image processing as described in claim 1, characterized in that, In step S50, based on the set of wear candidate stable regions The steps involved in performing wear region locking and quantitative wear parameter calculation using a wear region quantification evaluation mechanism based on spatial weighted constraints, and finally outputting the wear monitoring results of carbide tools, specifically include: Step S501: Obtain the set of candidate stable regions for wear For the corresponding contour height data, extract the weighted centroid of the contour height data, and use the weighted centroid as the center position of the wear area. Construct a two-dimensional Gaussian space weighted function with the center position of the wear area as the center. ; ;in, The pixel position with x-coordinate and y-coordinate The two-dimensional Gaussian space weighted function at the location; This is a spatial expansion scale parameter for the wear region, used to adjust the decay rate of the contribution weight of the center and edge pixels in the wear amount calculation; The x-coordinate of the pixel at the center of the wear area; The ordinate of the pixel at the center of the wear area; Step S502: Weighting function based on two-dimensional Gaussian space Set of candidate stable regions for wear The contour height data within the field is spatially weighted and locked to suppress abnormal height values ​​caused by rotation reconstruction errors or residual phase perturbations, and the spatially weighted contour height field is output. Step S503: Calculate the wear parameters based on the weighted height difference statistical analysis method of neighborhood height difference according to the spatial weighted profile height field. The wear parameters include at least the weighted maximum wear depth, weighted wear area and weighted wear volume. Finally, output the wear monitoring results of carbide tools according to the wear parameters.

8. A cemented carbide tool wear monitoring system based on grating projection image processing, applied to the cemented carbide tool wear monitoring method based on grating projection image processing as described in any one of claims 1 to 7, characterized in that, The carbide tool wear monitoring system based on grating projection image processing includes: The tool rotation synchronization and stroboscopic acquisition module is used to acquire the tool rotation synchronization information of cemented carbide tools. Based on the tool rotation synchronization information, a synchronous stroboscopic acquisition method based on phase-gated consistency verification is used to acquire grating projection images, resulting in the original grating projection image. ; Rotation domain residual dynamic blur suppression module, used for original raster projection images An image optimization task is performed using a residual dynamic blur suppression mechanism based on rotation domain constraints, outputting an optimized raster projection image. ; The grating phase reliable decoding and 3D reconstruction module is used for optimizing grating projection images. A reliable grating phase decoding mechanism is used to perform the tool 3D contour reconstruction task, and outputs a 3D contour data set P. The morphological mutation screening module is used to perform the task of constructing wear candidate stable regions based on a 3D contour data set P using a 3D morphological mutation screening mechanism based on rotation period consistency constraints, and outputs a set of wear candidate stable regions. ; The spatially weighted wear quantization assessment module is used to evaluate wear based on the set of candidate stable regions. A wear region quantitative evaluation mechanism based on spatial weighted constraints is adopted to perform wear region locking and quantitative wear parameter calculation tasks, and finally output the wear monitoring results of cemented carbide tools.

9. A cemented carbide cutting tool wear monitoring device based on grating projection image processing, characterized in that, The carbide tool wear monitoring device based on raster projection image processing includes: a memory, a processor, and a carbide tool wear monitoring program based on raster projection image processing stored in the memory and executable on the processor. When the carbide tool wear monitoring program based on raster projection image processing is executed by the processor, it implements a carbide tool wear monitoring method based on raster projection image processing according to any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a carbide tool wear monitoring program based on raster projection image processing. When the carbide tool wear monitoring program based on raster projection image processing is executed by the processor, it implements a carbide tool wear monitoring method based on raster projection image processing according to any one of claims 1 to 7.

Images