CN-121981975-A - Cutter rear tool face abrasion quantification method based on preset reference mark

CN121981975ACN 121981975 ACN121981975 ACN 121981975ACN-121981975-A

Abstract

The invention relates to the technical field of cutter state detection, and discloses a cutter rear cutter face abrasion quantification method based on a preset reference mark, wherein the reference mark is preset in a non-cutting area of the cutter rear cutter face which does not participate in cutting, and the relative spatial relation between the reference mark and the original contour of the cutting edge of the cutter is fixed; the method comprises the steps of obtaining a calibration image containing a reference mark and an original contour of a cutting edge, obtaining an image to be detected containing the reference mark and a rear cutter face abrasion zone after cutting and abrasion of a cutter, carrying out image registration processing on the image to be detected and the calibration image by taking reference mark pixel level alignment as a benchmark, extracting the original contour of the cutting edge and the contour of the rear cutter face abrasion zone from the registered image, and calculating a quantization index of the rear cutter face abrasion degree through a preset algorithm. According to the invention, the stable and external reference mark is introduced into the non-cutting area of the rear cutter surface of the cutter to anchor the initial state information of the cutter, and the abrasion is quantified after pixel-level image registration, so that the efficient, high-precision and automatic friendly cutter abrasion detection is realized.

Inventors

MA JIN
WEN LIDONG
JIA HAODONG
CHEN JIAJUN

Assignees

成都工具研究所有限公司

Dates

Publication Date: 20260505
Application Date: 20260107

Claims (10)

1. The method for quantifying the wear of the tool rear tool face based on the preset reference mark is characterized by comprising the following steps: s1, presetting a reference mark in a non-cutting area of a rear tool face of a tool which does not participate in cutting, wherein the relative spatial relation between the reference mark and the original contour of the cutting edge of the tool is fixed; s2, acquiring a calibration image containing a reference mark and an original contour of the cutting edge; S3, after the cutter is cut and worn, obtaining an image to be detected, which comprises a reference mark and a rear cutter face wear belt; s4, performing image registration processing on the image to be detected and the calibration image by taking reference mark pixel level alignment as a benchmark; S5, extracting an original contour of the cutting edge and a contour of a rear cutter face abrasion zone from the registered image, and calculating a quantization index of the rear cutter face abrasion degree through a preset algorithm.
2. The method for quantifying tool flank wear based on preset reference marks according to claim 1, wherein in S1, corresponding reference marks are respectively set for one or more cutting edge flank of a corresponding tool to quantify corresponding cutting edges.
3. The method according to claim 1, wherein in S1, the reference mark has a fixed and identifiable geometry and an optical characteristic different from the tool body, and wherein the geometry is a regular pattern including at least one characteristic element.
4. A method of quantifying tool flank wear based on preset reference marks as recited in claim 3 wherein the feature elements comprise one of point feature elements, line feature elements, or a combination thereof.
5. The method for quantifying tool flank wear based on preset reference marks is characterized in that in S2, initial information representing the reference marks, original independent features of the cutting edges and relative features in the calibration image is obtained, wherein the reference marks and the original independent features of the cutting edges comprise the shapes and positions of the reference marks and the cutting edges, and the relative features comprise the relative spatial relations of the reference marks and the cutting edges.
6. The method for quantifying tool flank wear based on preset reference marks of claim 1, wherein S4 comprises: s41, acquiring pixel coordinates of N preset points on the reference marks of the calibration image and the image to be detected; S42, solving a transformation matrix based on a one-to-one correspondence established by two groups of preset points; S43, the transformation matrix is applied to the image to be detected, so that the transformation matrix is aligned with the calibration image at the pixel level at the reference mark, and the current edge contour and the flank wear strip contour in the image to be detected are restored to the initial reference position under the coordinate system of the calibration image.
7. The method for quantifying tool flank wear based on preset reference marks according to claim 6, wherein in S41, when the reference mark adopts a marking method of a point feature, a center coordinate of a mark point is extracted, or when the reference mark adopts a marking method of a line feature, a coordinate at which an intersection point between adjacent straight lines is located is extracted.
8. The method according to claim 1, wherein in S5, the preset algorithm calculates a pixel level quantization index based on a pixel distance between contour points, and determines an image scale with reference to a reference mark or a standard reference plate to convert the pixel level index into a final quantization index having an actual physical size.
9. The method for quantifying tool flank wear based on preset reference marks according to any one of claims 1 or 8, wherein in S5, the quantification index comprises at least one of an average width, a maximum width, and a wear area of the wear strip.
10. The method for quantifying tool relief wear based on preset reference marks according to claim 1, wherein when the tool is a rotary tool, at least one of the following adjustment modes is adopted: In the first adjustment mode, S1, a plurality of groups of reference marks are arranged in an interested area of the rear cutter face of a cutter, and subsequent processing is carried out by using each group of reference marks; In the second adjustment mode, in the period that the cutter rotates within a certain angle range, multiple frames of images containing the reference mark and the original contour of the cutting edge are collected, and the image with the largest imaging definition of the reference mark is selected as a calibration image to be subjected to subsequent processing.

Description

Cutter rear tool face abrasion quantification method based on preset reference mark Technical Field The invention relates to the technical field of cutter state detection, in particular to a cutter rear cutter face abrasion quantification method based on preset reference marks. Background In the metal cutting process, the abrasion state of the cutter directly influences the processing quality, the processing efficiency and the production cost. The tool rear surface abrasion (FLANK WEAR) is one of key indexes for measuring the use state of the tool, and is also the most common tool abrasion form. The abrasion loss of the rear cutter surface of the cutter is timely and accurately obtained, and the method has important significance in judging the service life of the cutter, optimizing the processing parameters and preventing processing failure. Conventional tool wear detection methods mainly include manual visual inspection, contact measurement, and image-based detection methods. Among them, image-based detection methods are becoming the mainstream of research and application due to their non-contact and easy integration characteristics. However, the following problems still exist in the practical application of the prior art: 1. most methods rely on cutting edges or blade edges as measurement references, but in the machining process, the cutting edges are easily affected by factors such as chip adhesion, abrasion form change and the like, so that the identified references are not real references or are difficult to identify, and the accuracy and the stability of measurement are affected. 2. The cutting edge position is not accurately identified, namely, the cutting edge position is gradually deviated relative to an initial state, which is difficult to accurately predict, and the cutting edge can adhere to workpiece materials or collapse and lack, so that the measuring standard based on the cutting edge is unreliable. 3. The detection efficiency is low, the traditional method often needs to accurately control the clamping position and the shooting angle of the cutter, has high requirements on the detection environment, and is difficult to realize rapid batch detection. 4. The reference object is difficult to select, a part of the cutter structure lacks stable and obvious reference characteristics, or the application of the reference characteristics is difficult, and the automatic identification and the benchmark restoration of the image are difficult to realize efficiently. Therefore, there is a need for a tool wear quantization method that can provide stable reference, not rely on cutting edge references, simplify shooting conditions, and improve detection efficiency and accuracy. Disclosure of Invention The invention aims to provide a cutter rear tool face abrasion quantification method based on a preset reference mark, which utilizes initial state information of an anchor cutter of the preset reference mark, and quantifies abrasion after pixel-level image registration, so that efficient, high-precision and automatic friendly cutter abrasion detection is realized. The basic scheme provided by the invention is that the cutter rear surface abrasion quantization method based on the preset reference mark comprises the following steps: s1, presetting a reference mark in a non-cutting area of a rear tool face of a tool which does not participate in cutting, wherein the relative spatial relation between the reference mark and the original contour of the cutting edge of the tool is fixed; s2, acquiring a calibration image containing a reference mark and an original contour of the cutting edge; S3, after the cutter is cut and worn, obtaining an image to be detected, which comprises a reference mark and a rear cutter face wear belt; s4, performing image registration processing on the image to be detected and the calibration image by taking reference mark pixel level alignment as a benchmark; S5, extracting an original contour of the cutting edge and a contour of a rear cutter face abrasion zone from the registered image, and calculating a quantization index of the rear cutter face abrasion degree through a preset algorithm. The working principle and the advantages of the invention are as follows: The invention successfully constructs an absolute space coordinate system throughout detection by deeply analyzing the cutting wear characteristic of the cutter, finding a stable non-cutting area on a non-stable cutting surface of the cutter, presetting a special reference mark which is completely decoupled with the wear process, has fixed relative space relation with the original contour of the cutting edge of the cutter, fixed geometric form and identifiable and high contrast optical characteristic with the cutter in the stable area, and fundamentally changing the bottom logic of the traditional measurement by relying on the variable characteristic of the cutter. Based on stable preset reference marks, the initial state of