CN-121612870-B - Graphite morphology and spheroidization analysis system and method based on image processing

CN121612870BCN 121612870 BCN121612870 BCN 121612870BCN-121612870-B

Abstract

The invention relates to a graphite morphology and spheroidization analysis system and method based on image processing, and belongs to the technical field of graphite morphology analysis. The method comprises the steps of S1, obtaining a two-dimensional metallographic image of spheroidal graphite cast iron to be detected, S2, processing the two-dimensional metallographic image to divide graphite particles in the image, S3, extracting and calculating a plurality of shape factors of each graphite particle, wherein the shape factors at least comprise roundness, convexity and sharpness, S4, comprehensively judging whether each graphite particle is qualified spheroidal graphite or not based on a preset judging threshold, S5, calculating the spheroidization rate of the spheroidal graphite cast iron to be detected based on the area of the qualified spheroidal graphite, and S6, outputting an analysis report containing the spheroidization rate. The invention solves the technical problems that the existing spheroidization rate analysis method depends on single roundness and is easy to misjudge high-risk graphite.

Inventors

JIN CHENGGONG
ZHANG LIMING
ZHAO JUNYING
LI JIANFENG
LIU ZHIFENG
ZHANG XIAOYU
ZHANG JINGYU
LI ZHANTAO

Assignees

禹州市恒利来新材料股份有限公司

Dates

Publication Date: 20260512
Application Date: 20251202

Claims (5)

1. The graphite morphology and spheroidization analysis method based on image processing is characterized by comprising the following steps of: S1, acquiring a two-dimensional metallographic image of spheroidal graphite cast iron to be detected; S2, processing the two-dimensional metallographic image to separate each graphite particle in the image; s3, extracting and calculating a plurality of shape factors of each graphite particle, wherein the shape factors at least comprise roundness, convexity and sharpness, and the shape factors comprise the following steps: The roundness is obtained by calculating the ratio of the area of graphite particles to the maximum Buddha Reid circle area; The convexity is obtained by calculating the ratio of the area of graphite particles to the area of the convex hull of the graphite particles; The sharpness is obtained by analyzing and calculating the curvature of the contour line of the graphite particles; S4, comprehensively judging whether each graphite particle is qualified spherical graphite based on a preset judging threshold value, wherein the method comprises the following steps: S41, acquiring a dynamic judgment threshold corresponding to the specification of the spheroidal graphite cast iron to be tested, wherein the dynamic judgment threshold at least comprises a roundness threshold, a convexity threshold and a sharpness threshold; S42, judging that the graphite particles are qualified spherical graphite when the roundness of the graphite particles is larger than or equal to the threshold value of the roundness, the convexity of the graphite particles is larger than or equal to the threshold value of the convexity, and the sharpness of the graphite particles is smaller than or equal to the threshold value of the sharpness; s5, calculating the spheroidization rate of the spheroidal graphite cast iron to be tested based on the area of the qualified spheroidal graphite, wherein the spheroidization rate comprises the following steps: S51, distributing a weight coefficient for each spherical graphite particle judged to be qualified; The distribution of the weight coefficient is dynamically adjusted according to the values of the roundness and convexity of the qualified spherical graphite particles; The dynamic adjustment logic is that a weight basic value is set for the weight coefficient, and floating is carried out on the weight basic value based on the values of the roundness and convexity; S511, obtaining a weight basic value and a floating rule corresponding to the specification of the spheroidal graphite cast iron to be tested; the weight basic value and the floating rule are obtained by inquiring a spheroidal graphite cast iron specification database; the floating rule is that a numerical value interval and a corresponding weight correction factor are respectively set for the roundness and convexity; S512, calculating a final weight coefficient of the current graphite particles based on the roundness and convexity of the current graphite particles according to the floating rule; S5121, judging a first numerical value interval to which the roundness of the current graphite particles belongs, and acquiring a corresponding first weight correction factor; S5122, judging a second numerical value interval to which convexity of the current graphite particles belongs, and acquiring a corresponding second weight correction factor; s5123, calculating and correcting the weight basic value through the first weight correction factor and the second weight correction factor respectively to obtain the final weight coefficient of the graphite particles, S52, calculating spheroidization rate based on a weighted area method, wherein the calculation formula is spheroidization rate= (sum of weighted areas of all qualified spherical graphite particles/total area of all effective graphite particles participating in calculation) ×100%; Wherein the weighted area of a single qualified spherical graphite particle is the product of its actual area and its weight coefficient; s6, outputting an analysis report containing the spheroidization rate.
2. The method according to claim 1, characterized in that in said step S41: The dynamic judgment threshold value is obtained by inquiring a pre-established spheroidal graphite cast iron specification database; the database stores a plurality of spheroidal graphite cast iron specification codes, each specification code is associated with a threshold value set, and the threshold value set at least comprises a roundness threshold value, a convexity threshold value and a sharpness threshold value; The step S41 specifically includes: s411, acquiring a specification code number of the spheroidal graphite cast iron to be tested; And S412, retrieving and calling the corresponding threshold value set from the spheroidal graphite cast iron specification database by taking the specification code number as an index.
3. The method according to claim 1, wherein the step S2 comprises: S21, preprocessing the two-dimensional metallographic image, wherein the preprocessing at least comprises graying, contrast enhancement and noise filtering; S22, carrying out binarization segmentation on the preprocessed image, and separating graphite particles from a matrix; s23, carrying out morphological treatment on the binarized image, and separating adhered graphite particles by adopting a watershed algorithm.
4. A graphite morphology and spheroidization analysis system based on image processing for implementing the method of any one of claims 1 to 3, comprising: the image acquisition module is used for acquiring a two-dimensional metallographic image of the spheroidal graphite cast iron to be detected; The image processing module is used for processing the image to divide each graphite particle; The feature analysis module is used for extracting a plurality of shape factors of each graphite particle and judging qualification based on the dynamic judgment threshold value; The spheroidization rate calculation module is used for calculating spheroidization rate based on a weighted area method; and the report generation module is used for generating and outputting an analysis report.
5. The system of claim 4, further comprising: The spheroidal graphite cast iron specification database is used for storing the dynamic judging threshold values of the shape factors and the weight distribution rules corresponding to different spheroidal graphite cast iron specifications; the characteristic analysis module and the spheroidization rate calculation module are configured to call parameters corresponding to the current spheroidal graphite cast iron specification to be measured from the specification database to calculate and judge.

Description

Graphite morphology and spheroidization analysis system and method based on image processing Technical Field The invention relates to a graphite morphology and spheroidization analysis system and method based on image processing, and belongs to the technical field of graphite morphology analysis. Background The mechanical properties of spheroidal graphite cast iron, such as strength, toughness and fatigue life, are largely determined not by the chemical composition but by the microstructure within. Among them, the morphology of graphite is a critical influencing factor. In an ideal state, graphite should exist in a perfect sphere shape, and at the moment, the cracking effect of the graphite on a ferrite matrix is minimum, and the stress concentration degree is minimum, so that the excellent comprehensive mechanical property of the material is provided. In contrast, when graphite exists in a non-spherical form such as a flake form, a worm form, or the like, serious stress concentration is liable to be induced, and toughness, strength, and fatigue properties of the material are drastically deteriorated. Therefore, the graphite morphology is accurately detected and evaluated, is a core link for controlling and improving the quality of spheroidal graphite cast iron products, and has great significance for ensuring the safety and reliability of key equipment components such as wind power hubs, high-speed rail gearboxes, heavy-duty engine bodies and the like. At present, the detection method of the spheroidization rate of the graphite morphology of the spheroidal graphite cast iron is mainly divided into an indirect method and a direct method. The indirect method comprises an ultrasonic detection method, a knocking method and the like, the nodulizing rate is indirectly calculated by measuring the sound velocity or the vibration frequency of the material by the ultrasonic detection method and the knocking method, the rapid and nondestructive detection can be realized, but the precision is greatly influenced by the material components, the tissue structure and the workpiece shape, and the reliability is insufficient. The direct rule is based on the metallographic examination method, i.e. the evaluation is carried out by preparing a sample and observing its two-dimensional cross section under a microscope. In the metallographic examination, the traditional visual comparison method relies on experience of an inspector, and is strong in subjectivity and poor in repeatability. To overcome the subjectivity of manual assessment, image analysis techniques were introduced. Through searching, a method for quantitatively calculating spheroidization rate of spheroidal graphite cast iron based on a shape factor is disclosed in the publication No. CN106127821A, wherein the shape factor is adopted as a characteristic parameter for distinguishing spherical graphite, and the spheroidization rate is calculated based on a counting method. This approach takes an important step in quantification but is still essentially a simplified geometric approximation. However, with the increasing requirements for detection accuracy, the drawbacks of the prior art solutions are increasingly pronounced. First, the counting method represented by CN106127821a gives each graphite particle an equal weight regardless of size, which is not in agreement with the objective fact that large-size graphite has a greater influence on mechanical properties, and may lead to a disjoint assessment result from the actual properties of the material. For this reason, the latest national standard GB/T9441-2021 has turned to the area method, i.e. the spheroidization rate is defined as the percentage of the area of the acceptable spheroidal graphite to the total graphite area. The area method reflects the contribution difference of graphite with different sizes more scientifically and is an important progress in the assessment method. The introduction of the area method solves the problem of unreasonable particle weight in the counting method and represents a more advanced calculation concept. However, the method does not solve the defects of the criterion of the spherical graphite, namely whether the method is based on a counting method or an area method, the underlying logic of the method is seriously dependent on a single 'roundness' parameter to judge whether the graphite is 'qualified'. This decision logic has the fundamental technical problem that rounding is an integral geometric descriptor that only considers the "amount" of graphite, but severely ignores the effect of its "quality" on the material properties. In the actual metallographic structure, there is graphite in the form of bloom. Such graphites may have large projected areas overall, resulting in spherical graphites that may not necessarily have very low calculated rounding levels, and may even exceed a threshold of 0.6 and be judged to be "acceptable". However, from the point of view of material mechan