CN-122016654-A - Metal surface micro-defect polarization detection method based on Gram determinant

CN122016654ACN 122016654 ACN122016654 ACN 122016654ACN-122016654-A

Abstract

The application relates to the technical field of metal surface defect detection and discloses a Gram determinant-based metal surface microdefect polarization detection method which comprises the steps of obtaining reflected light intensity data of each pixel point on the surface of a finely bored tolerance hole to be detected under illumination of a plurality of different polarization states, calculating a defect matrix of each pixel point according to the reflected light intensity data, taking a row vector of the defect matrix corresponding to each pixel point as a base vector, calculating a determinant value of the Gram matrix of the base vector as a Gram determinant value of a current pixel point, generating a Gram distribution map according to the Gram determinant values of all the pixel points, carrying out image processing on the Gram distribution map, identifying and marking a microdefect area on the finely bored tolerance hole surface, constructing a defect matrix, calculating the Gram determinant, converting polarization information change of the microdefect into a scalar index, enhancing contrast ratio of the microdefect and normal processing texture, and realizing high-sensitivity detection of the finely bored tolerance hole surface microdefect.

Inventors

LIN BAOQUAN
CAI HUALI
HOU JUNMING

Assignees

南京南特精密机械有限公司

Dates

Publication Date: 20260512
Application Date: 20260413

Claims (10)

1. The utility model provides a metal surface micro-defect polarization detection method based on Gram determinant, the metal surface is the tolerance hole surface of finely boring, and characterized by includes: s1, obtaining reflection light intensity data of each pixel point on the surface of a fine boring tolerance hole to be measured under illumination of a plurality of different polarization states; S2, solving a defect matrix of each pixel point of the fine boring tolerance hole according to the reflected light intensity data; S3, taking four row vectors of a defect matrix corresponding to each pixel point of the fine boring tolerance hole as a group of base vectors, and calculating a determinant value of a Gram matrix of the current group of base vectors to be used as a Gram determinant value of the current pixel point; s4, generating a Gram distribution map according to Gram determinant values of all the pixel points; And S5, carrying out image processing on the Gram distribution map, and identifying and marking the micro defect area on the surface of the fine boring tolerance hole.
2. The Gram determinant-based metal surface micro defect polarization detection method according to claim 1, wherein the method comprises the following steps: carrying out regional positioning on the finish boring tolerance hole according to the clamping position of the cradle of the machine tool; Dividing the fine boring tolerance hole into a plurality of detection layers along the axis direction, controlling a polarization detection probe to sequentially extend into the positions of the detection layers, and performing circumferential scanning on the hole wall of each detection layer; in the circumferential scanning process of each detection layer, illuminating light with a plurality of different polarization states sequentially irradiates each microcell of the hole wall, and the emergent light intensity of at least four different analysis polarization states under each illumination polarization state is collected through a polarization state analyzer to obtain a plurality of groups of reflected light intensity data of each microcell; And carrying out axial attenuation compensation on the collected reflected light intensity according to the distance difference between each detection layer and the light source, and eliminating brightness difference caused by different optical paths.
3. The Gram determinant-based metal surface micro defect polarization detection method according to claim 2, wherein the method comprises the following steps: the defect matrix calculation and Gram determinant calculation in S2 and S3 comprise: Calculating a defect matrix of each hole wall microcell by the reflected light intensity data after axial attenuation compensation, and calculating Gram determinant values of each microcell; generating annular Gram expanded graphs of all the detection layers, wherein the abscissa is a circumferential angle, and the ordinate is a Gram determinant value; and performing directional filtering treatment on the Gram distribution gray scale map to enhance defect characteristics, wherein a filtering core extending along the axial direction is used for enhancing axial scratch, and a filtering core extending along the spiral direction is used for enhancing spiral scratch.
4. A Gram determinant-based metal surface micro defect polarization detection method according to claim 3, wherein: The defect identification and marking in S5 includes: in the Gram distribution gray level graph after directional filtering, identifying a connected region with abnormal Gram values as a candidate defect; extracting geometric features of candidate defects, including axial length, circumferential span and extending direction; The defects are classified according to the geometric characteristics, namely axial scratch is judged, spiral scratch is judged, point-like defects are judged, material burn or layering is judged, wherein the axial scratch is large in axial length, small in circumferential span and close to axial in extending direction, the axial scratch is judged, the axial scratch is large in axial length, the extending direction is consistent with a spiral angle, the point-like defects are judged, the area is small, the circularity is high, the area is large, the shape is irregular, and the Gram abnormal value is high; and outputting classification results, axial depth positions and circumferential angle positions of the defects.
5. The Gram determinant-based metal surface micro defect polarization detection method according to claim 1, wherein the method comprises the following steps: And S2, when the defect matrix corresponding to each pixel point is solved, respectively establishing corresponding incident light instrument matrixes for different detection layers of the fine boring tolerance hole according to the distances between each layer and the light source, and correcting the measurement data.
6. The Gram determinant-based metal surface micro defect polarization detection method according to claim 1, wherein the method comprises the following steps: and the Gram determinant value is used for representing the linear independence degree among four row vectors of the defect matrix, and micro defects on the surface of the fine boring tolerance hole are identified through the abnormality of the Gram determinant value.
7. The Gram determinant-based metal surface micro defect polarization detection method according to claim 2, wherein the method comprises the following steps: Measuring the reflected light intensity of the standard sample block under a plurality of different distances in advance, and establishing an attenuation curve of the light intensity changing along with the distance; During detection, according to the actual distance between the current detection layer and the light source, inquiring the attenuation curve to obtain a compensation coefficient, and multiplying the collected reflected light intensity by the compensation coefficient to eliminate brightness attenuation caused by the optical path difference.
8. A Gram determinant-based metal surface micro defect polarization detection method according to claim 3, wherein: The directional filtering is realized by a Gabor filter bank, and the Gabor filter bank comprises a plurality of two-dimensional Gabor kernels with different directions and scales; Filtering each layer of the Gram distribution gray level graph, and taking the maximum value of response in each direction as a reinforced characteristic graph, wherein Gabor cores corresponding to the axial direction are used for reinforcing axial scratches, gabor cores corresponding to the spiral direction are used for reinforcing spiral scratches, and the angle of the spiral direction is determined according to the feeding helix angle during fine boring tolerance hole processing.
9. The Gram determinant-based metal surface micro defect polarization detection method according to claim 4, wherein the method comprises the following steps: when identifying a connected region with abnormal Gram value as a candidate defect, adopting a self-adaptive threshold segmentation method: calculating a local mean mu and a standard deviation sigma of the Gram distribution gray map, and setting a threshold value T=mu+ksigma, wherein k is a preset sensitivity coefficient; Marking pixels with Gram values larger than T as abnormal pixels; and carrying out connected domain analysis on the abnormal pixels, merging the connected domains with the distance smaller than the preset interval, and removing isolated areas with the area smaller than the minimum defect area to obtain candidate defects.
10. The Gram determinant-based metal surface micro defect polarization detection method according to claim 5, wherein the method comprises the following steps: Establishing depth correction of the incident light instrument matrix includes: Placing a standard reflection template at different depth positions in a fine boring tolerance hole, respectively measuring the response of the standard reflection template under illumination of each polarization state, calculating the deviation between instrument matrix elements corresponding to different depths and the reference depth, and fitting a depth deviation function; and when in detection, correcting a pre-calibrated reference instrument matrix by utilizing the depth deviation function according to the depth position of the current detection layer to obtain an instrument matrix actually used by the current layer for solving the defect matrix.

Description

Metal surface micro-defect polarization detection method based on Gram determinant Technical Field The application relates to the technical field of metal surface defect detection, and discloses a metal surface micro-defect polarization detection method based on Gram determinant. Background The fine boring tolerance hole is an important structure on key parts in the fields of aerospace, precision machinery and the like, and the quality of the inner wall surface directly influences the matching precision, the sealing performance and the service life. In the traditional fine boring process, the cutter is worn, the cutting parameters fluctuate, the cooling is insufficient or the internal defects of the material are caused. The method is characterized in that the surface of the hole wall is easy to generate various microscopic defects such as axial scratch, spiral scratch, point-like defect, material burn or layering and the like, however, in the prior art, the inner wall of the finely bored tolerance hole is detected by adopting a contact coarser instrument or an industrial endoscope, the contact method is low in efficiency and is easy to scratch the surface and difficult to realize full detection, the subjectivity of the manual visual inspection is high, the microscopic defects are extremely easy to miss detection, the polarization optical detection technology can acquire the microstructure information of the surface of the material, the surface roughness change, subsurface damage and material physical property change are unique sensibility, however, the existing polarization detection method generally only analyzes single polarization parameters, the complete polarization information carried by reflected light cannot be fully utilized, and in addition, a specific detection scheme and a data processing model are lacked for the finely bored tolerance Kong Zhelei deep hole structure, and the polarization information is difficult to be effectively converted into the identification index of the microscopic defects. Disclosure of Invention This section is intended to outline some aspects of embodiments of the application and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section as well as in the description of the application and in the title of the application, which may not be used to limit the scope of the application. The specific application aims are as follows: The application aims to solve the technical problems that in the existing precision boring tolerance hole inner wall detection, the contact type coarseness gauge has low detection efficiency, is easy to scratch the surface and is difficult to fully detect, the industrial endoscope manual visual inspection has strong subjectivity, and the tiny defect is easy to miss. In order to solve the technical problems, the application provides a metal surface micro-defect polarization detection method based on Gram determinant. In one aspect, the application provides a Gram determinant-based metal surface micro-defect polarization detection method, which comprises the following steps: the metal surface is a finely bored tolerance hole surface, and is characterized by comprising: s1, obtaining reflection light intensity data of each pixel point on the surface of a fine boring tolerance hole to be measured under illumination of a plurality of different polarization states; S2, solving a defect matrix of each pixel point of the fine boring tolerance hole according to the reflected light intensity data; S3, taking four row vectors of a defect matrix corresponding to each pixel point of the fine boring tolerance hole as a group of base vectors, and calculating a determinant value of a Gram matrix of the current group of base vectors to be used as a Gram determinant value of the current pixel point; s4, generating a Gram distribution map according to Gram determinant values of all the pixel points; And S5, carrying out image processing on the Gram distribution map, and identifying and marking the micro defect area on the surface of the fine boring tolerance hole. As a preferable scheme of the metal surface micro-defect polarization detection method based on Gram determinant, the application comprises the following steps: carrying out regional positioning on the finish boring tolerance hole according to the clamping position of the cradle of the machine tool; Dividing the fine boring tolerance hole into a plurality of detection layers along the axis direction, controlling a polarization detection probe to sequentially extend into the positions of the detection layers, and performing circumferential scanning on the hole wall of each detection layer; in the circumferential scanning process of each detection layer, illuminating light with a plurality of different polarization states sequentially irradiates each microcell of the hole wall, and the emergent light intensity of at least four different analysis polarizati