CN-121298734-B - Automobile mold data acquisition method and system

Abstract

The application provides an automobile die data acquisition method and system, which comprise the steps of acquiring normal direction information of each micro area on the surface of a cavity of an automobile part die to be detected, controlling a narrow-band light source with a specific wavelength based on the acquired normal direction information, carrying out time-sequence illumination on any micro area on the surface of the cavity by a plurality of preset azimuth angles, synchronously acquiring a frame of image during each time-sequence illumination, thus obtaining an image sequence corresponding to the micro area under the plurality of preset azimuth angles, carrying out time-sequence comparison analysis on brightness values of pixels corresponding to the same physical position in the micro area in the obtained image sequence, identifying a pixel point with the brightness value which changes in a preset manner along with the plurality of preset azimuth angles, determining the identified pixel point as a coating defect, effectively identifying the coating defect with the micrometer scale, overcoming the problems of uneven illumination and background interference caused by a complex curved surface, and improving the detection precision and reliability.

Inventors

• ZENG QINGFENG
• XIAO QIU
• ZHANG CHANGYI

Assignees

• 昆山市海雅特汽车科技有限公司

Dates

Publication Date

20260508

Application Date

20251014

Claims (6)

1. 1. The data acquisition method for the automobile die is characterized by comprising the following steps of: Acquiring normal direction information of each micro area on the cavity surface of the automobile part mold to be detected; Based on the obtained normal direction information, controlling a narrow-band light source with a specific wavelength, and carrying out time-sharing illumination on any micro area on the surface of the cavity in a plurality of preset azimuth angles, wherein during illumination of each azimuth angle, light beams of the narrow-band light source are incident at a preset glancing angle relative to a surface tangential plane of the micro area; synchronously acquiring a frame of image during each time-sequential illumination, thereby obtaining an image sequence corresponding to the micro-area under the plurality of preset azimuth angles; Performing time sequence comparison analysis on brightness values of pixel points corresponding to the same physical position in the micro area in the obtained image sequence to identify the pixel points with the brightness values changed in a preset manner along with the plurality of preset azimuth angles, and determining the identified pixel points as coating defects; the step of performing time sequence comparison analysis on the obtained brightness values of the pixels corresponding to the same physical position in the micro area in the image sequence to identify the pixels with brightness values changed in a preset manner along with the preset azimuth angles comprises the following steps: Acquiring brightness values of the pixel points under the preset azimuth angles to form a brightness response sequence; Calculating a first morphological feature parameter of the brightness response sequence based on the brightness response sequence, wherein the first morphological feature parameter is used for representing the significance degree of brightness value change in the brightness response sequence; Calculating a second shape characteristic parameter of the brightness response sequence based on the brightness response sequence, wherein the second shape characteristic parameter is used for representing the distribution concentration degree of brightness value change among the plurality of preset azimuth angles in the brightness response sequence; When the first morphological characteristic parameter exceeds a first preset threshold value and the second morphological characteristic parameter exceeds a second preset threshold value, determining the pixel point as a coating defect, thereby identifying the pixel point with the preset change of the brightness value along with the plurality of preset azimuth angles; The step of calculating a second morphological feature parameter of the luminance response sequence includes: Analyzing the acquired brightness response sequence to identify a set of peak azimuth angles at which brightness value changes exhibit local peaks in the brightness response sequence; based on the identified group of peak azimuth angles, counting the number of the peak azimuth angles, and analyzing the distribution mode of the peak azimuth angles in the preset azimuth angles; When the number of peak azimuth angles is smaller than a first number threshold and the distribution mode of the peak azimuth angles is characterized by single aggregation, determining the second form characteristic parameter as a first reference value, wherein the first reference value represents high distribution concentration degree; when the number of peak azimuth angles is greater than or equal to the first number threshold and less than a second number threshold, and the distribution pattern of peak azimuth angles is characterized by a plurality of separate aggregation areas, determining the second state characteristic parameter as a second reference value, wherein the distribution concentration degree characterized by the second reference value is lower than the distribution concentration degree characterized by the first reference value, and the second reference value is set to be higher than the expected range of the second state characteristic parameter corresponding to the brightness response sequence generated by the transparent residual film; The step of calculating a first morphological feature parameter of the luminance response sequence based on the luminance response sequence comprises: Calculating a statistical reference quantity of the brightness response sequence, wherein the statistical reference quantity is used for quantifying the inherent optical background response intensity of the coating of the micro-area corresponding to the brightness response sequence; According to a preset functional relation between each brightness value in the brightness response sequence and the calculated statistical reference quantity, carrying out normalization operation on each brightness value in the brightness response sequence to generate a group of normalized brightness values, wherein the normalized brightness values are used for compensating local differences of inherent optical background response intensities of the coating; Calculating a statistical index representing the internal numerical fluctuation amplitude of the group of normalized brightness value sequences based on the generated group of normalized brightness values, and determining the statistical index as the first morphological feature parameter; The step of calculating a statistical index representing the magnitude of the numerical fluctuation within the set of normalized luminance value sequences based on the generated set of normalized luminance values, and determining the statistical index as the first morphological feature parameter includes: Calculating a variance of the normalized luminance value sequence based on the generated set of normalized luminance values; and taking the calculated variance as the statistical index, and determining the statistical index as the first morphological characteristic parameter.
2. 2. The method of claim 1, wherein the step of calculating the statistical reference of the sequence of luminance responses comprises: Calculating, for each luminance value in the sequence of luminance responses, an index that characterizes the degree of dispersion of that luminance value relative to the other luminance values in the sequence of luminance responses; identifying whether an isolated luminance value exists in the luminance response sequence according to the calculated index of the discrete degree of each luminance value; If the isolated brightness value is identified to exist, eliminating the isolated brightness value from the brightness response sequence to form a brightness sequence, and calculating a statistical reference based on the brightness sequence, wherein the statistical reference is used for quantifying the inherent optical background response intensity of the coating of the micro-area corresponding to the brightness response sequence; if the isolated brightness value is not identified, calculating the statistical reference quantity based on the brightness response sequence, wherein the statistical reference quantity is used for quantifying the inherent optical background response intensity of the coating of the micro-area corresponding to the brightness response sequence.
3. 3. The method for collecting automotive mold data according to claim 1, wherein the step of normalizing each luminance value in the luminance response sequence according to a preset functional relationship between each luminance value in the luminance response sequence and the calculated statistical reference amount comprises: Determining the preset function relation as a function model, wherein the function model represents the association relation among each brightness value in the brightness response sequence, the calculated statistical reference quantity and the surface state change; Based on the determined function model, and by using the calculated statistical reference quantity, processing each brightness value in the brightness response sequence to obtain a group of normalized brightness values, wherein the normalized brightness values weaken the influence of the inherent optical background response intensity of the coating, and the normalized brightness values represent the surface state change.
4. 4. The method for collecting automotive mold data according to claim 1, wherein the step of determining the pixel as a coating defect when the first morphological feature parameter exceeds a first predetermined threshold and the second morphological feature parameter exceeds a second predetermined threshold comprises: Acquiring the first morphological characteristic parameters and the second morphological characteristic parameters of a plurality of pixel points in a micro area where the pixel points are located; determining a first preset threshold corresponding to the micro area based on the first morphological feature parameters of the acquired multiple pixel points; Determining a second preset threshold corresponding to the micro area based on the acquired second form characteristic parameters of the pixel points; and when the first morphological characteristic parameter of the pixel point exceeds the first preset threshold value and the second morphological characteristic parameter of the pixel point exceeds the second preset threshold value, determining the pixel point as a coating defect.
5. 5. A method of collecting automotive die data as described in claim 3 wherein said step of analyzing a distribution pattern of said peak azimuth angle among said plurality of preset azimuth angles comprises: Calculating an angular variance or an angular standard difference of the set of peak azimuth angles based on the identified set of peak azimuth angles; And determining the calculated angular variance or the angular standard difference as a distribution mode of the peak azimuth in the plurality of preset azimuths.
6. 6. A data acquisition system for implementing only the method for acquiring automotive mold data according to any one of claims 1 to 5, characterized in that it comprises: The normal information acquisition module is used for acquiring normal direction information of each micro area on the cavity surface of the automobile part die to be detected; The illumination control module is used for controlling a narrow-band light source with a specific wavelength based on the obtained normal direction information, and carrying out time-sharing sequential illumination on any micro area on the surface of the cavity in a plurality of preset azimuth angles, wherein during illumination of each azimuth angle, light beams of the narrow-band light source are incident at a preset glancing angle relative to a surface tangential plane of the micro area; the image acquisition module is used for synchronously acquiring one frame of image during each time of time-sharing illumination so as to obtain an image sequence corresponding to the micro area under a plurality of preset azimuth angles; The defect identification module is used for carrying out time sequence comparison analysis on the brightness information of the pixel points corresponding to the same physical position in the micro area in the obtained image sequence so as to identify the pixel points with the brightness value changed in a preset manner along with the plurality of preset azimuth angles, and determining the identified pixel points as coating defects.

Description

Automobile mold data acquisition method and system Technical Field The application relates to the field of machine vision detection, in particular to the field of mold surface defect detection, and specifically relates to an automobile mold data acquisition method and an automobile mold data acquisition system. Background In the field of precision manufacturing of automobile parts, in particular to parts with higher requirements on appearance, such as transparent lamp covers of automobile lamps, transparent protective cover plates of instrument panels, or plastic parts with fine textures in inner decorations, the surface quality of a cavity of a corresponding injection mold or a die-casting mold directly determines the quality of a final product, and in order to further improve the demolding smoothness of the high-finish molds in the injection molding or die-casting process, a special thin coating is often prepared on a key working surface of a mold cavity in production. In the periodic maintenance or quality sampling inspection link of the die, it becomes important to accurately detect the surface state of the coating die. Conventional machine vision inspection systems, in the face of such micron-scale coating surface defects, pose significant challenges in terms of accuracy and reliability of data acquisition. Under normal illumination conditions, the contrast ratio generated on the image may be very low, and the image is almost submerged in extremely fine processing marks or random ambient light spot noise inherent to the surface of the mold, which may exist even after mirror polishing, and is extremely prone to missed inspection. Further, the cavity of such precision molds often is not a simple plane, but rather contains a large number of complex three-dimensional free-form surfaces, such as headlight cover molds, which are internally designed with numerous micro-optical prism structures and curved transitions of varying radii of curvature in order to achieve a particular light distribution effect. Some areas may be strongly overexposed due to near specular reflection, masking all details, while others in beveled or recessed areas may appear dull and blurry due to the difficulty of light to reach or shadow effectively. The uneven illumination and the focus drift caused by the complex curved surface make it extremely difficult to stably and reliably collect high-quality image data capable of reflecting the real state of the defect of the micro coating on the whole working surface of the die. Even if an image is barely acquired, the subsequent defect recognition and size quantization are greatly impaired by uneven image quality. Disclosure of Invention The technical aim of the application is to provide a method and a system for acquiring data of an automobile die, which have the advantages of effectively identifying the coating defect of a micrometer scale, overcoming the problems of uneven illumination and background interference caused by a complex curved surface and improving the detection precision and reliability. The core technical scheme of the application is an automobile die data acquisition method, which specifically comprises the following steps: Acquiring normal direction information of each micro area on the cavity surface of the automobile part mold to be detected; Based on the obtained normal direction information, controlling a narrow-band light source with a specific wavelength, and carrying out time-sharing illumination on any micro area on the surface of the cavity by a plurality of preset azimuth angles, wherein during illumination of each azimuth angle, light beams of the narrow-band light source are incident at a preset glancing angle relative to the surface tangential plane of the micro area; Synchronously acquiring one frame of image during each time-sequential illumination, so as to obtain an image sequence corresponding to a micro area under a plurality of preset azimuth angles; And carrying out time sequence comparison analysis on the brightness values of the pixels corresponding to the same physical position in the micro area in the obtained image sequence so as to identify the pixels with preset changes of the brightness values along with a plurality of preset azimuth angles, and determining the identified pixels as coating defects. Through the scheme, the coating defects of the micrometer scale can be effectively identified, the problems of uneven illumination and background interference caused by complex curved surfaces are overcome, and the detection precision and reliability are improved. The application also provides an automobile die data acquisition system, which specifically comprises: The normal information acquisition module is used for acquiring normal direction information of each micro area on the cavity surface of the automobile part die to be detected; the illumination control module is used for controlling a narrow-band light source with a specific wavelength b