Search

CN-121997076-A - Method for detecting spraying quality of coating of aircraft component

CN121997076ACN 121997076 ACN121997076 ACN 121997076ACN-121997076-A

Abstract

The invention relates to the field of spray quality detection, in particular to a spray quality detection method for an aircraft component coating, which comprises the steps of determining a category of a concerned point according to a neighborhood vector difference degree and a maximum reference difference degree, determining whether an adaptive division mode is adopted or not based on a second category concerned point proportion and an analysis area proportion, determining a seed point according to a curvature processing value under the adaptive division mode, obtaining an initial selection sequence of the seed point based on the curvature processing value, determining whether sequential adjustment is carried out according to a curvature processing value corresponding to a comparison condition to obtain a final sequence, determining an associated point in a target area corresponding to seed point growth based on a curvature processing difference value and a symbol similarity index, determining whether merging processing is carried out according to an area of the area, and determining the merging processing mode according to a curvature similarity parameter difference value and a spray distance difference index. The invention improves the efficiency of spray quality detection.

Inventors

  • ZHU DAN
  • ZHU YUETING
  • YU YING

Assignees

  • 张家港市晨艺电子有限公司

Dates

Publication Date
20260508
Application Date
20260410

Claims (10)

  1. 1. The method for detecting the spraying quality of the coating of the aircraft component is characterized by comprising the following steps of: Determining the category of the attention point according to the neighborhood vector difference degree and the maximum reference difference degree, and determining whether to adopt an adaptive division mode or not based on the proportion of the attention points of the category and the proportion of the area of the analysis area; Under the adaptive division mode, a first class of attention points and a second class of attention points are respectively screened according to the curvature processing values to obtain a first seed point and a second seed point, and the seed points are arranged according to the order of the curvature processing values from large to small to obtain a primary selection sequence; determining whether to sequentially adjust the initial sequence according to curvature processing values corresponding to the comparison conditions so as to obtain a final sequence; Sequentially carrying out region growth on various seed points in the final sequence to obtain a plurality of target regions comprising the seed points and corresponding association points, wherein the association points are selected based on curvature processing difference values and symbol similarity indexes; Aiming at a target area with the area smaller than the area of a preset area, determining to carry out merging treatment; in the merging process, the states of the adjacent target areas are determined according to the curvature similarity parameter difference value and the spraying distance difference index, and the merging process mode is curvature similarity merging or spraying similarity merging based on the states of the adjacent target areas.
  2. 2. The method for detecting the coating spraying quality of the aircraft component according to claim 1, wherein the region construction is performed on the second class of interest points of the target component model to obtain a model region with the smallest area capable of comprising all the second class of interest points, and the model region is recorded as an analysis region; If the second class attention point occupation ratio is larger than the preset second class attention point occupation ratio or the analysis area occupation ratio is smaller than or equal to the preset analysis area occupation ratio, adopting an adaptive division mode; and if the second class of attention point occupation ratio is smaller than or equal to the preset second class of attention point occupation ratio and the analysis area occupation ratio is larger than the preset analysis area occupation ratio, adopting a reference division mode.
  3. 3. The method for detecting the coating spraying quality of the aircraft component according to claim 2, wherein the attention points with the neighborhood vector difference degree larger than the preset neighborhood vector difference degree or the maximum reference difference degree larger than the preset maximum reference difference degree are marked as one attention point; And aiming at the attention points with the neighborhood vector difference degree smaller than or equal to the preset neighborhood vector difference degree and the maximum reference difference degree smaller than or equal to the preset maximum reference difference degree, marking the attention points as the class II attention points.
  4. 4. A method of detecting the spray quality of a coating of an aircraft component according to claim 3, wherein the adaptive division comprises: Performing filtering processing and normalization processing on curvature reference values of all the attention points to obtain curvature processing values; Marking a class of attention points with curvature processing values larger than a preset highest curvature processing value as first seed points; And marking the second class of interest points with curvature processing values smaller than or equal to the preset minimum curvature processing value as second seed points.
  5. 5. The method for detecting the spraying quality of the coating of the aircraft component according to claim 4, wherein the determination is based on the neighborhood similar reference value to sequentially adjust the primary selection sequence according to the comparison result that the comparison degree of the curvature processing value is larger than the comparison degree of the preset curvature processing value.
  6. 6. The method for inspecting the spray quality of a coating of an aircraft component according to claim 5, wherein the sequential adjustment for the preliminary sequence comprises: sequentially carrying out neighborhood analysis on various sub-points based on the order of the curvature processing values from large to small; The neighborhood analysis for a single seed point comprises detecting the difference value of neighborhood similar reference values between the single seed point and the adjacent seed points positioned behind the sequence, and if the difference value of the neighborhood similar reference values is greater than or equal to 0, the sequence replacement is not needed; if the difference value of the neighborhood similar reference values is smaller than 0, the two seed points are sequentially exchanged.
  7. 7. The method according to claim 6, wherein, when the region growth is performed for a single seed point, a point of interest having a curvature process difference value smaller than a preset curvature process difference value and a sign similarity index larger than a preset sign similarity index corresponding to the seed point is marked as a relevant point, and a region of the target component model including a minimum area of the seed point and the corresponding relevant point is marked as a target region; if the sequence adjustment is carried out, marking the initial sequence after the sequence adjustment is finished as a final sequence; if no sequence adjustment is made, the initially selected sequence is noted as the final sequence.
  8. 8. The method for detecting the coating spraying quality of the aircraft component according to claim 7, wherein the merging processing mode is determined to be curvature similarity merging according to the states of adjacent target areas with curvature similarity parameter difference values smaller than preset curvature similarity parameter difference values and spraying distance difference indexes smaller than preset spraying distance difference indexes, and adjacent areas with minimum curvature similarity parameters are selected to be merged.
  9. 9. The method for detecting the coating spraying quality of the aircraft component according to claim 8, wherein the merging processing mode is determined to be spraying similar merging for the states of adjacent target areas with curvature similar parameter difference values larger than or equal to preset curvature similar parameter difference values or spraying distance difference indexes larger than or equal to preset spraying distance difference indexes, the track reference directions of the adjacent areas are detected, and if the number of the adjacent areas with the same track reference direction as that of the target areas is 1, the target areas are merged to the adjacent areas with the same track reference direction.
  10. 10. The method according to claim 9, wherein if the number of adjacent areas in the same reference direction as the trajectory of the target area is not 1, the target area is merged into the adjacent area having the greatest similarity of the spray output parameters.

Description

Method for detecting spraying quality of coating of aircraft component Technical Field The invention relates to the field of spray quality detection, in particular to a method for detecting the spray quality of a coating of an aircraft component. Background Along with the gradual maturity of aircraft automatic spraying technology, more and more aircraft parts select to carry out automatic spraying by the robot, but the coating spraying quality that its corresponds still is mostly carried out the visual inspection by the manual work, in order to improve detection efficiency and detection correct rate, adopt laser light source gradually to detect spraying quality, but the detection of aircraft parts that the structure is complicated often is difficult to once accomplish, so need carry out regional setting to it according to the structure of actual detection object when detecting, how to promote the setting effect in detection region with the improvement detection efficiency is the problem that the person of the art is more concerned. The Chinese patent publication No. CN108212719A discloses an automatic spraying area division and robot station planning method for an aircraft complete machine, which comprises the steps of firstly dividing the aircraft into three parts including a machine head, a machine body and wings (including tail wings) by combining the surface appearance characteristics of the complete machine, abstracting the area into basic shapes such as a round table surface, a cylindrical surface and a plane, establishing a single spraying area unit (simply referred to as an area unit) with regular geometric appearance by using a single maximum intersection of a usable working space of the robot and the surface of the geometric body as a basis, and finally finishing the arrangement of areas on the outer surface of the aircraft by using the area unit to realize station planning of the robot. Disclosure of Invention Therefore, the invention provides a method for detecting the spraying quality of an aircraft component coating, which is used for solving the problems that the area acquisition is only carried out according to the simple shape of the aircraft component in the prior art, and the area acquisition mode cannot meet the detection efficiency and the detection flexibility of the structural characteristics of the aircraft component. In order to achieve the above purpose, the invention provides a method for detecting the spraying quality of an aircraft component coating, which comprises the following steps: Determining the category of the attention point according to the neighborhood vector difference degree and the maximum reference difference degree, and determining whether to adopt an adaptive division mode or not based on the proportion of the attention points of the category and the proportion of the area of the analysis area; Under the adaptive division mode, a first class of attention points and a second class of attention points are respectively screened according to the curvature processing values to obtain a first seed point and a second seed point, and the seed points are arranged according to the order of the curvature processing values from large to small to obtain a primary selection sequence; determining whether to sequentially adjust the initial sequence according to curvature processing values corresponding to the comparison conditions so as to obtain a final sequence; Sequentially carrying out region growth on various seed points in the final sequence to obtain a plurality of target regions comprising the seed points and corresponding association points, wherein the association points are selected based on curvature processing difference values and symbol similarity indexes; Aiming at a target area with the area smaller than the area of a preset area, determining to carry out merging treatment; in the merging process, the states of the adjacent target areas are determined according to the curvature similarity parameter difference value and the spraying distance difference index, and the merging process mode is curvature similarity merging or spraying similarity merging based on the states of the adjacent target areas. Further, performing region construction on the second class of interest points of the target component model to obtain a model region with the smallest area capable of comprising all the second class of interest points, and marking the model region as an analysis region; If the second class attention point occupation ratio is larger than the preset second class attention point occupation ratio or the analysis area occupation ratio is smaller than or equal to the preset analysis area occupation ratio, adopting an adaptive division mode; and if the second class of attention point occupation ratio is smaller than or equal to the preset second class of attention point occupation ratio and the analysis area occupation ratio is larger than the preset analysis area occupatio