WO2021227288A1 - 蜂窝产品的几何形态评估方法和检测及评估系统 - Google Patents

蜂窝产品的几何形态评估方法和检测及评估系统 Download PDF

Info

Publication number
WO2021227288A1
WO2021227288A1 PCT/CN2020/109731 CN2020109731W WO2021227288A1 WO 2021227288 A1 WO2021227288 A1 WO 2021227288A1 CN 2020109731 W CN2020109731 W CN 2020109731W WO 2021227288 A1 WO2021227288 A1 WO 2021227288A1
Authority
WO
WIPO (PCT)
Prior art keywords
cell
top surface
honeycomb
deviation angle
value
Prior art date
Application number
PCT/CN2020/109731
Other languages
English (en)
French (fr)
Inventor
王中钢
崔灿
施冲
梁习峰
李振东
孙博
雷紫平
Original Assignee
中南大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from CN202010387864.9A external-priority patent/CN111583234B/zh
Priority claimed from CN202010388343.5A external-priority patent/CN111583238B/zh
Application filed by 中南大学 filed Critical 中南大学
Priority to US17/422,721 priority Critical patent/US11893725B2/en
Publication of WO2021227288A1 publication Critical patent/WO2021227288A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G06T7/0006—Industrial image inspection using a design-rule based approach
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/90—Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/695—Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20021—Dividing image into blocks, subimages or windows
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30164—Workpiece; Machine component

Definitions

  • the invention relates to the fields of design, manufacture and application of lightweight structural products for transportation, machinery, aerospace, shipbuilding and other equipment, and particularly relates to a geometric form evaluation method and a detection and evaluation system for honeycomb products.
  • the lightweight honeycomb structure is widely used in various engineering fields due to its excellent load-bearing and energy-absorbing properties.
  • various structural defects such as the bend, warpage, and cell deformity of the honeycomb core block inevitably appear, and these defects have been proved to have a greater bearing and energy absorption performance. Influence.
  • the honeycomb product has a periodic arrangement of porous structures, with typical features such as multiple vertices, thin walls, and wide bearing surfaces, traditional ultrasonic inspection technology cannot obtain the characteristic information of its structural defects. Therefore, work on the regularity inspection and evaluation of cellular products to avoid the risk of using inferior products is urgently needed.
  • the object of the present invention is to provide a geometric shape evaluation method and geometric shape detection and evaluation system of a honeycomb product that can accurately evaluate the geometric shape of the honeycomb product.
  • the first aspect of the present invention provides a method for evaluating the geometric form of a honeycomb product, which includes: acquiring a top surface image and a side image of the honeycomb product; extracting vertices from the top surface image to obtain vertex coordinates; Perform cell reconstruction on the top surface image to obtain the serial number of the six vertices of each cell; based on the serial number of the six vertices of each cell and the vertex coordinates of the six vertices of each cell, Calculate the deviation value of the six internal angles of each cell; extract the top edge and side edge of the honeycomb product from the top image and the side image; calculate the top edge and the side edge based on the top edge and the side edge The maximum deflection of the top surface of the honeycomb product and the maximum deflection of the side surface; determine whether the honeycomb product is qualified based on the deviation angle, the maximum deflection of the top surface and the maximum deflection of the side surface.
  • the determining whether the honeycomb product is qualified based on the deviation angle, the maximum deflection of the top surface, and the maximum deflection of the side surface includes: based on the deviation angle, the maximum deflection of the top surface, and the side surface
  • the maximum deflection determines the geometric regularity of the honeycomb product, and if the geometric regularity reaches a preset standard, the honeycomb product is qualified.
  • the geometric regularity includes: top cell regularity, side flatness, and top flatness; the determination is based on the deviation angle, the maximum deflection of the top surface, and the maximum deflection of the side surface
  • the geometric regularity of the honeycomb product if the geometric regularity reaches a preset standard, the honeycomb product is qualified, including: determining the cell regularity of the top surface based on the deviation angle; Deflection judges the straightness of the side surface; judges the straightness of the top surface based on the maximum deflection of the side; If the standard is preset, the honeycomb product is qualified.
  • the judgment index of the regularity of the top surface cell includes: the maximum deviation angle, the average deviation angle, the maximum value of the average value of the deviation angle of the cell, the standard deviation of the deviation angle, and the standard deviation of the average value of the deviation angle of the cell.
  • the maximum deviation angle, the average deviation angle, the maximum value of the average value of the cell deviation angle, the standard deviation of the deviation angle, and the standard of the average value of the cell deviation angle If one or more of the differences are less than the preset threshold, the regularity of the top surface cell reaches the preset standard.
  • the judgment index of the regularity of the top surface cell includes: the maximum deviation angle, the average deviation angle, the maximum value of the average value of the deviation angle of the cell, the standard deviation of the deviation angle, and the standard deviation of the average value of the deviation angle of the cell.
  • the maximum deviation angle, the average deviation angle, the maximum value of the average value of the cell deviation angle, the standard deviation of the deviation angle, and the standard of the average value of the cell deviation angle If the ratio of one or more of the differences greater than the preset threshold is less than the preset ratio, the regularity of the top surface cell reaches the preset standard.
  • the judgment index of the regularity of the top surface cell includes: the average value of the deviation angle of the cell; If the percentage of the whole cell is smaller than the corresponding thresholds B1, B2, B3, B4, and B5, the regularity of the top surface cell reaches the standard.
  • the method for extracting vertices includes: obtaining a skeleton diagram of a honeycomb product; extracting boundary pixels belonging to the same cell from the skeleton diagram to obtain the sequence of boundary pixels of the cell; Obtain the included angles corresponding to the boundary pixels to obtain the sequence of included angles corresponding to the boundary pixels; the included angles are in the sequence of boundary pixels, between two points of the boundary pixels Side, the angle between two pixels that are separated from the boundary pixel by a first preset value and the line connecting the boundary pixel of the cell; using a window of a second preset size, the clip
  • the angle sequence performs non-minimum value suppression, and the included angle excluding the minimum value of the included angle in the window is set as the first preset included angle; when the non-minimum value suppression of the included angle sequence is completed , Determining that the boundary pixels corresponding to the included angles that are smaller than the second preset included angle among all the minimum values are the vertices.
  • the method for extracting vertices includes: obtaining a skeleton diagram of a vertex image; searching the eight neighborhoods of each pixel on the skeleton diagram in a preset order for a week, and if the number of pixel value changes is 6, then The pixel points are vertices.
  • the method for cell reconstruction includes: using a Moore neighborhood tracking algorithm to track the boundary of each cell; taking each traced boundary point as the center, establishing a window of a preset size, if there is Vertex, the number of the vertex is recorded and stored under the name of the cell, and the vertices of the cell are sorted according to the order of the vertices encountered. After the tracking is completed, the vertices of each cell are connected in order to obtain the cell weight Composition.
  • the second aspect of the present invention provides a geometric form detection and evaluation system for honeycomb products, which is used to implement the above geometric form evaluation method, including: a table for placing honeycomb products; The face and the side are used to take pictures of the honeycomb product to obtain the original top surface image and the original side image of the honeycomb product; the camera moving device is used to move the camera on the top and side surfaces of the honeycomb product, In order to obtain a complete original top image and a complete original side image; an analysis and evaluation module, used to obtain the complete original top image and the complete original side image, and perform noise reduction filtering, binarization and morphology Filter processing to obtain the top surface image and the side image; perform vertex extraction on the top surface image to obtain the vertex coordinates; perform cell reconstruction on the top surface image to obtain the serial number of the six vertices of each cell; based on The serial numbers of the six vertices of each cell and the coordinates of the vertices of the six vertices of each cell to calculate the deviation value of the six internal angles of
  • the camera moving device includes: a walking gantry, the camera is connected to the walking gantry; a slide rail is arranged along the length of the honeycomb product, and the walking gantry is arranged at the place On the slide rail, move along the slide rail.
  • the geometric form evaluation method of the present invention in a non-contact manner, simply processes the image to accurately and harmlessly judge whether the geometric form of the honeycomb product is qualified.
  • Figure 1 is a flow chart of the method for evaluating the geometric form of a honeycomb product of the present invention
  • Figure 2 is a top view of the honeycomb product of the present invention.
  • Figure 3 is a side profile view of the honeycomb product of the present invention.
  • FIG. 4 is a schematic diagram of a top view structure of a geometric form detection and evaluation system for honeycomb products according to Embodiment 3 of the present invention.
  • Fig. 5 is a schematic diagram of the test structure of the geometric form detection and evaluation system of the honeycomb product according to the third embodiment of the present invention.
  • the drawing shows a schematic diagram of a layer structure according to an embodiment of the present invention.
  • the figures are not drawn to scale, in which some details are exaggerated for clarity purposes, and some details may be omitted.
  • the shapes of the various regions and layers shown in the figure and the relative size and positional relationship between them are only exemplary. In practice, there may be deviations due to manufacturing tolerances or technical limitations. Areas/layers with different shapes, sizes, and relative positions can be designed as needed.
  • Figure 2 is a top view of the honeycomb product of the present invention
  • Figure 3 is a side profile view of the honeycomb product of the present invention.
  • FIG 2. (a) is the top view of the standard honeycomb product and (b) is the top view of the honeycomb product to be evaluated.
  • the original side image of the honeycomb product of the present application can be obtained by looking at the arrow, which is shown in Figure 3.
  • Figure 3 where (a) is the side profile view of the standard honeycomb product, (b) is the side profile view of the honeycomb product to be evaluated, the original honeycomb product of the application can be obtained by looking at the arrow from the direction Top image.
  • Fig. 1 is a flowchart of a method for evaluating the geometric form of a honeycomb product according to Embodiment 1 of the present application.
  • this embodiment provides a method for evaluating the geometric form of a honeycomb product, which includes: obtaining a top surface image and a side image of the honeycomb product; extracting the vertices of the top surface image to obtain the vertex coordinates; Perform cell reconstruction to obtain the serial number of the six vertices of each cell; calculate the deviation value of the six internal angles of each cell based on the serial number of the six vertices of each cell and the vertex coordinates of the six vertices of each cell ; Extract the top and side edges of the honeycomb product from the preprocessed top and side images; calculate the maximum deflection of the honeycomb product's top surface and the maximum side deflection based on the top and side edges; based on the deviation angle and the top surface The maximum deflection and the maximum lateral deflection determine whether the honeycomb product is qualified.
  • FIG 2 (b) is a top view of the honeycomb product to be evaluated, in which the dotted line is the top profile view of the standard honeycomb product, and the maximum side deflection can be seen in Figure 2.
  • FIG 3 (b) is a top view of the honeycomb product to be evaluated, where the dotted line is a top profile view of the standard honeycomb product, and the maximum lateral deflection can be seen in Figure 2.
  • the invention judges whether the geometric regularity is qualified by acquiring the top and side images of the honeycomb product, and in a non-contact manner, by simply processing the image, it can accurately determine whether the geometric form of the honeycomb product is qualified.
  • the mechanical performance of the honeycomb product conforms to the wing surface, cabin surface, hatch cover, floor, engine shield, tail nozzle, muffler board, heat insulation board, satellite star shell, rigid solar battery wing, parabolic surface Requirements for products such as antennas and rocket propellant tank bottoms.
  • the absolute value of the difference between the internal angle and 120° is the deviation angle of the corresponding vertex of the cell.
  • determining whether the honeycomb product is qualified based on the deviation angle, the maximum deflection of the top surface and the maximum deflection of the side surface includes: determining the geometric regularity of the honeycomb product based on the deviation angle, the maximum deflection of the top surface and the maximum deflection of the side surface, If the geometric regularity reaches the preset standard, the honeycomb product is qualified.
  • Geometric regularity includes: top cell regularity, side flatness and top flatness;
  • acquiring the top image and side image of the honeycomb product includes acquiring a complete original top image and acquiring a complete original side image, and comparing the original top image and acquiring a complete original side image.
  • the image is preprocessed to obtain the top image and the side image.
  • the preprocessing includes the following steps:
  • the image processing sequence required for the calculation of the regularity of the top cell includes noise reduction filtering, binarization, and morphological filtering: the noise reduction filtering adopts the median filter method to reduce the noise of the image; the binarization adopts the Otsu method and the honeycomb wall is executed.
  • the pixel of is 1, and the pixel of the background or hole is 0; Morphological filtering can correct the error of binarization;
  • Image processing required for side flatness calculation based on the above steps, perform hole filling, edge extraction, and contour filtering;
  • Image processing required for top surface flatness calculation perform edge extraction on the basis of noise reduction filtering, binarization, and morphological filtering; edge extraction is to use canny algorithm or/and sobel algorithm to extract the edge of the top image or side image
  • the contour line of the hole is a method of morphological hole filling, and the cell holes in the top image are filled with pixels with a pixel value of 1
  • the contour filtering is to filter the contour lines in the image by using the Gaussian filtering method To make it smoother;
  • the geometric regularity includes: top cell regularity, side flatness, and top flatness; determine the geometry of the honeycomb product based on the deviation angle, the maximum deflection of the top surface, and the maximum deflection of the side Regularity, if the geometric regularity reaches the preset standard, the honeycomb product is qualified, including: determining the regularity of the top surface cell based on the deviation angle; determining the straightness of the side based on the maximum deflection of the top surface; determining the straightness of the top surface based on the maximum deflection of the side Degree; if the top cell regularity, side flatness and top flatness all meet the preset standards, the honeycomb product is qualified.
  • the criteria for determining the regularity of the top surface cell include: the maximum deviation angle, the average deviation angle, the maximum value of the average deviation angle of the cell, the standard deviation of the deviation angle, and the average deviation angle of the cell.
  • the mechanical properties of the honeycomb product conform to the wing surface, cabin surface, hatch cover, floor, engine guard, tail nozzle, muffler board, heat insulation board, satellite star shell, rigid solar battery wing, parabolic antenna, Requirements for products such as the bottom of the rocket propellant tank.
  • the criteria for determining the regularity of the top surface cell include: the maximum deviation angle, the average deviation angle, the maximum value of the average deviation angle of the cell, the standard deviation of the deviation angle, and the average deviation angle of the cell.
  • the mechanical properties of the honeycomb product conform to the wing surface, cabin surface, hatch cover, floor, engine guard, tail nozzle, muffler board, heat insulation board, satellite star shell, rigid solar battery wing, parabolic antenna, Requirements for products such as the bottom of the rocket propellant tank.
  • the judgment index of the regularity of the top surface cell includes: the average value of the deviation angle of the cell; if the average value of the deviation angle of the cell is not greater than the cells of A1, A2, A3, A4, and A5 in sequence If the percentage of the whole cell is smaller than the corresponding thresholds B1, B2, B3, B4 and B5, the regularity of the top cell reaches the standard.
  • the mechanical properties of the honeycomb product conform to the wing surface, cabin surface, hatch cover, floor, engine guard, tail nozzle, muffler board, heat insulation board, satellite star shell, rigid solar battery wing, parabolic antenna, Requirements for products such as the bottom of the rocket propellant tank.
  • the maximum deviation angle the maximum value of all deviation angles
  • Average deviation angle the average value of all deviation angles
  • the maximum value of the average value of the deviation angle of a cell first calculate the average value of each deviation angle of a single cell, which is the average value of the deviation angle of the cell, and the maximum value of the average value of the deviation angle of all cells is the cell deviation angle average value;
  • Standard deviation of deviation angle Take all deviation angles as samples and calculate the standard deviation of deviation angle
  • Standard deviation of the average value of the deviation angle of the cell Take the average value of all the deviation angles of the cell as the sample, and calculate the standard deviation of the average value of the deviation angle of the cell;
  • the methods for extracting vertices include: contour angle method, branch point method, HARRIS method, and moving window method.
  • the contour angle method includes:
  • the included angles are in the sequence of boundary pixels, between two points of the boundary pixels Side, the angle between two pixel points that are separated from the boundary pixel point by a first preset value and the line connecting the cell boundary pixel point;
  • the branch point method includes: obtaining a skeleton diagram of the vertex image; searching the eight neighborhoods of each pixel on the skeleton diagram in a preset order for a week, and if the number of pixel value changes is 6, then the pixel is a vertex.
  • a line with a pixel value of 1 is drawn into a skeleton diagram using a line segment with a line width of 1 pixel;
  • a skeleton diagram with k pixels is used as the basis to traverse pixel 1
  • search for a circle around the eight neighborhoods of the pixel in a clockwise or counterclockwise direction to obtain a number of pixel value changes; if the number of pixel value changes is 4 , It has been shown that there are two straight lines at the pixel, and when the two straight lines have a reasonable angle through the coordinate calculation, the pixel is determined as the edge vertex and recorded; if the number of pixel value changes is 6, the pixel has been displayed If there are three straight lines, the pixel is determined to be the middle vertex and recorded, and the vertex extraction is completed.
  • the morphological image is obtained by preprocessing.
  • the morphological image is obtained by preprocessing.
  • the moving window method includes: S1: Obtain a square window with a preset side length L. After the square window traverses the vertex image, the minimum value of the number of pixels in the square window with a pixel value of 0 is non-zero; among them, the preset side length is obtained
  • S2 Use a square window to traverse each pixel with a pixel value of 1 in the vertex image, and then assign the sum of the number of pixels with a pixel value of 1 in each square window to the pixel at the center of the square window;
  • the method of cell reconstruction includes: a neighborhood window recursive method.
  • the neighborhood window recursive method includes: using the Moore neighborhood tracking algorithm to track the boundary of each cell; taking each tracked boundary point as the center, establishing a window of a preset size, if there is a vertex in the window, record the vertex The number of is stored under the name of the cell, and the vertices of the cell are sorted according to the order of the vertices encountered. After the tracking is completed, the vertices of each cell are connected in order to obtain a cell reconstruction graph.
  • cell reconstruction refers to the establishment of the mapping relationship between cells and vertices, that is, which vertices belong to the same cell, and the vertices are connected accordingly to obtain the image of the honeycomb structure.
  • the method used is adjacent Domain window recursive method: Use the Moore neighborhood tracking algorithm to track the boundary of each cell, and establish a window of appropriate size with each boundary point tracked as the center. If there is a vertex in the window, record the vertex number and Store it under the name of the cell, and sort the vertices of the cell according to the sequence of the vertices encountered. After the tracking is completed, the vertices of each cell are connected in order to obtain a cell reconstruction graph.
  • the step “reconstructing cells” includes edge expansion and vertex connection;
  • the step “edge expansion” is based on the morphological image, and at least one of the four edges of the morphological image is expanded outward
  • the width of the pixels forms the expansion area, and the pixel values of all pixels in the expansion area are all set to 1, to obtain the expanded image;
  • FIG. 3 is a schematic diagram of the top view structure of the geometric form detection and evaluation system for honeycomb products according to Embodiment 2 of the present invention
  • FIG. 4 is a schematic diagram of the test structure of the geometric form detection and evaluation system for honeycomb products according to Embodiment 2 of the present invention.
  • this embodiment provides a geometric form detection and evaluation system for honeycomb products, including:
  • the camera 3 is set on the top surface and the side surface of the honeycomb product 2 for taking pictures of the honeycomb product 2 to obtain the original top surface image and the original side image of the honeycomb product 2;
  • the camera moving device 4 is used to move the camera 3 on the top surface and the side surface of the honeycomb product 2 to obtain a complete original top surface image and a complete original side image;
  • Analysis and evaluation module 5 used to obtain a complete original top image and a complete original side image, and perform noise reduction filtering, binarization and morphological filter processing to obtain the top image and the side image, and perform vertices on the top image Extract to obtain the vertex coordinates; perform cell reconstruction on the top surface image to obtain the serial number of the six vertices of each cell; based on the serial number of the six vertices of each cell and the vertex coordinates of the six vertices of each cell, Calculate the deviation value of the six internal angles of each cell; extract the top and side edges of the honeycomb product 2 from the preprocessed top and side images; calculate the maximum top surface of the honeycomb product 2 based on the top and side edges Deflection and maximum lateral deflection; determine whether the honeycomb product 2 is qualified based on the deviation angle, the maximum deflection of the top surface and the maximum deflection of the side.
  • the present invention judges whether the geometric form of the honeycomb product is qualified by acquiring the top and side images of the honeycomb product, and in a non-contact manner, by simply processing the image, it can accurately judge whether the geometric form of the honeycomb product 2 is qualified.
  • the analysis and evaluation module 5 of this embodiment is also used to implement all the content of the embodiment 1, which will not be repeated here.
  • the camera moving device 4 includes: a walking gantry 41, the camera 3 is connected to the walking gantry 41; a slide rail 42 is arranged along the length of the honeycomb product 2, and the walking gantry 41 It is arranged on the sliding rail 42 and moves along the sliding rail 42.
  • the camera moving device 4 further includes: a camera 3 sliding component.
  • the camera moving component is slidably connected with the walking gantry 41, the camera 3 is arranged on the camera moving component, and the camera 3 slides on the walking gantry 41 through the machine moving component.
  • the camera moving component includes: a roller 43 and a moving block 44.
  • the moving block 44 is sleeved on the walking gantry 41, and is connected to the walking gantry 41 through the roller 43, so that the moving block 44 slides on the walking gantry 41, the camera 3 is detachably connected with the moving block 44, and the complete original top surface image and original side image of the honeycomb product 2 are obtained by the movement of the moving block 44.
  • the geometric form detection and evaluation system of the honeycomb product further includes a clamp 6 for fixing the position of the honeycomb product 2 on the stage 1 so that the camera 3 can take pictures.
  • the clamp 6 is composed of four flat plates and a driving device, which can be moved closer to the honeycomb product 2 under the action of the driving device, and locked after being close to the honeycomb product 2 to be tested, for positioning the tested honeycomb product 2 Honeycomb products 2.
  • the side of the fixture 6 close to the honeycomb product 2 is painted bright yellow, so that the boundary of the honeycomb product 2 can be clearly defined in the image to assist image processing.
  • the geometric form detection and evaluation system of the honeycomb product 2 further includes a lifting device 7 of the storage table 1 for adjusting the height of the honeycomb product 2 so that the honeycomb product 2 is set on the top surface of the honeycomb product 2
  • the camera 3 is at a suitable distance so that the camera 3 can take pictures.
  • the upper plane of the honeycomb product 2 and the upper plane of the clamp 6 can be maintained at the same height by the lifting device 7, and the vertical distance between the clamp 6 and the camera 3 is unchanged, and the upper plane of the honeycomb product 2 and the camera 3 are further maintained. Keep a fixed distance so that the camera 3 can take pictures.
  • the lifting device 7 includes a hydraulic rod.
  • the geometric form detection and evaluation system of the honeycomb product 2 further includes a display module connected to the analysis and evaluation module 5 for displaying whether the geometric form of the honeycomb product 2 is qualified.
  • the display module includes a display lamp. If the honeycomb product 2 is qualified, it will turn on a green light, and if it fails, it will turn on a red light.
  • the geometric form detection and evaluation system of the honeycomb product 2 also includes a calibration module:
  • the calibration board is a display board with an electronic ink screen that can display standard honeycombs with adjustable side length and wall thickness. The outside of the screen displays a color that contrasts with the honeycomb.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

蜂窝产品的几何形态评估方法和检测及评估系统，其中评估方法，包括：获取蜂窝产品的顶面图像和侧面图像；对顶面图像进行顶点提取，得到顶点坐标；对顶面图像进行胞元重构，得到各胞元的六个顶点的序号编号；基于各胞元的六个顶点的序号编号和各胞元六个顶点的顶点坐标，计算各胞元六个内角的偏差值；从顶面图像和侧面图像中提取蜂窝产品的顶面边线和侧面边线；基于顶面边线和侧面边线计算蜂窝产品的顶面最大挠度和侧面最大挠度；基于偏差角、顶面最大挠度和侧面最大挠度判定蜂窝产品是否合格。上述几何形态评估方法，以非接触式的方式，通过对图像进行简单处理，以准确并且对蜂窝产品无损伤的判断蜂窝产品的几何形态是否合格。

Description

蜂窝产品的几何形态评估方法和检测及评估系统 技术领域
本发明涉及交通、机械、航空航天、船舶等装备的轻质结构产品设计、制造及应用等领域，特别涉及蜂窝产品的几何形态评估方法和检测及评估系统。
背景技术
轻质蜂窝结构以其优异的承载与吸能特性而被广泛应用到各种工程领域。然而，在该产品的生产制造过程中，不可避免地出现蜂窝芯块拱弯、翘曲、胞孔畸形等各型结构性缺陷，而这些缺陷已被证实对其承载与吸能性能产生较大影响。并且，由于蜂窝产品为周期排列多孔结构，具有典型的多顶点、细薄壁、承载面宽等特征，传统超声检测技术无法获得其结构性缺陷的特征信息。因此，有关蜂窝产品规整性检测与评估以规避低劣产品的使用风险的工作亟待开展。
发明内容
(一)发明目的
本发明的目的是提供一种可准确评估蜂窝产品的几何形态的蜂窝产品的几何形态评估方法和几何形态检测及评估系统。
(二)技术方案
为解决上述问题，本发明的第一方面提供了一种蜂窝产品的几何形态评估方法，包括：获取蜂窝产品的顶面图像和侧面图像；对所述顶面图像进行顶点提取，得到顶点坐标；对所述顶面图像进行胞元重构，得到各胞元的六个顶点的序号编号；基于所述各胞元的六个顶点的序号编号和所述各胞元六 个顶点的顶点坐标，计算各胞元六个内角的偏差值；从所述顶面图像和所述侧面图像中提取所述蜂窝产品的顶面边线和侧面边线；基于所述顶面边线和所述侧面边线计算所述蜂窝产品的顶面最大挠度和侧面最大挠度；基于所述偏差角、所述顶面最大挠度和所述侧面最大挠度判定所述蜂窝产品是否合格。
可选地，所述基于所述偏差角、所述顶面最大挠度和所述侧面最大挠度判定所述蜂窝产品是否合格，包括：基于所述偏差角、所述顶面最大挠度和所述侧面最大挠度判定所述蜂窝产品的几何规整度，若所述几何规整度达到预设标准，则所述蜂窝产品合格。
可选地，所述几何规整度包括：顶面胞元规整度、侧面平直度和顶面平直度；所述基于所述偏差角、所述顶面最大挠度和所述侧面最大挠度判定所述蜂窝产品的几何规整度，若所述几何规整度达到预设标准，则所述蜂窝产品合格，包括：基于所述偏差角判定所述顶面胞元规整度；基于所述顶面最大挠度判定所述侧面平直度；基于所述侧面最大挠度判定所述顶面平直度；若所述顶面胞元规整度、所述侧面平直度和所述顶面平直度均达到预设标准，则所述蜂窝产品合格。
可选地，所述顶面胞元规整度的判定指标包括：最大偏差角、平均偏差角、胞元偏差角平均值的最大值、偏差角的标准差和胞元偏差角平均值的标准差中的一种或多种；所述最大偏差角、所述平均偏差角、所述胞元偏差角平均值的最大值、所述偏差角的标准差和所述胞元偏差角平均值的标准差中的一种或多种均小于预设阈值，则所述顶面胞元规整度达到预设标准。
可选地，所述顶面胞元规整度的判定指标包括：最大偏差角、平均偏差角、胞元偏差角平均值的最大值、偏差角的标准差和胞元偏差角平均值的标准差中的一种或多种；所述最大偏差角、所述平均偏差角、所述胞元偏差角平均值的最大值、所述偏差角的标准差和所述胞元偏差角平均值的标准差中的一种或多种大于预设阈值的比例小于预设比例，则所述顶面胞元规整度达到预设标准。
可选地，所述顶面胞元规整度的判定指标包括：胞元偏差角平均值；若 所述胞元偏差角平均值依次不大于A1、A2、A3、A4和A5的胞元所占整体胞元的百分比，依次小于相应阈值B1、B2、B3、B4和B5，则所述顶面胞元规整度达到标准。
可选地，所述顶点提取的方法包括：获取蜂窝产品的骨架图；从所述骨架图中提取出属于同一个胞元的边界像素点，得到所述胞元的所述边界像素点序列；分别获取与所述边界像素点对应的夹角，得到与所述边界像素点对应的所述夹角序列；所述夹角为在所述边界像素点序列中，在所述边界像素点的两侧，分别与所述边界像素点相隔第一预设值个像素的两个像素点与所述胞元边界像素点连线的夹角；采用第二预设值大小的窗口，对所述夹角序列进行非极小值抑制，将除了所述窗口中夹角的最小值之外的夹角设置为第一预设夹角；当完成所述夹角序列的所述非极小值抑制时，确定所有所述最小值中小于第二预设夹角的夹角对应的所述边界像素点为所述顶点。
可选地，所述顶点提取的方法包括：获取顶点图像的骨架图；在所述骨架图上每个像素点的八邻域按照预设顺序搜寻一周，若像素值变化次数为6，则所述像素点为顶点。
可选地，所述顶点提取的方法包括S1：获取预设边长L的正方形窗口，设置一个边长从3个像素依次增大的正方形窗口，所述正方形窗口遍历所述顶点图像后，所述正方形窗口内的像素值为0的像素数的最小值首次为非零S2：以所述正方形窗口遍历所述顶点图像中每个像素值为1的像素点，然后将每个所述正方形窗口内的像素值为1的像素个数的总和赋值给所述正方形窗口的中心点的像素点；S3：以最大赋值的像素点为中心建立初始边长E＝L的正方形区域，计算出所述正方形区域四个边界上各像素点的赋值与正方形区域中心点的赋值的差值绝对值，记录最小的差值绝对值Z及其对应的边界像素点的坐标；S4：以E＝E+2个像素建立新的正方形区域，重复S3，直到最小的差值绝对值有明显的反向增大的趋势为止，以此时取得最小的差值绝对值Z的边界像素点坐标与对应的正方形区域中心点的坐标，通过所述Z值得边界像素点坐标和所述对应的正方形区域中心点的坐标的坐标值求解得到蜂 窝胞元边长A；S5：将最大赋值的像素点，确定为顶点，再以所述顶点为中心，建立一个以蜂窝胞元边长A为边长的正方形的湮灭窗口，将所述湮灭窗口内所有像素值＝1的像素点上的赋值全部清零，在此基础上，再在剩余的赋值中再找到最大赋值的像素点确定为顶点并记录，重复湮灭窗口的操作，直到像素值＝1的像素点上的赋值小于给定阈值为止，提取所有顶点。
可选地，所述胞元重构的方法包括：采用摩尔邻域追踪算法追踪各胞元的边界；以追踪到的每个边界点为中心，建立一个预设大小的窗口，若窗口内有顶点，则记录该顶点的编号并存储在该胞元名下，并根据所遇顶点的先后顺序将胞元的顶点排序，完成追踪后将各胞元的顶点按顺序连线，得到胞元重构图。
本发明的第二方面提供了一种蜂窝产品的几何形态检测及评估系统，用于实现上述几何形态评估方法，包括：置物台，用于放置蜂窝产品；相机，设置于所述蜂窝产品的顶面和侧面，用于对所述蜂窝产品进行拍照，得到所述蜂窝产品原始顶面图像和原始侧面图像；相机移动装置，用于将所述相机在所述蜂窝产品的顶面和侧面移动，以得到完整的原始顶面图像和完整的原始侧面图像；分析评估模块，用于获取所述完整的原始顶面图像和所述完整的原始侧面图像，并进降噪滤波、二值化和形态学滤波处理，得到顶面图像和侧面图像；对所述顶面图像进行顶点提取，得到顶点坐标；对所述顶面图像进行胞元重构，得到各胞元的六个顶点的序号编号；基于所述各胞元的六个顶点的序号编号和所述各胞元六个顶点的顶点坐标，计算各胞元六个内角的偏差值；从所述预处理后的顶面图像和侧面图像中提取所述蜂窝产品的顶面边线和侧面边线；基于所述顶面边线和所述侧面边线计算所述蜂窝产品的顶面最大挠度和侧面最大挠度；基于所述偏差角、所述顶面最大挠度和所述侧面最大挠度判定所述蜂窝产品是否合格。
可选地，所述相机移动装置包括：行走式龙门架，所述相机与所述行走式龙门架连接；滑轨，沿所述蜂窝产品的长度方向设置，所述行走式龙门架设置于所述滑轨上，沿所述滑轨移动。
(三)有益效果
本发明的上述技术方案具有如下有益的技术效果：
本发明的几何形态评估方法，以非接触式的方式，通过对图像进行简单处理，以准确并且对蜂窝产品无损伤的判断蜂窝产品的几何形态是否合格。
附图说明
图1是本发明的蜂窝产品的几何形态评估方法的流程图；
图2是本发明的蜂窝产品俯视图；
图3是本发明的蜂窝产品侧视轮廓图；
图4是本发明的实施例3的蜂窝产品的几何形态检测及评估系统俯视结构示意图；
图5是本发明的实施例3的蜂窝产品的几何形态检测及评估系统测试结构示意图。
附图标记：
1：置物台；2：蜂窝产品；3：相机；4：相机移动装置；41：行走式龙门架；42：滑轨；43：滚轮；44：移动块；5：分析评估模块；6：夹具；7：升降装置。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚明了，下面结合具体实施方式并参照附图，对本发明进一步详细说明。应该理解，这些描述只是示例性的，而并非要限制本发明的范围。此外，在以下说明中，省略了对公知结构和技术的描述，以避免不必要地混淆本发明的概念。
在附图中示出了根据本发明实施例的层结构示意图。这些图并非是按比例绘制的，其中为了清楚的目的，放大了某些细节，并且可能省略了某些细节。图中所示出的各种区域、层的形状以及它们之间的相对大小、位置关系 仅是示例性的，实际中可能由于制造公差或技术限制而有所偏差，并且本领域技术人员根据实际所需可以另外设计具有不同形状、大小、相对位置的区域/层。
显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
在本发明的描述中，需要说明的是，术语“第一”、“第二”、“第三”仅用于描述目的，而不能理解为指示或暗示相对重要性。
此外，下面所描述的本发明不同实施方式中所涉及的技术特征只要彼此之间未构成冲突就可以相互结合。
图2是本发明的蜂窝产品俯视图；图3是本发明的蜂窝产品侧视轮廓图。
在描述本发明的实施例之前，先对本发明的方位用词做一个介绍，请参照图2，其中，(a)为标准蜂窝产品的俯视图，(b)为待评估的蜂窝产品的俯视图，由箭头防向看去可得到本申请的蜂窝产品的原始侧面图像，即图3。请参照图3，其中，(a)为标准蜂窝产品的侧视轮廓图，(b)为待评估的蜂窝产品的侧视轮廓图，由箭头防向看去可得到本申请的蜂窝产品的原始顶面图像。
实施例1
图1是本申请实施例1的蜂窝产品的几何形态评估方法的流程图。
如图1所示，本实施例提供了一种蜂窝产品的几何形态评估方法，包括：获取蜂窝产品的顶面图像和侧面图像；对顶面图像进行顶点提取，得到顶点坐标；对顶面图像进行胞元重构，得到各胞元的六个顶点的序号编号；基于各胞元的六个顶点的序号编号和各胞元六个顶点的顶点坐标，计算各胞元六个内角的偏差值；从预处理后的顶面图像和侧面图像中提取蜂窝产品的顶面边线和侧面边线；基于顶面边线和侧面边线计算蜂窝产品的顶面最大挠度和侧面最大挠度；基于偏差角、顶面最大挠度和侧面最大挠度判定蜂窝产品是否合格。参照图2，(b)为待评估的蜂窝产品的俯视图，其中，虚线部分为 为标准蜂窝产品的俯视轮廓图，由图2中可见侧面最大挠度。参照图3，(b)为待评估的蜂窝产品的俯视图，其中，虚线部分为标准蜂窝产品的俯视轮廓图，由图2中可见侧面最大挠度。
本发明通过获取蜂窝产品的顶面及侧面图像判定几何规整度是否合格，以非接触式的方式，通过对图像进行简单处理，以准确判断蜂窝产品的几何形态是否合格。蜂窝产品的几何形态合格后，蜂窝产品的力学性能符合翼面、舱面、舱盖、地板、发动机护罩、尾喷管、消音板、隔热板、卫星星体外壳、刚性太阳电池翼、抛物面天线、火箭推进剂贮箱箱底等产品的要求。
具体地，内角与120°差的绝对值，即为该胞元对应顶点的偏差角。
本实施例的可选实施方案中，基于偏差角、顶面最大挠度和侧面最大挠度判定蜂窝产品是否合格，包括：基于偏差角、顶面最大挠度和侧面最大挠度判定蜂窝产品的几何规整度，若几何规整度达到预设标准，则蜂窝产品合格。几何规整度包括：顶面胞元规整度、侧面平直度和顶面平直度；
本实施例的可选实施方案中，获取蜂窝产品的顶面图像和侧面图像，包括，获取完整的原始顶面图像和获取完整的原始侧面图像，并对原始顶面图像和获取完整的原始侧面图像进行预处理，得到顶面图像和侧面图像。
其中，预处理包括以下步骤：
顶面胞元规整度计算所需图像处理顺序包括降噪滤波、二值化、形态学滤波：降噪滤波采用中值滤波法，降低图像的噪声；二值化采用大津法，执行后蜂窝壁的像素为1，背景即孔洞部分的像素为0；形态学滤波可以纠正二值化的误差；
侧面平直度计算所需图像处理：在上述步骤基础上，再执行孔洞填充、边缘提取、轮廓线滤波；
顶面平直度计算所需图像处理：在降噪滤波、二值化、形态学滤波基础上，再执行边缘提取；边缘提取是采用canny算法或/和sobel算法提取顶面图像或侧面图像边缘的轮廓线；孔洞填充是采用形态学孔洞填充的方法，将顶面图像中的元胞孔采用像素值为1的像素填满；轮廓线滤波是采用高斯滤 波法对图像中的轮廓线进行滤波，使其更加平滑；
本实施例的可选实施方案中，几何规整度包括：顶面胞元规整度、侧面平直度和顶面平直度；基于偏差角、顶面最大挠度和侧面最大挠度判定蜂窝产品的几何规整度，若几何规整度达到预设标准，则蜂窝产品合格，包括：基于偏差角判定顶面胞元规整度；基于顶面最大挠度判定侧面平直度；基于侧面最大挠度判定顶面平直度；若顶面胞元规整度、侧面平直度和顶面平直度均达到预设标准，则蜂窝产品合格。
本实施例的可选实施方案中，顶面胞元规整度的判定指标包括：最大偏差角、平均偏差角、胞元偏差角平均值的最大值、偏差角的标准差和胞元偏差角平均值的标准差中的一种或多种；最大偏差角、平均偏差角、胞元偏差角平均值的最大值、偏差角的标准差和胞元偏差角平均值的标准差中的一种或多种均小于预设阈值，则顶面胞元规整度达到预设标准。达到预设标准后，蜂窝产品的力学性能符合翼面、舱面、舱盖、地板、发动机护罩、尾喷管、消音板、隔热板、卫星星体外壳、刚性太阳电池翼、抛物面天线、火箭推进剂贮箱箱底等产品的要求。
本实施例的可选实施方案中，顶面胞元规整度的判定指标包括：最大偏差角、平均偏差角、胞元偏差角平均值的最大值、偏差角的标准差和胞元偏差角平均值的标准差中的一种或多种；最大偏差角、平均偏差角、胞元偏差角平均值的最大值、偏差角的标准差和胞元偏差角平均值的标准差中的一种或多种大于预设阈值的比例小于预设比例，则顶面胞元规整度达到预设标准。达到预设标准后，蜂窝产品的力学性能符合翼面、舱面、舱盖、地板、发动机护罩、尾喷管、消音板、隔热板、卫星星体外壳、刚性太阳电池翼、抛物面天线、火箭推进剂贮箱箱底等产品的要求。
本实施例的可选实施方案中，顶面胞元规整度的判定指标包括：胞元偏差角平均值；若胞元偏差角平均值依次不大于A1、A2、A3、A4和A5的胞元所占整体胞元的百分比，依次小于相应阈值B1、B2、B3、B4和B5，则顶面胞元规整度达到标准。达到预设标准后，蜂窝产品的力学性能符合翼面、舱 面、舱盖、地板、发动机护罩、尾喷管、消音板、隔热板、卫星星体外壳、刚性太阳电池翼、抛物面天线、火箭推进剂贮箱箱底等产品的要求。
其中，最大偏差角：即所有的偏差角的最大值
平均偏差角：即所有的偏差角的平均值；
胞元偏差角平均值的最大值：先计算单个胞元的各偏差角的平均值，为此胞元的偏差角平均值，所有胞元的偏差角平均值的最大值即为胞元偏差角平均值；
偏差角的标准差：以所有的偏差角为样本，计算得到偏差角的标准差；
胞元偏差角平均值的标准差：以所有的胞元偏差角的平均值为样本，计算得到胞元偏差角平均值的标准差；
本实施例的可选实施方案中，顶点提取的方法包括：轮廓线夹角法、分支点方法、HARRIS方法和移动窗口法。
其中，轮廓线夹角法包括：
获取蜂窝产品的骨架图；
从所述骨架图中提取出属于同一个胞元的边界像素点，得到所述胞元的所述边界像素点序列；
分别获取与所述边界像素点对应的夹角，得到与所述边界像素点对应的所述夹角序列；所述夹角为在所述边界像素点序列中，在所述边界像素点的两侧，分别与所述边界像素点相隔第一预设值个像素的两个像素点与所述胞元边界像素点连线的夹角；
采用第二预设值大小的窗口，对所述夹角序列进行非极小值抑制，将除了所述窗口中夹角的最小值之外的夹角设置为第一预设夹角；
当完成所述夹角序列的所述非极小值抑制时，确定所有所述最小值中小于第二预设夹角的夹角对应的所述边界像素点为所述顶点。
分支点方法，包括：获取顶点图像的骨架图；在骨架图上每个像素点的八邻域按照预设顺序搜寻一周，若像素值变化次数为6，则像素点为顶点。具体来说，首先以形态图像为基础，将像素值为1的线条采用线宽为1个像 素的线段绘制成骨架图；其次是以具有k个像素点的骨架图为基础，遍历像素点1至像素点k，每当遇到像素值＝1的像素点时，以顺时针或逆时针方向绕该像素点的八邻域搜寻一周，得到一个像素值变化次数；若像素值变化次数为4，显示过该像素点存在两根直线，通过坐标计算得知这两根直线存在合理的夹角时，则确定该像素点为边缘顶点并记录；若像素值变化次数为6，显示过该像素点存在三根直线，则确定该像素点为中间顶点并记录，顶点提取完毕。其中，形态图像是由预处理得到的。
HARRIS方法：首先是以形态图像为基础，将像素值为1的线条采用线宽为1个像素的线段绘制骨架图；第二步是在骨架图的基础上，以像素值＝1的像素点处为中心点建立尺寸为5×5个像素的窗口，如果该窗口的部分区域溢出骨架图时，先将溢出区域的像素点的像素值全部赋值为0，然后采用Harris算法计算出该中心点对应的角点响应函数值R；在同一窗口内的全部的R值中，将小于最大的R值的1％的像素点处的R值置为零，重复以上操作遍历整个骨架图；下一步，以像素值＝1且R值大于零的像素点处为中心点建立尺寸为3×3个像素的窗口，若该中心点的R值为本窗口内的最大值，则记录该点为顶点，重复以上操作遍历整个骨架图，顶点提取完毕。其中，形态图像是由预处理得到的。
移动窗口法，包括：S1：获取预设边长L的正方形窗口，正方形窗口遍历顶点图像后，正方形窗口内的像素值为0的像素数的最小值为非零；其中，获取预设边长L的正方形窗口，包括：以形态图像为对象，设置一个边长能够从小变大变化的正方形窗口，当某一边长的窗口遍历形态图像后，如果窗口内的像素值＝0的像素数的最小值为非零时，将该正方形窗口的边长定义为壁厚L；
S2：以正方形窗口遍历顶点图像中每个像素值为1的像素点，然后将每个正方形窗口内的像素值为1的像素个数的总和赋值给正方形窗口的中心点的像素点；
S3：以最大赋值的像素点为中心建立初始边长E＝L的正方形区域，计算 出正方形区域四个边界上各像素点的赋值与正方形区域中心点的赋值的差值绝对值，记录最小的差值绝对值Z及其对应的边界像素点的坐标；
S4：以E＝E+2个像素建立新的正方形区域，重复S3，直到最小的差值绝对值有明显的反向增大的趋势为止，以此时取得最小的差值绝对值Z的边界像素点坐标与对应的正方形区域中心点的坐标，通过Z值得边界像素点坐标和对应的正方形区域中心点的坐标的坐标值求解得到蜂窝胞元边长A；其中，明显的反向增大的趋势是指，最小的差值绝对值会先减小再增大。
S5：将最大赋值的像素点，确定为顶点，再以顶点为中心，建立一个以蜂窝胞元边长A为边长的正方形的湮灭窗口，将湮灭窗口内所有像素值＝1的像素点上的赋值全部清零，在此基础上，再在剩余的赋值中再找到最大赋值的像素点确定为顶点并记录，重复湮灭窗口的操作，直到像素值＝1的像素点上的赋值小于给定阈值为止，提取所有顶点。其中，给定阈值与具体的蜂窝壁厚度L大小有关，没有一个统一的值，阈值选取的原则是将图像中最大的赋值与蜂窝壁中心像素点对应的赋值取平均值。
本实施例的可选实施方案中，胞元重构的方法包括：邻域窗口递推方法。邻域窗口递推方法包括：采用摩尔邻域追踪算法追踪各胞元的边界；以追踪到的每个边界点为中心，建立一个预设大小的窗口，若窗口内有顶点，则记录该顶点的编号并存储在该胞元名下，并根据所遇顶点的先后顺序将胞元的顶点排序，完成追踪后将各胞元的顶点按顺序连线，得到胞元重构图。
具体来说，胞元重构指的是建立胞元与顶点的映射关系，即哪几个顶点同属于一个胞元，并据此将顶点连线，重新得到蜂窝结构的图像，所用方法为邻域窗口递推方法：采用摩尔邻域追踪算法追踪各胞元的边界，以追踪到的每个边界点为中心，建立一个合适大小的窗口，若窗口内有顶点，则记录该顶点的编号并存储在该胞元名下，并根据所遇顶点的先后顺序将该胞元的顶点排序，完成追踪后将各胞元的顶点按顺序连线，得到胞元重构图。
具体来说，是步骤“重构胞元”包括边缘扩展和顶点连线；步骤“边缘扩展”是以形态图像为基础，在形态图像四个边的最外缘，均向外扩展至少 1个像素的宽度形成扩展区，扩展区内所有像素点的像素值全部置为1，得到扩展图像；步骤“顶点连线”是以扩展图像为对象，以从左至右、自上而下的顺序遍历该扩展图像，当遇到像素值＝0的像素点时，就采用摩尔邻域跟踪算法寻找并记录同一胞元的顶点及其连接顺序并记录在该胞元的名下，以每个胞元最多保留六个顶点为原则，删除重复的记录，按照保留的记录进行顶点连线，画出完整的胞元；随后将该胞元内所有像素点的像素值全部置为1；在此基础上，寻找下一个像素值＝0的像素点，重复以上操作，遍历完成的同时也完成了胞元重构图。
实施例2
图3是本发明的实施例2的蜂窝产品的几何形态检测及评估系统俯视结构示意图；图4是本发明的实施例2的蜂窝产品的几何形态检测及评估系统测试结构示意图。
如图4和5所示，本实施例提供了一种蜂窝产品的几何形态检测及评估系统，包括：
置物台1，用于放置蜂窝产品2；
相机3，设置于蜂窝产品2的顶面和侧面，用于对蜂窝产品2进行拍照，得到蜂窝产品2原始顶面图像和原始侧面图像；
相机移动装置4，用于将相机3在蜂窝产品2的顶面和侧面移动，以得到完整的原始顶面图像和完整的原始侧面图像；
分析评估模块5，用于获取完整的原始顶面图像和完整的原始侧面图像，并进降噪滤波、二值化和形态学滤波处理，得到顶面图像和侧面图像，；对顶面图像进行顶点提取，得到顶点坐标；对顶面图像进行胞元重构，得到各胞元的六个顶点的序号编号；基于各胞元的六个顶点的序号编号和各胞元六个顶点的顶点坐标，计算各胞元六个内角的偏差值；从预处理后的顶面图像和侧面图像中提取蜂窝产品2的顶面边线和侧面边线；基于顶面边线和侧面边线计算蜂窝产品2的顶面最大挠度和侧面最大挠度；基于偏差角、顶面最大挠度和侧面最大挠度判定蜂窝产品2是否合格。
本发明通过获取蜂窝产品的顶面及侧面图像判定蜂窝产品的几何形态是否合格，以非接触式的方式，通过对图像进行简单处理，以准确判断蜂窝产品2的几何形态是否合格。
本实施例的分析评估模块5还用于实现实施例1的全部内容，此处不再多做赘述。
本实施例的可选实施方案中，相机移动装置4包括：行走式龙门架41，相机3与行走式龙门架41连接；滑轨42，沿蜂窝产品2的长度方向设置，行走式龙门架41设置于滑轨42上，沿滑轨42移动。
本实施例的可选实施方案中，相机移动装置4还包括：相机3滑动组件。相机移动组件与行走式龙门架41滑动连接，相机3设置于相机移动组件上，相机3通过机移动组件在行走式龙门架41上滑动。
本实施例的可选实施方案中，相机移动组件包括：滚轮43和移动块44，移动块44套设于行走式龙门架41上，通过滚轮43与行走式龙门架41连接，以使移动块44在行走式龙门架41上滑动，相机3与移动块44可拆卸连接，通过移动块44的移动，来获取蜂窝产品2完整的原始顶面图像和原始侧面图像。可选地，蜂窝产品的几何形态检测及评估系统还包括夹具6，用于固定蜂窝产品2在置物台1上的位置，以便相机3拍照。
本实施例的可选实施方案中，夹具6由四块平板及驱动装置组成，可在驱动装置的作用下向蜂窝产品2靠拢，靠紧被测蜂窝产品2后锁死，用于定位被测蜂窝产品2。
本实施例的可选实施方案中，夹具6靠近蜂窝产品2一侧均涂成亮黄色，以便图像中清楚的确定蜂窝产品2的边界，以辅助图像处理。
本实施例的可选实施方案中，蜂窝产品2的几何形态检测及评估系统还包括置物台1升降装置7，用于调节蜂窝产品2的高度，以便蜂窝产品2与设置于蜂窝产品2顶面的相机3在适宜的距离，以便相机3拍照。进一步可选地，可以通过升降装置7将蜂窝产品2的上平面与夹具6的上平面保持在同一高度，夹具6与相机3的垂直距离不变，进一步保持蜂窝产品2的上平 面与相机3保持的固定距离，以便相机3拍照。具体地，升降装置7包括液压杆。
本实施例的可选实施方案中，蜂窝产品2的几何形态检测及评估系统还包括显示模块，显示模块与分析评估模块5连接，用于显示蜂窝产品2的几何形态是否合格。
本实施例的可选实施方案中，显示模块包括显示灯，若蜂窝产品2合格，则亮绿灯，若不合格则亮红灯。
本实施例的可选实施方案中，蜂窝产品2的几何形态检测及评估系统还包括标定模块：标定板为一个采用电子墨水屏的显示板，可显示边长、壁厚可调的标准蜂窝，屏幕外侧显示与蜂窝成对比色的颜色。将该标定板放置于置物台16并用夹具6定位后，调整相机3至合适位置，获取该标定板的照片，传递给分析评估模块5进行标定，校核系统的检测准确性。
应当理解的是，本发明的上述具体实施方式仅仅用于示例性说明或解释本发明的原理，而不构成对本发明的限制。因此，在不偏离本发明的精神和范围的情况下所做的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。此外，本发明所附权利要求旨在涵盖落入所附权利要求范围和边界、或者这种范围和边界的等同形式内的全部变化和修改例。

Claims (12)

  1. 一种蜂窝产品的几何形态评估方法，其特征在于，包括：
    获取蜂窝产品的顶面图像和侧面图像；
    对所述顶面图像进行顶点提取，得到顶点坐标；
    对所述顶面图像进行胞元重构，得到各胞元的六个顶点的序号编号；
    基于所述各胞元的六个顶点的序号编号和所述各胞元六个顶点的顶点坐标，计算各胞元六个内角的偏差值；
    从所述顶面图像和所述侧面图像中提取所述蜂窝产品的顶面边线和侧面边线；
    基于所述顶面边线和所述侧面边线计算所述蜂窝产品的顶面最大挠度和侧面最大挠度；
    基于所述偏差角、所述顶面最大挠度和所述侧面最大挠度判定所述蜂窝产品是否合格。
  2. 根据权利要求1所述的蜂窝产品的几何形态评估方法，其特征在于，所述基于所述偏差角、所述顶面最大挠度和所述侧面最大挠度判定所述蜂窝产品是否合格，包括：
    基于所述偏差角、所述顶面最大挠度和所述侧面最大挠度判定所述蜂窝产品的几何规整度，若所述几何规整度达到预设标准，则所述蜂窝产品合格。
  3. 根据权利要求2所述的蜂窝产品的几何形态评估方法，其特征在于，所述几何规整度包括：顶面胞元规整度、侧面平直度和顶面平直度；
    所述基于所述偏差角、所述顶面最大挠度和所述侧面最大挠度判定所述蜂窝产品的几何规整度，若所述几何规整度达到预设标准，则所述蜂窝产品合格，包括：
    基于所述偏差角判定所述顶面胞元规整度；
    基于所述顶面最大挠度判定所述侧面平直度；
    基于所述侧面最大挠度判定所述顶面平直度；
    若所述顶面胞元规整度、所述侧面平直度和所述顶面平直度均达到预设标准，则所述蜂窝产品合格。
  4. 根据权利要求3所述的蜂窝产品的几何形态评估方法，其特征在于，所述顶面胞元规整度的判定指标包括：最大偏差角、平均偏差角、胞元偏差角平均值的最大值、偏差角的标准差和胞元偏差角平均值的标准差中的一种或多种；
    所述最大偏差角、所述平均偏差角、所述胞元偏差角平均值的最大值、所述偏差角的标准差和所述胞元偏差角平均值的标准差中的一种或多种均小于预设阈值，则所述顶面胞元规整度达到预设标准。
  5. 根据权利要求3所述的蜂窝产品的几何形态评估方法，其特征在于，所述顶面胞元规整度的判定指标包括：最大偏差角、平均偏差角、胞元偏差角平均值的最大值、偏差角的标准差和胞元偏差角平均值的标准差中的一种或多种；
    所述最大偏差角、所述平均偏差角、所述胞元偏差角平均值的最大值、所述偏差角的标准差和所述胞元偏差角平均值的标准差中的一种或多种大于预设阈值的比例小于预设比例，则所述顶面胞元规整度达到预设标准。
  6. 根据权利要求3所述的蜂窝产品的几何形态评估方法，其特征在于，所述顶面胞元规整度的判定指标包括：胞元偏差角平均值；
    若所述胞元偏差角平均值依次不大于A1、A2、A3、A4和A5的胞元所占整体胞元的百分比，依次小于相应阈值B1、B2、B3、B4和B5，则所述顶面胞元规整度达到标准。
  7. 根据权利要求1所述的蜂窝产品的几何形态评估方法，其特征在于，所述顶点提取的方法包括：
    获取蜂窝产品的骨架图；
    从所述骨架图中提取出属于同一个胞元的边界像素点，得到所述胞元的所述边界像素点序列；
    分别获取与所述边界像素点对应的夹角，得到与所述边界像素点对应的所述夹角序列；所述夹角为在所述边界像素点序列中，在所述边界像素点的两侧，分别与所述边界像素点相隔第一预设值个像素的两个像素点与所述胞元边界像素点连线的夹角；
    采用第二预设值大小的窗口，对所述夹角序列进行非极小值抑制，将除了所述窗口中夹角的最小值之外的夹角设置为第一预设夹角；
    当完成所述夹角序列的所述非极小值抑制时，确定所有所述最小值中小于第二预设夹角的夹角对应的所述边界像素点为所述顶点。
  8. 根据权利要求1所述的蜂窝产品的几何形态评估方法，其特征在于，所述顶点提取的方法包括：
    获取顶点图像的骨架图；
    在所述骨架图上每个像素点的八邻域按照预设顺序搜寻一周，若像素值变化次数为6，则所述像素点为顶点。
  9. 根据权利要求1所述的蜂窝产品的几何形态评估方法，其特征在于，所述顶点提取的方法包括：
    S1：获取预设边长L的正方形窗口，设置一个边长从3个像素依次增大的正方形窗口，所述正方形窗口遍历所述顶点图像后，所述正方形窗口内的像素值为0的像素数的最小值首次为非零；
    S2：以所述正方形窗口遍历所述顶点图像中每个像素值为1的像素点，然后将每个所述正方形窗口内的像素值为1的像素个数的总和赋值给所述正方形窗口的中心点的像素点；
    S3：以最大赋值的像素点为中心建立初始边长E＝L的正方形区域，计算出所述正方形区域四个边界上各像素点的赋值与正方形区域中心点的赋值的差值绝对值，记录最小的差值绝对值Z及其对应的边界像素点的坐标；
    S4：以E＝E+2个像素建立新的正方形区域，重复S3，直到最小的差值绝对值有明显的反向增大的趋势为止，以此时取得最小的差值绝对值Z的边界像素点坐标与对应的正方形区域中心点的坐标，通过所述Z值得边界像素点坐标和所述对应的正方形区域中心点的坐标的坐标值求解得到蜂窝胞元边长A；
    S5：将最大赋值的像素点，确定为顶点，再以所述顶点为中心，建立一个以蜂窝胞元边长A为边长的正方形的湮灭窗口，将所述湮灭窗口内所有像素值＝1的像素点上的赋值全部清零，在此基础上，再在剩余的赋值中再找到最大赋值的像素点确定为顶点并记录，重复湮灭窗口的操作，直到像素值＝1的像素点上的赋值小于给定阈值为止，提取所有顶点。
  10. 根据权利要求1所述的蜂窝产品的几何形态评估方法，其特征在于，所述胞元重构的方法包括：
    采用摩尔邻域追踪算法追踪各胞元的边界；
    以追踪到的每个边界点为中心，建立一个预设大小的窗口，若窗口内有顶点，则记录该顶点的编号并存储在该胞元名下，并根据所遇顶点的先后顺序将胞元的顶点排序，完成追踪后将各胞元的顶点按顺序连线，得到胞元重构图。
  11. 一种蜂窝产品的几何形态检测及评估系统，其特征在于，包括如权利要求1-10任一项所述的蜂窝产品的几何形态评估系统，还包括：
    置物台(1)，用于放置蜂窝产品(2)；
    相机(3)，设置于所述蜂窝产品(2)的顶面和侧面，用于对所述蜂窝产品(2)进行拍照，得到所述蜂窝产品(2)原始顶面图像和原始侧面图像；
    相机移动装置(4)，用于将所述相机(3)在所述蜂窝产品(2)的顶面和侧面移动，以得到完整的原始顶面图像和完整的原始侧面图像；
    分析评估模块(5)，用于获取所述完整的原始顶面图像和所述完整的原始侧面图像，并进降噪滤波、二值化和形态学滤波处理，得到顶面图像 和侧面图像；对所述顶面图像进行顶点提取，得到顶点坐标；对所述顶面图像进行胞元重构，得到各胞元的六个顶点的序号编号；基于所述各胞元的六个顶点的序号编号和所述各胞元六个顶点的顶点坐标，计算各胞元六个内角的偏差值；从所述预处理后的顶面图像和侧面图像中提取所述蜂窝产品(2)的顶面边线和侧面边线；基于所述顶面边线和所述侧面边线计算所述蜂窝产品(2)的顶面最大挠度和侧面最大挠度；基于所述偏差角、所述顶面最大挠度和所述侧面最大挠度判定所述蜂窝产品(2)是否合格。
  12. 根据权利要求20所述的蜂窝产品的几何形态检测及评估系统，其特征在于，所述相机移动装置(4)包括：
    行走式龙门架(41)，所述相机(3)与所述行走式龙门架(41)连接；
    滑轨(42)，沿所述蜂窝产品的长度方向设置，所述行走式龙门架(41)设置于所述滑轨(42)上，沿所述滑轨(42)移动。
PCT/CN2020/109731 2020-05-09 2020-08-18 蜂窝产品的几何形态评估方法和检测及评估系统 WO2021227288A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/422,721 US11893725B2 (en) 2020-05-09 2020-08-18 Method for evaluating and system for detecting and evaluating geometric form of honeycomb product

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN202010388343.5 2020-05-09
CN202010387864.9 2020-05-09
CN202010387864.9A CN111583234B (zh) 2020-05-09 2020-05-09 蜂窝产品的几何形态评估方法和检测及评估系统
CN202010388343.5A CN111583238B (zh) 2020-05-09 2020-05-09 蜂窝的轮廓线夹角顶点提取方法、检测方法及装置

Publications (1)

Publication Number Publication Date
WO2021227288A1 true WO2021227288A1 (zh) 2021-11-18

Family

ID=78525961

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/109731 WO2021227288A1 (zh) 2020-05-09 2020-08-18 蜂窝产品的几何形态评估方法和检测及评估系统

Country Status (2)

Country Link
US (1) US11893725B2 (zh)
WO (1) WO2021227288A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114905229A (zh) * 2022-04-21 2022-08-16 北京九天行歌航天科技有限公司 一种一体式无主焊缝火箭贮箱箱底结构及其制备方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101915545A (zh) * 2010-08-13 2010-12-15 南京柯瑞特种陶瓷股份有限公司 一种蜂窝陶瓷自动检测方法与装置
CN107123108A (zh) * 2017-03-30 2017-09-01 上海卫星装备研究所 基于机器视觉的遥感器遮光罩蜂窝缺陷自动检测方法
CN110111331A (zh) * 2019-05-20 2019-08-09 中南大学 基于机器视觉的蜂窝纸芯缺陷检测方法
CN110232388A (zh) * 2019-06-11 2019-09-13 大连理工大学 一种从蜂窝芯表面测量数据中识别蜂窝边的方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4095768B2 (ja) * 2000-11-24 2008-06-04 株式会社日立製作所 画像処理方法及び、それを用いた非接触画像入力装置
JP4944057B2 (ja) * 2008-03-28 2012-05-30 日本碍子株式会社 ハニカム構造体
DE112016001723T5 (de) * 2015-10-06 2018-01-04 Ngk Insulators, Ltd. Verfahren zum Kontrollieren einer Keramikkörperoberfläche

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101915545A (zh) * 2010-08-13 2010-12-15 南京柯瑞特种陶瓷股份有限公司 一种蜂窝陶瓷自动检测方法与装置
CN107123108A (zh) * 2017-03-30 2017-09-01 上海卫星装备研究所 基于机器视觉的遥感器遮光罩蜂窝缺陷自动检测方法
CN110111331A (zh) * 2019-05-20 2019-08-09 中南大学 基于机器视觉的蜂窝纸芯缺陷检测方法
CN110232388A (zh) * 2019-06-11 2019-09-13 大连理工大学 一种从蜂窝芯表面测量数据中识别蜂窝边的方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
CUI CAN; WANG ZHONGGANG; ZHOU WEI; WU YONGJUN; WEI WANHUI: "Branch point algorithm for structural irregularity determination of honeycomb", COMPOSITES PART B, ELSEVIER, AMSTERDAM, NL, vol. 162, 19 October 2018 (2018-10-19), AMSTERDAM, NL, pages 323 - 330, XP085596185, ISSN: 1359-8368, DOI: 10.1016/j.compositesb.2018.10.062 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114905229A (zh) * 2022-04-21 2022-08-16 北京九天行歌航天科技有限公司 一种一体式无主焊缝火箭贮箱箱底结构及其制备方法

Also Published As

Publication number Publication date
US20230214986A1 (en) 2023-07-06
US11893725B2 (en) 2024-02-06

Similar Documents

Publication Publication Date Title
CN102171554B (zh) 用于分析基底的表面的装置
CN105335973B (zh) 运用于带钢加工生产线的视觉处理方法
CN110415300B (zh) 一种基于三靶标建面的立体视觉结构动态位移测量方法
CN106600593B (zh) 一种中铝瓷球表面缺陷检测方法
WO2021227288A1 (zh) 蜂窝产品的几何形态评估方法和检测及评估系统
CN108665458A (zh) 透明体表面缺陷提取及识别方法
CN106338277B (zh) 一种基于基线的建筑物变化检测方法
CN109523539A (zh) 基于多相机阵列的大尺寸的工业板材在线测量系统及方法
CN114219842A (zh) 港口集装箱自动装卸作业中的视觉识别、测距与定位方法
CN117288099B (zh) 基于机器视觉的铝塑复合板生产质量检测分析方法
CN111524134B (zh) 生产线上蜂窝产品规整度的检测方法及检测装置
CN115931898A (zh) 一种陶瓷基板表面缺陷视觉检测方法、装置及存储介质
Guan et al. Defect detection method for specular surfaces based on deflectometry and deep learning
CN111583234B (zh) 蜂窝产品的几何形态评估方法和检测及评估系统
CN111583237B (zh) 蜂窝规整度检测的邻域窗口递推胞元重构方法
Zhang et al. Defect inspection of coated automobile roofs using a single camera
CN111583239B (zh) 蜂窝结构几何规整度图像识别方法及系统
CN111583241B (zh) 超大面积蜂窝产品规整度的移动式检测方法及装置
WO2021227286A1 (zh) 超大面积蜂窝产品规整度的移动式检测方法及装置
CN106548474A (zh) 一种微结构表面检测方法
CN114187269B (zh) 小零器件表面缺陷边缘快速检测方法
WO2021227285A1 (zh) 蜂窝结构几何规整度图像识别方法及系统
CN111583243A (zh) 蜂窝规整度检测的相邻点胞元重构方法
CN111524133B (zh) 蜂窝规整度检测的窗口统计顶点提取方法及其系统
CN114565596A (zh) 基于深度学习和视频理解的钢表面裂缝检测和预测方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20936038

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20936038

Country of ref document: EP

Kind code of ref document: A1