Method for estimating an adhesion grade between the rubber composition and the reinforcing plies of a test specimen representative of a tire to be characterized

Patent metadata

Field	Value
Publication number	US-2021181087-A1
Application number	US-201816758657-A
Country	US
Kind code	A1
Filing date	Oct 16, 2018
Priority date	Oct 23, 2017
Publication date	Jun 17, 2021
Grant date	—

Abstract

A method comprises steps in which: the specimen is baked under pressure then put into the expected format, a stripping test is carried out by peeling apart two reinforcing plies, at least one image is acquired of the two half-specimens obtained at the end of the stripping test, the acquired images are trimmed, a first image separating reinforcement regions from regions of rubber composition is then determined a second image is determined depending on the spacing and on the diameter of the reinforcing cords of the tire, among a theoretical image and a fitted theoretical image, and a score depending on the ratio of the area of the reinforcing cords in the first image to the area of the reinforcing cords in the second image is determined.

First claim

1 .- 12 . (canceled)

13 . A method for estimating a score of adhesion between a rubber composition and reinforcing plies of a specimen representative of a tire to be characterized, the specimen comprising at least two reinforcing plies embedded in the rubber composition, each reinforcing ply itself comprising reinforcing cords placed beside one another, the method comprising the following steps:
baking the specimen under pressure; then carrying out a stripping test by peeling apart two reinforcing plies of the specimen, thus resulting in two half-specimens; acquiring at least one image of the two half-specimens obtained at the end of the stripping test; trimming the acquired at least one image; determining a first image separating reinforcement regions from regions of rubber composition based on the at least one trimmed acquired image via superpixel decomposition, the first image being selected from a segmentation binary mask image, a folded binary mask image, a modified segmentation binary mask image, a modified folded binary mask image and an aligned folded binary mask image; determining a second image of the reinforcing cords based on a spacing and on a diameter of the reinforcing cords of the tire, the second image being selected from a theoretical image and a fitted theoretical image; and determining a score based on a ratio of an area of the reinforcing cords in the first image to an area of the reinforcing cords in the second image.

14 . The method according to claim 13 , wherein the at least one image of the two half-specimens is acquired with a flatbed scanner.

15 . The method according to claim 13 , wherein the at least one image of the two half-specimens is acquired with a flatbed scanner in conjunction with use of a rigid half-specimen holder configured to control a manner in which each half-specimen is pressed against a window of the flatbed scanner and an alignment in an acquisition region of the flatbed scanner.

16 . The method according to claim 13 , wherein trimming the at least one acquired image comprises:
removing the color components of the at least one acquired image so as to obtain a greyscale image; applying a greyscale threshold so that pixels having greyscales lower than the threshold are attributed a first Boolean value and pixels having greyscales higher than or equal to the threshold are attributed a second Boolean value, the greyscale threshold being selected so as to remove background image and preserve the at least one image of the two half-specimens; determining a binary mask of same size as the greyscale image, in which mask each pixel of the at least one image to be trimmed is associated with the first or second Boolean value; carrying out closing morphological filtering on the binary mask in order to join the two half-specimens; and carrying out opening morphological filtering on the filtered binary mask in order to remove noise.

17 . The method according to claim 13 , wherein the first image is a segmentation binary mask image, and
wherein determining the first image comprises:
applying a superpixel decomposition to the at least one trimmed image; and
then determining a segmentation binary mask containing superpixels having both an eccentricity higher than an eccentricity threshold and a brightness higher than a brightness threshold.

18 . The method according to claim 13 , wherein the second image is a fitted theoretical image, and
wherein determining the second image comprises:
carrying out an AND logic operation between a segmentation binary mask image and a theoretical image;
then measuring a deviation in position between the cords of the theoretical image and the cords of the segmentation binary mask image by taking a sum of pixels of non-zero Boolean value, the measurement of the deviation being repeated in an interval of values of pitch and of positions of the first cord that are defined beforehand by initial values;
retaining the values of pitch and of positions of the first cord associated with the highest value among the determined deviation values; and
determining a fitted theoretical image based on the retained values of pitch of the cords and of position of the first cord and on a diameter of the cords.

19 . The method according to claim 13 , wherein the first image is a modified segmentation binary mask image, and
wherein determining the first image comprises:
if a fitted theoretical image is not available, expanding a segmentation binary mask using a structuring element representing a horizontal line of width depending on a distance between two reinforcing cords and on a diameter of a reinforcing cord;
then carrying out an AND logic operation between a theoretical image of the reinforcing cords and the image obtained at the end of the expanding so as to obtain an expanded segmentation binary mask image; and
carrying out a linewise morphological reconstruction with, as marker, the expanded segmentation binary mask image and, as mask, the theoretical image in order to obtain the modified segmentation binary mask image; or
if a fitted theoretical image is available, expanding a segmentation binary mask using a structuring element representing a horizontal line of width depending on a distance between two reinforcing cords and on a diameter of a reinforcing cord;
carrying out an AND logic operation between a fitted theoretical image of the reinforcing cords and the image obtained at the end of the expanding so as to obtain an expanded segmentation binary mask image; and
carrying out a linewise morphological reconstruction with, as marker, the expanded segmentation binary mask image and, as mask, the fitted theoretical image in order to obtain a modified segmentation binary mask image.

20 . The method according to claim 13 , wherein the first image is a folded binary mask image, and
wherein determining the first image in order to avoid false positives comprises:
carrying out a horizontal mirroring operation on a lower portion of a segmentation binary mask image corresponding to one half-specimen; and
determining a folded binary mask image by carrying out an OR logic operation between each pixel of the upper half-mask and each pixel of the horizontal mirror of the lower half-mask, so as to superpose the pixels corresponding to reinforcing cords visible in the two half-specimens.

21 . The method according to claim 20 , wherein the second image is a fitted theoretical image, and
wherein determining the second image comprises:
carrying out an AND logic operation between the folded binary mask image and the theoretical image;
then measuring a deviation in position between the cords of the theoretical image and the cords of the folded binary mask image by taking a sum of pixels of non-zero Boolean value, the measurement of the deviation being repeated in an interval of values of pitch and of positions of the first cord that are defined beforehand by initial values;
retaining the values of pitch and of positions of the first cord associated with the highest value among the determined deviation values; and
determining a fitted theoretical image based on the retained values of pitch of the cords and of position of the first cord and on diameter of the cords.

22 . The method according to claim 20 , wherein the first image is a modified folded binary mask image, and
wherein determining the second image comprises:
if a fitted theoretical image is not available, expanding the folded binary mask image using a structuring element representing a horizontal line of width depending on a distance between two reinforcing cords and on a diameter of a reinforcing cord;
then carrying out an AND logic operation betwe

Assignees

• Michelin & Cie

Inventors

• Arvis Vincent
• Maginier Pierre

Classifications

• G06T2207/20112

  Image segmentation details · CPC title

• G06T7/0004 Primary

  Industrial image inspection · CPC title

• G06T7/62

  of area, perimeter, diameter or volume · CPC title

• G06T2207/30108

  Industrial image inspection · CPC title

• G06T7/10

  Segmentation; Edge detection (motion-based segmentation G06T7/215) · CPC title