Classifications

• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T5/00—Image enhancement or restoration
  • G06T5/77—Retouching; Inpainting; Scratch removal
• • 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
  • G06T7/0008—Industrial image inspection checking presence/absence
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T7/00—Image analysis
  • G06T7/90—Determination of colour characteristics
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
  • G01N21/84—Systems specially adapted for particular applications
  • G01N21/88—Investigating the presence of flaws or contamination
  • G01N21/89—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles
  • G01N21/8914—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles characterised by the material examined
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
  • G01N21/84—Systems specially adapted for particular applications
  • G01N21/88—Investigating the presence of flaws or contamination
  • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
• • 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
  • G06T2207/10—Image acquisition modality
  • G06T2207/10028—Range image; Depth image; 3D point clouds
• • 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 field of tire manufacturing, and more particularly the field of automatic inspection of the surface of a tire in order to establish a conformity diagnosis with respect to pre-established references.
• One of the steps of this process consists, in known manner, in acquiring the three-dimensional image of the surface of the tire. [003]
• the acquisition of this image is done using means based on the principle of optical triangulation, implementing for example a 2D sensor coupled to a laser-type lighting source.
• the topographic image of the tire surface is in fact a two-dimensional image, called a gray level, in which, at any point, ie at any pixel (x, y) of the image, a value is associated with f (x, y), called gray level, and usually between 0 and 255.
• This gray level value can usefully be coded on 8, or 16 bits for better dynamics.
• the gray level represents the altitude of this point relative to the surface. For 8-bit encoding, the value 255 (white) corresponds to the highest altitude, and the value 0 (black) corresponds to the lowest altitude.
• the pixels of the image are arranged in line and in columns.
• the method of processing the three-dimensional digital image of the surface The purpose of a pneumatic tire is to detect and eliminate these points of non-measurement in order not to disturb the subsequent digital processing intended for example to identify the anomalies present on the surface of the tire.
• the image processing method provides the steps in which:
• areas of the surface to be inspected are searched, comprising pixels whose gray level value is below a given threshold
• each pixel whose gray level value is less than a given threshold is assigned a gray level value equal to the average gray level value of a set of pixels references located in an area in the immediate vicinity of the pixel in question.
• the method may provide that the set constituting the reference pixels is formed by a reference matrix comprising an odd number of rows and columns.
• the reference matrix comprises less than ten rows and less than ten columns.
• each point of the bounding box is modified by assigning to a given point the average value of gray level of the pixels of the reference matrix centered on said point.
• the points of the bounding box are successively processed in the increasing order of the rows and columns.
• the area of the pixel area having a gray level value below the given threshold is determined
• the zone is oriented in a direction OX extending over the main axis of the zone and originating from the foot of the vent,
• a reference segment is arranged disposed on the side of the main axis of said zone corresponding to the angular sector forming a positive angle g with the direction of the shadow zone,
• the average gray level value of the pixels of the reference segment is assigned to all the pixels of the pixel. the line containing said reference segment and lying between the middle of the reference segment and the intersection of said pixel line with the contour of the vent.
• FIGS. 1 and 2 show a schematic view of a means of capturing the image of the surface of a tire
• FIG. 3 schematically illustrates the causes causing the appearance of the points of non-measurement
• FIG. 4 represents the partial image of the surface of a tire having points of non-measurement
• FIG. 5 illustrates the image treatments used by the method according to the invention
• FIGS 6 and 7 illustrate the possible configurations of image processing in the vicinity of the vents
• FIG. 8 represents the image of FIG. 4 after treatment using the method according to the invention.
• the acquisition of the image of the surface of a tire is illustrated by FIGS. 1 and 2. This acquisition is effected by way of example using a slot light emitted by a laser 1 and a camera 3 capable of capturing the 2D image of the illuminated surface. The camera is positioned so that its aiming direction makes a given angle with the beam emitted by the laser source. By triangulation, it is then possible to determine the coordinates of the relief element 2 relative to the support surface 4.
• the slot light is directed in an axial or radial direction perpendicular to the circumferential direction corresponding to the direction the rotation imposed on the tire to capture a complete image of its surface.
• Figure 1 shows a configuration in which the device captures the image of a relief element 2
• Figure 2 the configuration in which the relief element is a vent.
• the non-measurement points 6 appear in the shadow areas not illuminated by the laser beam. In the absence of a luminous return, these points are considered as having a zero or very low altitude and appear in black on the image representing the surface.
• FIG. 3 illustrates the representation, in section, of the relief of FIG. 1 in which the measured relief appears in continuous line, while the real profile appears in dashed line.
• the location of the non-measurement points can be done simply by looking on the image pixels, whose gray level value, representing the altitude of the corresponding point relative to the surface to be inspected, is below a threshold given or even zero.
• FIG. 4 is a gray level image of the surface of a tire. which in contrast appear areas comprising non-measurement points, easily identifiable in that their gray level is zero or much lower than a predetermined threshold value. In this case, these points come from the shadow carried by a vent 5 (in the center of the image) or points along a relief (to the right of the image). It is observed that the shape and the area of these zones are different in nature, respectively large and elongated for the shadow of the vent, and small and of granular shape for the edge of a relief.
• the first step of the treatment after identifying the points of non-measurement, consists in framing the zones containing these points by a bounding box.
• said box may advantageously be of rectangular shape, and be aligned with the rows and columns of pixels forming the image to be processed.
• FIG. 5 makes it possible to visualize the zones containing non-measurement points surrounded by a bounding box 6, comprising n lines (h, l 2 , I 3 , U) and p columns (ci, c 2 , ... c p ).
• the further processing consists in assigning to each of the non-measurement points a gray level value equal to the average of the gray level value of the pixels situated in the immediate vicinity of the pixel in question. It is admitted that the point of no measurement is located at substantially the same altitude as its direct neighbors with whom they are aligned.
• a first mode of treatment consists in assigning to one pixel the average value of the pixels situated directly around it. This mode of treatment is particularly advantageous for correcting the non-measurement points included in areas of small area as illustrated on the right side of the image of FIG. 5.
• a set of reference pixels formed by a reference matrix K comprising an odd number of rows and columns, and whose central value is positioned on the pixel to be modified, is determined.
• the processing then consists in averaging the gray level values of the points of the matrix, including the central value, and replacing the gray level value of the central value with the newly calculated average value.
• This operation is preferably slippery, starting with the pixel located, for example, at the top left of the bounding box, and treating column by column in the manner indicated above all the points of the upper line. We then go down one line and repeat the same operation on all the pixels of this line up to the pixel of the last line and the last column included in the bounding box. [028] It will be observed that the calculated averages successively replace the pixels whose gray level value is lower than the predetermined threshold and that these averages calculated in turn in the determination of the gray level value of the following pixels, where the recursive nature of this moving average over the entire bounding box. [029] In order to obtain a stable result, the limits of the bounding box will preferably be placed so that the pixels representing the non-measurement points closest to these boundaries are arranged a few pixels apart.
• the second mode of treatment is of particular interest to the resulting non-measurement points of the shadow carried by a vent on the surface of the tire.
• the shape of the shadow 6 is generally highly elongated and occupies a relatively large area facing the surface of the previously described areas. This hypothesis is verified by ensuring that the area of the shadow zone is greater than a predetermined threshold, and the center of gravity W of the surface formed by the non-measurement points is located.
• the following operations are specific to this type of anomaly, while also considering that, in the case of a tire, these areas can not be confused with other types of zones containing points of no measured.
• the foot of the vent 5 is then searched at one of the two ends of the main axis of the zone containing the non-measurement points. This makes it possible to give an orientation to the main axis in a direction OX extending along the zone and originating from the foot of the vent, as shown in FIG.
• the next step is to determine the reference pixels that will be used to calculate the average value of gray level to be assigned to the non-measurement points.
• the reference pixels are constituted, in this second case, by reference segments s, each contained in a line 1, and comprising about twenty pixels.
• each of the lines I it is then appropriate for each of the lines I, to position said reference segment judiciously. Indeed, preferably, it is arranged for this segment I, is disposed on the side of the main axis 60 corresponding to the angular sector at a positive angle g with the direction OX of the shadow zone 6. In other words, it is preferable to position the reference segments on the opposite side to the general direction of the projection of the vent on the surface, so as to allow, as will be seen later, a complete treatment of all points in the area containing the non-measurement points.
• the average gray level value of the pixels of the reference segment s is assigned to all the pixels of the line I containing said reference segment and included between the medium m of the reference segment s, and the intersection of said line of pixels with the contour of the vent.
• the pixel lines intersect the zone of non-measurement points.
• Each line I intersects said zone 6 at two points.
• a first point of intersection is located on the edge of said zone 6 on the side where the reference segment s is located.
• the second point of intersection is located opposite to it on the opposite edge of this area of non-measurement points, and corresponds substantially to the point of intersection of the line comprising said reference segment with the foot of the vent whose presence is at the origin of the shadow zone corresponding to the non-measurement points.
• the first part contains the lines situated between the center of gravity W and the foot of the vent 5.
• the average value of gray level of the pixels of the reference segment s is then assigned to all the pixels of line I, containing said reference segment s, and between the middle m of the reference segment and the intersection of said pixel line with the contour of the foot of the vent.

Landscapes

• Engineering & Computer Science (AREA)
• Physics & Mathematics (AREA)
• General Physics & Mathematics (AREA)
• Theoretical Computer Science (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Quality & Reliability (AREA)
• Image Analysis (AREA)
• Length Measuring Devices By Optical Means (AREA)
• Image Processing (AREA)
• Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Applications Claiming Priority (2)

Publications (1)

Family

ID=45888207

Family Applications (1)

Country Status (7)

Families Citing this family (10)

Family Cites Families (19)

• 2011
  • 2011-04-18 FR FR1153343A patent/FR2974218A1/fr not_active Withdrawn
• 2012
  • 2012-03-21 US US14/112,452 patent/US9224198B2/en not_active Expired - Fee Related
  • 2012-03-21 EP EP12710714.2A patent/EP2700051A1/fr not_active Withdrawn
  • 2012-03-21 WO PCT/EP2012/055015 patent/WO2012143197A1/fr active Application Filing
  • 2012-03-21 JP JP2014505557A patent/JP2014513793A/ja active Pending
  • 2012-03-21 BR BR112013023977A patent/BR112013023977A2/pt not_active IP Right Cessation
  • 2012-03-21 CN CN201280019094.6A patent/CN103493095A/zh active Pending

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events