WO2011030360A1 - Machine for automatic detection of flaws on pieces made of leather or similar material - Google Patents

Machine for automatic detection of flaws on pieces made of leather or similar material Download PDF

Info

Publication number
WO2011030360A1
WO2011030360A1 PCT/IT2009/000407 IT2009000407W WO2011030360A1 WO 2011030360 A1 WO2011030360 A1 WO 2011030360A1 IT 2009000407 W IT2009000407 W IT 2009000407W WO 2011030360 A1 WO2011030360 A1 WO 2011030360A1
Authority
WO
WIPO (PCT)
Prior art keywords
piece
face
scanning
machine
flaws
Prior art date
Application number
PCT/IT2009/000407
Other languages
French (fr)
Inventor
Matteo Mascetti
Original Assignee
Camoga S.P.A.
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
Application filed by Camoga S.P.A. filed Critical Camoga S.P.A.
Priority to PCT/IT2009/000407 priority Critical patent/WO2011030360A1/en
Priority to TW099129651A priority patent/TW201115138A/en
Publication of WO2011030360A1 publication Critical patent/WO2011030360A1/en

Links

Classifications

    • 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
    • C—CHEMISTRY; METALLURGY
    • C14—SKINS; HIDES; PELTS; LEATHER
    • C14B—MECHANICAL TREATMENT OR PROCESSING OF SKINS, HIDES OR LEATHER IN GENERAL; PELT-SHEARING MACHINES; INTESTINE-SPLITTING MACHINES
    • C14B1/00—Manufacture of leather; Machines or devices therefor
    • C—CHEMISTRY; METALLURGY
    • C14—SKINS; HIDES; PELTS; LEATHER
    • C14B—MECHANICAL TREATMENT OR PROCESSING OF SKINS, HIDES OR LEATHER IN GENERAL; PELT-SHEARING MACHINES; INTESTINE-SPLITTING MACHINES
    • C14B17/00—Details of apparatus or machines for manufacturing or treating skins, hides, leather, or furs
    • C14B17/005—Inspecting hides or furs
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/44—Resins; Plastics; Rubber; Leather
    • G01N33/447—Leather

Definitions

  • the present invention relates to a machine for automatic detection of flaws on pieces made of leather or similar material, to be used in such production sectors as, for instance, manufacture of articles of clothing (shoes, bags and so on) and industrial articles (saddlery and inner parts of cars/boats, sofas, furniture and so on) .
  • the preliminary selection of the piece of leather from which possible various lengths of material can be obtained must therefore identify possible anomalies either in the colour (stains due to dyeing errors or to diseases of the animal, bleached regions or others) or in the surface finish (abrasions, cuts, holes and so on) : generally, this preliminary selection is carried out manually or, to be more exact, by an operator personally examining the material to be cut or to be sent to subsequent working operations.
  • This operating method has some unavoidable drawbacks, above all connected with the fact that a lot of time is required for performing this selection, and with the intrinsic arbitrariness in the operator's judgement who has to decide on "where" and "which type of flaw” is to be found and highlighted on the piece of material (which judgement is in turn influenced by the operator's personal experience).
  • a further drawback connected with a manual examination of the pieces of hide and/or leather resides in that, even if the above mentioned flaws can be detected, it becomes very difficult, through a manual operation, to establish optimisation of the cuts to be carried out on the piece of material in a reasonable period of time, in order to minimise the production of scraps and/or prepare the semifinished products adapted to be sent to the next steps of the production chain.
  • the piece of material to be examined is provided to be merely laid down on a surface and this may give rise to formation of bulged regions and/or localised raised areas on the. piece of material (which typically can be caused by the inner tension generated in the piece itself) ; these variations in shape and/or type are not necessarily flaws because they can be "eliminated” during the following working steps, but most likely the automatic optical methods of known type will classify them as "flaws" and therefore will decide upon elimination of same, which is clearly a waste of leather or similar material.
  • the structure for detection of big areas generally has an important bulkiness; in fact the optics imposes the observance of great focal distances (for instance beyond 600 mm if images of one-meter-large leather pieces are wished to be acquired) .
  • a further defect typical of known devices resides in that the image processing software on which they are generally based comprises a single operating "logic", so that this programming rigidity does not possess the same efficiency in identifying flaws having different types of visual appearance (let us think of the visual differences .between a longitudinal cut and a stain, for example) ; likewise, the software programs so far used are based on a standardised parameterisation of the type of leather being examined, and therefore within a given "predetermined typology" of the piece of material, the programs of known type are not able to identify, in an independent manner, the type of hide and/or leather that is being submitted to scanning.
  • the present invention therefore aims at devising a machine for automatic detection of flaws on pieces made of leather or similar material that is able to obviate the above mentioned limits.
  • the present invention aims at devising a machine for automatic detection of flaws on pieces made of leather or similar material having a high efficiency in identifying any type of surface flaw, with high resolution and as much as possible reduced scanning times .
  • the present invention further aims at devising a machine that can operate on different types of leather, recognising them in advance and in an automatic manner and spontaneously identifying the most important aesthetic-visual parameters of same.
  • the present invention aims at devising a machine enabling a substantial elimination of "false flaws" due to pleats or folds and/or localised raised areas on the pieces to be examined.
  • FIG. 1 shows a diagrammatic side view of a machine according to the invention.
  • Fig. 2 is a diagrammatic view of a detail of the machine seen in Fig. 1.
  • the machine according to the present invention has been identified with reference numeral 1 and substantially comprises a support element 2 (that, depending on the circumstances, can be flat or conveniently shaped following any geometry, or it may either consist of a supporting surface or be provided with feeding means to be activated on the piece of material itself) .
  • This support element 2 is adapted to receive at least one piece of leather S, while suitable scanning means 3 (described in more detail in the following) is adapted to detect an image of at least one first face SI of this piece S and to convert it into a so-called “analysable signal” that can be in an analogue or digital format (by suitable conversion) .
  • first face SI the face or surface of the piece of material S that faces (or at all events can be “inspected” or examined by) the scanning means 3, which means in turn defines an optical scanning line directed towards the first face SI.
  • processing means In order to enable automation of the process for recognition and localisation of flaws in a piece S, processing means (usually consisting of electronic apparatus provided with suitable software and/or firmware programs) are conveniently present, which means is adapted to process said analysable signal and to identify at least one anomaly (or in a quite similar manner, identify any thing that can be defined as "flaw" on the piece S) that can be correlated with a portion of the first face SI of the piece S itself.
  • the present machine further comprises supporting and centring means 4 to be activated on a second face S2 of the piece S (which is opposite to the first face SI), which means is also adapted to determine localised tensioning of the piece S at least at the scanning line 3a optically in alignment with said scanning means 3.
  • the present invention contemplates an innovatory and different interfacing mode between the scanning means and the piece of material; this interfacing is studied in such a manner as to provide at least one ideal line along which the deformation and/or stretching and/or curling up effects, as well as possible variations in height and/or surface tension (that do not necessarily indicate the presence of flaws in the piece) are eliminated from a geometric and/or optical point of view; in this way, the machine 1 according to the present invention will be able to eliminate the "false reading of flaws" in a systematic manner, thus, optimising the analysis of the piece (on which only "true” flaws or defects will be identified) and therefore increasing the production efficiency.
  • the presence of the supporting and centring means 4 not only allows elimination of the above mentioned “false flaws", but also simultaneously enables amplification of those flaws that from an optical point of view could be of difficult detection; for instance, due to the localised “stretching” and “flattening” action that is performed by the supporting and centring means 4 along the scanning line 3a, possible cuts and/or notches that are present in the piece S can be emphasised; in fact, should the "localised tensioning" effect put into action by the supporting and centring means 4 give rise to mutual spacing apart of the edges of a cut or an abrasion, it would be the inner part of the piece of material to be in sight (i.e. exposed to the scanning means 3) , which part would have a colour and/or aspect quite different from that of surface SI.
  • the present invention contemplates the presence of means for "highlighting" flaws belonging at least to the first face SI; these highlighting means can therefore be coincident with the support and centring means and is adapted to highlight possible flaws (and in particular flaws resulting from cuts and/or abrasions and/or notches in the integrity of the first face SI of the piece S) through stretching and/or localised extension of said flaws, at least at a scanning line 3a that can be ideally defined on the first face SI of said piece S.
  • the supporting and centring means 4 first of all comprises a tensioning element 4a which has an operating surface 4b adapted to receive the piece of material S thereon (which piece is received along a feeding axis 4c) ; the tensioning element 4a defines the aforesaid scanning line 3a transverse and preferably perpendicular to the feeding axis 4c.
  • adhesion means 4d to be activated between the tensioning element 4a and the piece S to promote adhesion of the second face S2 on the tensioning element 4a itself.
  • this adhesion means 4d performs at least two functions at the same time: on the one hand, said means promotes adhesion of the piece S to the tensioning element 4a and simultaneously operates in such a manner as to flatten the piece so that, on the other hand, the effects of deviation from an "ideal geometric condition" for carrying out scanning are eliminated.
  • the adhesion means 4b can be put into practice on a machine for visual detection of flaws which may even not be equipped with the tensioning element 4a; in particular, it is possible to implement a machine in which the piece of leather is merely laid on a planar table which is provided with the adhesion means 4b (which is therefore adapted to keep the whole piece S under conditions of perfect adhesion, and therefore in a fully planar condition, relative to the surface of the table itself) .
  • the tensioning element 4a can consist of a cylindrical support engaged in rotation around an operating axis transverse (for instance, perpendicular) to the feeding axis 4c; in accordance with this geometry, the scanning line 3a is ideally defined by a line tangent to a side surface of the cylindrical support and is therefore parallel to the operating axis around which the cylindrical support rotates.
  • Adhesion between piece S and tensioning element 4a can be optimised by the optional presence of a presser roller 4e to be activated between the piece S and tensioning element 4a to press the piece S against the operating surface 4b.
  • a driving device acting on the cylindrical support is further present to set said cylindrical support in rotation and thus promote feeding of the piece S along the feeding axis 4c; in other words, the present invention contemplates the presence of feeding means 5 acting at least on the tensioning element 4a to promote movement of the piece S along the feeding axis 4c.
  • the adhesion means 4d comprises a suction unit to be activated on the second face S2 of the piece S; this suction unit is active in an inner axial cavity of the cylindrical support (which therefore has a substantially hollow conformation) , so as to pull and/or retain the piece S on the operating surface 4b of the cylindrical support.
  • said suction unit comprises a pneumatic circuit extending between two opposite ends of said inner axial cavity of the cylindrical support and interlocked with a vacuum pump 7 (or an equivalent machine) .
  • this means comprises at least one optical sensor in alignment with the cylindrical support; this optical sensor has a scanning axis aligned with the scanning line 3a defined by the cylindrical support so that the overall analysis of the first face SI is carried out by a method involving an analysis "by successive lines".
  • the optical sensor that can be used in the present machine is of the "CIS" type (wherein CIS is the acronym of "Contact Image Sensor”) .
  • the present invention contemplates the presence of backlighting means 6 which can be operatively activated in the inner axial cavity of the cylindrical support (or more generally, of the tensioning element 4a, which for structural and functional purposes can practically be coincident with the cylindrical support) and is adapted to generate a light beam directed to the second surface S2 of the piece S.
  • this back-lighting means 6 acts in cooperation with the scanning means 3 for identifying flaws and/or anomalies, and/or holes (or cuts, notches or others) passing through the surfaces SI and S2 of the piece S; this particular technical effect is made possible because the light beams emitted starting from the inner axial cavity of the cylindrical support find a "natural exit" in the cavities defined by the through holes/cuts, and can therefore reach the optical sensor 3a in an undisturbed manner thereby generating a very "clear" area of light that is uniquely connected with the presence of a discontinuity passing through the piece .
  • the technical effect consisting in "visual amplification" of the flaws with a structure "passing through” the surfaces SI and S2 of the piece S can be also implemented in a manner independent of the presence of the supporting and centring means 4; for instance, turning back to the already made example (according to which a planar table can be arranged which is possibly and optionally equipped with the adhesion means alone) , the presence of scanning means and back-lighting means can be provided, said means being disposed in opposite half-spaces respectively, which are divided by the surface of the table itself and face each other (so that possible rays or light beams emerging from the back-lighting means may be received by the scanning means) . It is a further aim of the present invention also to provide an innovative and original method of automatically detecting flaws on pieces made of leather or similar material, which comprises the following steps :
  • the aforesaid step of processing the analysable signal comprises the following sub-steps:
  • determining at least one "average identifier parameter" that can be correlated with a general type of said piece S (to the aims of the present invention, as "general type” it is intended the whole assembly of values and/or reference information on a given type of animal from which the piece S can be drawn and/or on visual surface features that can be, on average, ascribed to the piece S depending on its coming from a given type of animal);
  • this describing parameter will be correlated with a colour that can be locally ascribed to a scanning line and/or with surface features to be, on average, ascribed to the scanning area itself) ;
  • the method hitherto described can have a particular development of the step of determining the extension and/or localisation of a flaw; in particular, to the aims of the present invention (and for maximising accuracy and speed) , this operating step can advantageously comprise a sub-step consisting in carrying out a "selective choice" of one or more processing algorithms that can also be of an already known type.
  • this sub-step concerning the selective choice is correlated at least with the average identifier parameter and preferably comprises a sub- step consisting in establishing (or, in other words, adopting and loading in the operating memory of the processing system that will have to work on the "analysable signal) a given "strategy of intervention" on the analysable signal.
  • the present method can therefore establish use of:
  • the three alternative solutions discussed above can also be combined with each other, and therefore they can be performed following any time combination/sequence and/or in parallel, all that as a function of the available calculation power and/or the processing accuracy and/or the processing speed that one wishes to have at one's disposal.
  • the "preliminary" identification of the piece S is connected with the fact that the pieces of leather (or also the whole hide) can greatly vary as regards their texture; this variability can be seen both between the different types of animals (cattle, sheep, goats, saurian animals and others) , and within the same family of animals.
  • firmware part or also only the software part of the machine to comprise an appropriate series of "libraries" of processing programs, which are specifically developed for each type of hide or leather that is wished to be analysed; therefore if the operating capacity of the machine 1 is wished to be maximised, it will be necessary to provide it with the greatest possible number of libraries and therefore the greatest possible number of processing strategies.
  • each library is different from the others due both to the presence of one or more specific "filters", and to the possible presence of image enhancement procedures (carried out on some factors of the image itself) , and also due to the arrangement and pre-storage of specific tables containing and assigning values and/or parameters used for development of the filters themselves.
  • pattern matching is utilised in cascade, which technique is substantially based on segmentation of the detected image into elementary parts (named “textons” in current bibliography) , which are then compared with those stored in an experimentally created database.
  • the possible flaw found is thus highlighted, also through display on a monitor, and the Cartesian coordinates thereof are calculated with respect to a reference point of the piece of material; these coordinates can therefore be subsequently automatically treated for exclusion of this region of the piece that is concerned with the flaw, during a possible subsequent cutting step by means of cutting tables and/or semi-automatic shears (that are therefore drivable by suitable data processing systems for optimisation of the cuts in the piece S) .
  • the invention achieves many advantages.
  • the present invention is able to offer an interfacing mode between the piece of leather and the scanning means that is capable of obtaining a much higher accuracy than that obtainable with the presently known automated systems.
  • the present invention can be also interfaced with programs (and automated machines controlled by such programs) that can define an optimal die cutting operation (or a cutting path for automatic cutting tables) on the piece, so as to minimise the portions of leather (or similar material) to be rejected and to cut out as many semifinished patterns as possible from the piece itself .
  • programs and automated machines controlled by such programs
  • a further favourable point of the invention resides in that the particular structure used for positioning and analysing the piece allows a more efficient identification of possible through holes (or more generally, damages in the piece concerning both faces of the piece itself) , which is a clear progress relative to known methods.
  • the machine can be provided with capability of self-learning and independent recognition of the type of leather that at a given moment is submitted to scanning; this makes the preliminary work for setting/parameterisation that is required to an operator of the machine, much simpler and increases the overall production quality (reducing the possibility of errors due to a wrong identification of the type of leather) .
  • the present invention can have the possibility of selecting the most appropriate processing algorithm (or the most suitable sequence of one o more processing algorithms) in an independent manner, as a function of the detected type of leather and/or of other parameters that can be easily and quickly inputted by the machine's operator.
  • the present invention enables low manufacturing costs of the machine to be ensured and does not involve particular complications, or the necessity of important modifications or adaptations to be carried out on machines of known type, which is advantageous for the overall production economy and the final price of the product .

Landscapes

  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Mechanical Engineering (AREA)
  • Pathology (AREA)
  • Organic Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Manufacturing & Machinery (AREA)
  • General Physics & Mathematics (AREA)
  • Medicinal Chemistry (AREA)
  • Food Science & Technology (AREA)
  • Textile Engineering (AREA)
  • Treatment Of Fiber Materials (AREA)

Abstract

A machine for automatic detection of flaws on pieces made of leather or similar material comprises a support element adapted to receive at least one piece of leather, scanning means adapted to detect an image of a first face of the piece and to convert it into an analysable signal, and processing means adapted to process the analysable signal and to identify at least one anomaly that can be correlated with a portion of the first face of the piece; the machine further comprises supporting and centring means to be activated on a second face of the piece, opposite to the first face, and adapted to determine localised tensioning of the piece at least at a scanning line optically aligned with the scanning means.

Description

MACHINE FOR AUTOMATIC DETECTION OF FLAWS ON PIECES MADE OF LEATHER OR SIMILAR MATERIAL
D e s c r i p t i o n
The present invention relates to a machine for automatic detection of flaws on pieces made of leather or similar material, to be used in such production sectors as, for instance, manufacture of articles of clothing (shoes, bags and so on) and industrial articles (saddlery and inner parts of cars/boats, sofas, furniture and so on) .
It is known that production of articles made of hide and/or leather (or similar material) is based on a suitable selection of the "base" material, i.e. the pieces of hide and/or leather; in fact, these pieces are required to have a level of surface homogeneity (in addition to features of consistency and structural compactness) that must be in each case suitable for the type of goods that is wished to be manufactured.
The preliminary selection of the piece of leather from which possible various lengths of material can be obtained (which lengths are to be in turn differently shaped) must therefore identify possible anomalies either in the colour (stains due to dyeing errors or to diseases of the animal, bleached regions or others) or in the surface finish (abrasions, cuts, holes and so on) : generally, this preliminary selection is carried out manually or, to be more exact, by an operator personally examining the material to be cut or to be sent to subsequent working operations.
This operating method has some unavoidable drawbacks, above all connected with the fact that a lot of time is required for performing this selection, and with the intrinsic arbitrariness in the operator's judgement who has to decide on "where" and "which type of flaw" is to be found and highlighted on the piece of material (which judgement is in turn influenced by the operator's personal experience).
A further drawback connected with a manual examination of the pieces of hide and/or leather resides in that, even if the above mentioned flaws can be detected, it becomes very difficult, through a manual operation, to establish optimisation of the cuts to be carried out on the piece of material in a reasonable period of time, in order to minimise the production of scraps and/or prepare the semifinished products adapted to be sent to the next steps of the production chain.
As an alternative to a manual inspection or examination, methods involving automated optical detection of flaws have been recently introduced in which an optical sensor based on the "CCD" technology catches and processes an image of the piece of material and, based on an algorithm of comparison, confines areas of the piece where flaws of various nature are present .
While the new art briefly discussed above is quicker and more efficient than a manual inspection, it however has some operating drawbacks.
First of all, in the automatic devices of known type the piece of material to be examined is provided to be merely laid down on a surface and this may give rise to formation of bulged regions and/or localised raised areas on the. piece of material (which typically can be caused by the inner tension generated in the piece itself) ; these variations in shape and/or type are not necessarily flaws because they can be "eliminated" during the following working steps, but most likely the automatic optical methods of known type will classify them as "flaws" and therefore will decide upon elimination of same, which is clearly a waste of leather or similar material.
In addition, in the automatic detection devices the structure for detection of big areas generally has an important bulkiness; in fact the optics imposes the observance of great focal distances (for instance beyond 600 mm if images of one-meter-large leather pieces are wished to be acquired) . This involves an important vulnerability of the acquisition method to variations in the environmental work conditions (for example, light and vibrations that can heavily invalidate the detection operations).
The above mentioned drawbacks could be obviated by use of several CCD devices for sharing out the overall acquisition width. However, in this way there will be an increase in the complexity of the mechanical mounting, interfacing and integration between devices, which as a result will increase the manufacturing and servicing costs. A further defect typical of known devices resides in that the image processing software on which they are generally based comprises a single operating "logic", so that this programming rigidity does not possess the same efficiency in identifying flaws having different types of visual appearance (let us think of the visual differences .between a longitudinal cut and a stain, for example) ; likewise, the software programs so far used are based on a standardised parameterisation of the type of leather being examined, and therefore within a given "predetermined typology" of the piece of material, the programs of known type are not able to identify, in an independent manner, the type of hide and/or leather that is being submitted to scanning.
In the light of the above discussed known art, the present invention therefore aims at devising a machine for automatic detection of flaws on pieces made of leather or similar material that is able to obviate the above mentioned limits.
Mainly, the present invention aims at devising a machine for automatic detection of flaws on pieces made of leather or similar material having a high efficiency in identifying any type of surface flaw, with high resolution and as much as possible reduced scanning times .
The present invention further aims at devising a machine that can operate on different types of leather, recognising them in advance and in an automatic manner and spontaneously identifying the most important aesthetic-visual parameters of same.
Accordingly, the present invention aims at devising a machine enabling a substantial elimination of "false flaws" due to pleats or folds and/or localised raised areas on the pieces to be examined.
The technical task mentioned and the aims specified are substantially achieved by a machine for automatic detection of. flaws on pieces made of leather or similar material having the features reproduced in one or more of the appended claims.
Description of a preferred but not exclusive embodiment of a machine according to the invention is now given hereinafter, by way of non-limiting example, and illustrated in the accompanying drawings, in which:
- Fig. 1 shows a diagrammatic side view of a machine according to the invention; and
- Fig. 2 is a diagrammatic view of a detail of the machine seen in Fig. 1.
With reference to the drawings, the machine according to the present invention has been identified with reference numeral 1 and substantially comprises a support element 2 (that, depending on the circumstances, can be flat or conveniently shaped following any geometry, or it may either consist of a supporting surface or be provided with feeding means to be activated on the piece of material itself) .
This support element 2 is adapted to receive at least one piece of leather S, while suitable scanning means 3 (described in more detail in the following) is adapted to detect an image of at least one first face SI of this piece S and to convert it into a so-called "analysable signal" that can be in an analogue or digital format (by suitable conversion) .
To the aims of the present invention it is denoted as "first face SI" the face or surface of the piece of material S that faces (or at all events can be "inspected" or examined by) the scanning means 3, which means in turn defines an optical scanning line directed towards the first face SI.
In order to enable automation of the process for recognition and localisation of flaws in a piece S, processing means (usually consisting of electronic apparatus provided with suitable software and/or firmware programs) are conveniently present, which means is adapted to process said analysable signal and to identify at least one anomaly (or in a quite similar manner, identify any thing that can be defined as "flaw" on the piece S) that can be correlated with a portion of the first face SI of the piece S itself.
Advantageously, in order to ensure a high scanning accuracy, and elimination of "systematic errors" due to raising, accidental folding and/or inner tensions of the piece of material, the present machine further comprises supporting and centring means 4 to be activated on a second face S2 of the piece S (which is opposite to the first face SI), which means is also adapted to determine localised tensioning of the piece S at least at the scanning line 3a optically in alignment with said scanning means 3.
In other words, the present invention contemplates an innovatory and different interfacing mode between the scanning means and the piece of material; this interfacing is studied in such a manner as to provide at least one ideal line along which the deformation and/or stretching and/or curling up effects, as well as possible variations in height and/or surface tension (that do not necessarily indicate the presence of flaws in the piece) are eliminated from a geometric and/or optical point of view; in this way, the machine 1 according to the present invention will be able to eliminate the "false reading of flaws" in a systematic manner, thus, optimising the analysis of the piece (on which only "true" flaws or defects will be identified) and therefore increasing the production efficiency. It is also to be pointed out that the presence of the supporting and centring means 4 not only allows elimination of the above mentioned "false flaws", but also simultaneously enables amplification of those flaws that from an optical point of view could be of difficult detection; for instance, due to the localised "stretching" and "flattening" action that is performed by the supporting and centring means 4 along the scanning line 3a, possible cuts and/or notches that are present in the piece S can be emphasised; in fact, should the "localised tensioning" effect put into action by the supporting and centring means 4 give rise to mutual spacing apart of the edges of a cut or an abrasion, it would be the inner part of the piece of material to be in sight (i.e. exposed to the scanning means 3) , which part would have a colour and/or aspect quite different from that of surface SI.
Therefore, it is possible to see that the present invention contemplates the presence of means for "highlighting" flaws belonging at least to the first face SI; these highlighting means can therefore be coincident with the support and centring means and is adapted to highlight possible flaws (and in particular flaws resulting from cuts and/or abrasions and/or notches in the integrity of the first face SI of the piece S) through stretching and/or localised extension of said flaws, at least at a scanning line 3a that can be ideally defined on the first face SI of said piece S. In more detail as regards the structure of the invention, it is possible to see that the supporting and centring means 4 first of all comprises a tensioning element 4a which has an operating surface 4b adapted to receive the piece of material S thereon (which piece is received along a feeding axis 4c) ; the tensioning element 4a defines the aforesaid scanning line 3a transverse and preferably perpendicular to the feeding axis 4c.
Also present and included in the supporting and centring means 4 is adhesion means 4d to be activated between the tensioning element 4a and the piece S to promote adhesion of the second face S2 on the tensioning element 4a itself.
Within the scope of the present invention, it is possible to see that this adhesion means 4d performs at least two functions at the same time: on the one hand, said means promotes adhesion of the piece S to the tensioning element 4a and simultaneously operates in such a manner as to flatten the piece so that, on the other hand, the effects of deviation from an "ideal geometric condition" for carrying out scanning are eliminated.
It is to be noted that at the light of the just mentioned second technical effect, the adhesion means 4b can be put into practice on a machine for visual detection of flaws which may even not be equipped with the tensioning element 4a; in particular, it is possible to implement a machine in which the piece of leather is merely laid on a planar table which is provided with the adhesion means 4b (which is therefore adapted to keep the whole piece S under conditions of perfect adhesion, and therefore in a fully planar condition, relative to the surface of the table itself) . With reference to the accompanying figures, and for achieving an optimal compromise between compactness of the machine and accuracy in ensuring the "localised tensioning", the tensioning element 4a can consist of a cylindrical support engaged in rotation around an operating axis transverse (for instance, perpendicular) to the feeding axis 4c; in accordance with this geometry, the scanning line 3a is ideally defined by a line tangent to a side surface of the cylindrical support and is therefore parallel to the operating axis around which the cylindrical support rotates.
Adhesion between piece S and tensioning element 4a can be optimised by the optional presence of a presser roller 4e to be activated between the piece S and tensioning element 4a to press the piece S against the operating surface 4b.
From a functional point of view, a driving device acting on the cylindrical support is further present to set said cylindrical support in rotation and thus promote feeding of the piece S along the feeding axis 4c; in other words, the present invention contemplates the presence of feeding means 5 acting at least on the tensioning element 4a to promote movement of the piece S along the feeding axis 4c.
From a structural point of view, the adhesion means 4d comprises a suction unit to be activated on the second face S2 of the piece S; this suction unit is active in an inner axial cavity of the cylindrical support (which therefore has a substantially hollow conformation) , so as to pull and/or retain the piece S on the operating surface 4b of the cylindrical support.
In order to enable correct operation of the suction unit, and at the same time allow rotation of the cylindrical support, said suction unit comprises a pneumatic circuit extending between two opposite ends of said inner axial cavity of the cylindrical support and interlocked with a vacuum pump 7 (or an equivalent machine) .
Paying particular attention to the scanning means 3, it is possible to see that this means comprises at least one optical sensor in alignment with the cylindrical support; this optical sensor has a scanning axis aligned with the scanning line 3a defined by the cylindrical support so that the overall analysis of the first face SI is carried out by a method involving an analysis "by successive lines".
In order to maximise the scanning speed and rebuilding of the "virtual" image (or in other words, the "analysable signal" that is conveniently correlated with the surface aspect of the whole scanned area of the piece S) and in order to have the greatest possible image resolution, the optical sensor that can be used in the present machine is of the "CIS" type (wherein CIS is the acronym of "Contact Image Sensor") .
Advantageously, for improving efficiency in detection of flaws "passing through" the whole thickness of the piece (or also flaws that reach a great depth in the piece of material S starting from the first face SI, while not passing through the whole thickness) , the present invention contemplates the presence of backlighting means 6 which can be operatively activated in the inner axial cavity of the cylindrical support (or more generally, of the tensioning element 4a, which for structural and functional purposes can practically be coincident with the cylindrical support) and is adapted to generate a light beam directed to the second surface S2 of the piece S. Conveniently, this back-lighting means 6 acts in cooperation with the scanning means 3 for identifying flaws and/or anomalies, and/or holes (or cuts, notches or others) passing through the surfaces SI and S2 of the piece S; this particular technical effect is made possible because the light beams emitted starting from the inner axial cavity of the cylindrical support find a "natural exit" in the cavities defined by the through holes/cuts, and can therefore reach the optical sensor 3a in an undisturbed manner thereby generating a very "clear" area of light that is uniquely connected with the presence of a discontinuity passing through the piece .
In addition, the technical effect consisting in "visual amplification" of the flaws with a structure "passing through" the surfaces SI and S2 of the piece S can be also implemented in a manner independent of the presence of the supporting and centring means 4; for instance, turning back to the already made example (according to which a planar table can be arranged which is possibly and optionally equipped with the adhesion means alone) , the presence of scanning means and back-lighting means can be provided, said means being disposed in opposite half-spaces respectively, which are divided by the surface of the table itself and face each other (so that possible rays or light beams emerging from the back-lighting means may be received by the scanning means) . It is a further aim of the present invention also to provide an innovative and original method of automatically detecting flaws on pieces made of leather or similar material, which comprises the following steps :
- first of all, properly disposing a first face SI of a piece of material S;
- carrying out scanning, preferably along successive lines, of at least one portion of said first face SI, so as to obtain an "analysable signal" that practically virtually (or in terms of qualitative and/or quantitative information) rebuilds said surface SI; and
- processing this "analysable signal" so as to identify and/or localise at least one area of said first face (SI) having a flaw in terms of shape and/or colour and/or surface finish and/or structural integrity.
Advantageously, in the present method the aforesaid step of processing the analysable signal comprises the following sub-steps:
- first of all, determining at least one "average identifier parameter" that can be correlated with a general type of said piece S (to the aims of the present invention, as "general type" it is intended the whole assembly of values and/or reference information on a given type of animal from which the piece S can be drawn and/or on visual surface features that can be, on average, ascribed to the piece S depending on its coming from a given type of animal);
- assigning a general identification to the piece S through comparison between said "average identifier parameter" and a plurality of pre-stored average identifier parameters (which therefore are indicative of different types and/or origins of the pieces);
- ideally dividing the first face SI of the piece S into a plurality of scanning areas (that in turn can be defined by a multiplicity of successive scanning lines) ;
- determining at least one describing parameter for each of said scanning areas (for example, this describing parameter will be correlated with a colour that can be locally ascribed to a scanning line and/or with surface features to be, on average, ascribed to the scanning area itself) ; and
- determining an extension and/or localisation of at least one "flaw", through determination of at least one deviation between the "average identifier parameter" and the aforesaid "describing parameter" .
Conveniently, the method hitherto described (and claimed in the following) can have a particular development of the step of determining the extension and/or localisation of a flaw; in particular, to the aims of the present invention (and for maximising accuracy and speed) , this operating step can advantageously comprise a sub-step consisting in carrying out a "selective choice" of one or more processing algorithms that can also be of an already known type. In more detail, this sub-step concerning the selective choice is correlated at least with the average identifier parameter and preferably comprises a sub- step consisting in establishing (or, in other words, adopting and loading in the operating memory of the processing system that will have to work on the "analysable signal) a given "strategy of intervention" on the analysable signal.
Depending on the decisions taken "upstream", and m particular depending on the type of piece that has been identified, thanks to the "average identifier parameter", the present method can therefore establish use of:
- a single processing algorithm;
- a sequence of two or more processing algorithms, that will then be executed "in cascade", starting from the data relating to the analysable signal;
- an execution in parallel of two or more processing algorithms .
Conveniently, the three alternative solutions discussed above can also be combined with each other, and therefore they can be performed following any time combination/sequence and/or in parallel, all that as a function of the available calculation power and/or the processing accuracy and/or the processing speed that one wishes to have at one's disposal.
From a practical point of view, it is to be pointed out that the "preliminary" identification of the piece S is connected with the fact that the pieces of leather (or also the whole hide) can greatly vary as regards their texture; this variability can be seen both between the different types of animals (cattle, sheep, goats, saurian animals and others) , and within the same family of animals.
At the same time, it is required that the great variability of findable flaws be recorded not only in terms of possible nature, conformation and "symptomatology" of the flaw, but also in terms of statistical recurrence of some types of flaws in relation to specific general types of leather; for instance, flaws in terms of desquamation are very likely to occur for pieces of hides of saurian animals (crocodiles, snakes and so on) while flaws in terms of surface cuts or notches frequently occur in bovine hides . Taking into due account this great variability, it is therefore suitable for the so-called firmware part (or also only the software part) of the machine to comprise an appropriate series of "libraries" of processing programs, which are specifically developed for each type of hide or leather that is wished to be analysed; therefore if the operating capacity of the machine 1 is wished to be maximised, it will be necessary to provide it with the greatest possible number of libraries and therefore the greatest possible number of processing strategies.
Taking into account what has been hitherto discussed, it is therefore possible to note that each library is different from the others due both to the presence of one or more specific "filters", and to the possible presence of image enhancement procedures (carried out on some factors of the image itself) , and also due to the arrangement and pre-storage of specific tables containing and assigning values and/or parameters used for development of the filters themselves.
In particular, in some libraries the so-called "pattern matching" technique is utilised in cascade, which technique is substantially based on segmentation of the detected image into elementary parts (named "textons" in current bibliography) , which are then compared with those stored in an experimentally created database.
The possible flaw found is thus highlighted, also through display on a monitor, and the Cartesian coordinates thereof are calculated with respect to a reference point of the piece of material; these coordinates can therefore be subsequently automatically treated for exclusion of this region of the piece that is concerned with the flaw, during a possible subsequent cutting step by means of cutting tables and/or semi-automatic shears (that are therefore drivable by suitable data processing systems for optimisation of the cuts in the piece S) .
The invention achieves many advantages.
In fact, due to the particular construction architecture of the above described machine (as hereinafter claimed) it is possible to obtain an optimal positioning of the piece to be analysed, eliminating variations in the shape and/or geometry that could be detected as "flaws" but that in fact are not so; in other words, the present invention is able to offer an interfacing mode between the piece of leather and the scanning means that is capable of obtaining a much higher accuracy than that obtainable with the presently known automated systems.
On the other hand, it is to be noted that the present invention can be also interfaced with programs (and automated machines controlled by such programs) that can define an optimal die cutting operation (or a cutting path for automatic cutting tables) on the piece, so as to minimise the portions of leather (or similar material) to be rejected and to cut out as many semifinished patterns as possible from the piece itself . A further favourable point of the invention resides in that the particular structure used for positioning and analysing the piece allows a more efficient identification of possible through holes (or more generally, damages in the piece concerning both faces of the piece itself) , which is a clear progress relative to known methods.
From the point of view of the process to be implemented by the machine of the invention, it is possible to notice that the machine can be provided with capability of self-learning and independent recognition of the type of leather that at a given moment is submitted to scanning; this makes the preliminary work for setting/parameterisation that is required to an operator of the machine, much simpler and increases the overall production quality (reducing the possibility of errors due to a wrong identification of the type of leather) . At the same time, still from the point of view of the process, it is to be pointed out that the present invention can have the possibility of selecting the most appropriate processing algorithm (or the most suitable sequence of one o more processing algorithms) in an independent manner, as a function of the detected type of leather and/or of other parameters that can be easily and quickly inputted by the machine's operator.
Finally, it will be recognised that the present invention enables low manufacturing costs of the machine to be ensured and does not involve particular complications, or the necessity of important modifications or adaptations to be carried out on machines of known type, which is advantageous for the overall production economy and the final price of the product .

Claims

C L A I M S
1. A machine for automatic detection of flaws on pieces made of leather or similar material, comprising:
- a support element (2) adapted to receive at least one piece of leather (S) ;
- scanning means (3) adapted to detect an image of at least one first face (SI) of said piece (S) and to convert it into an analysable signal; and
- processing means adapted to process said analysable signal and to identify at least one anomaly that can be correlated with a portion of said first face (SI) of the piece (S) ,
characterised in that it further comprises supporting and centring means (4) to be activated on a second face (S2) of the piece (S) , opposite to said first face (SI), which means is adapted to determine localised tensioning of the piece (S) at least at a scanning line (3a) optically aligned with the scanning means (3) .
2. A machine as claimed in claim 1, wherein said supporting and centring means (4) comprises:
- a tensioning element (4a) having an operating surface (4b) adapted to receive the piece (S) thereon, along a feeding axis (4c), said tensioning element (4a) defining said scanning line (3a) in a transverse and preferably perpendicular direction to the feeding axis (4c) ; and
adhesion means (4d) to be activated between the tensioning element (4a) and the piece (4) for promoting an adhesion of the second face (S2) to the tensioning element ( 4a ) .
3. A machine as claimed in claim 2, wherein said tensioning element (4a) comprises a cylindrical support engaged in rotation around an operating axis transverse and preferably perpendicular to the feeding axis (4c), the scanning line (3a) being more preferably ideally defined by a line tangent to a side surface of said cylindrical support, which line is parallel to said operating axis.
4. A machine as claimed in claim 2, wherein said adhesion means (4d) comprises a suction unit to be activated on the second face (S2) of the piece (S) and preferably acting in an inner axial cavity of the cylindrical support for drawing and/or retaining the piece (S) on the operating surface (4b) of the cylindrical support, said suction unit more preferably comprising a pneumatic circuit extending between two opposite ends of said inner axial cavity of the cylindrical support and being interlocked with a vacuum pump .
5. A machine as claimed in anyone of the preceding claims, wherein the scanning means (3) comprises at least one optical sensor aligned with said cylindrical support, and preferably having a scanning axis aligned with the scanning line (3a) defined by the cylindrical support itself.
6. A machine as claimed in claim 5, wherein said optical sensor is of the Contact Image Sensor (CIS) type .
7. A machine as claimed in anyone of the preceding claims, wherein also present is feeding means (5) acting at least on the tensioning element for promoting movement of the piece (S) along the feeding axis (4c).
8. A machine as claimed in anyone of the preceding claims, wherein also present is back-lighting means (6) to be operatively activated in the inner axial cavity of the cylindrical support and adapted to generate a light beam directed to the inner surface (S2) of the piece (S) , said back-lighting means (6) acting in cooperation with the scanning means (3) for identifying flaws and/or anomalies and/or through holes passing through the surfaces (SI, S2) of the piece (S) .
9. A method for automatic detection of flaws on pieces of leather or similar material, comprising the following steps:
- properly disposing a first face (SI) of a piece (S) ; - carrying out scanning, preferably along successive lines, of at least one portion of said first face (SI) , so as to obtain an analysable signal; and
- processing said analysable signal so as to identify and/or localise at least one area of said first face (SI) having a flaw in terms of shape and/or colour and/or surface finish and/or structural integrity, characterised in that the step of processing the analysable signal comprises the following sub-steps:
- determining at least one average identifier parameter that can be correlated with a general type of the piece
(S) , said general type preferably comprising values and/or reference information on a given type of animal from which the piece (S) can be drawn and/or on visual surface features that can be, on average, ascribed to the piece (S) depending on its coming from a given type of animal;
- assigning a general identification to the piece (S) through comparison of said at least one average identifier parameter with a plurality of pre-stored average identifier parameters; - ideally diyiding the first face (SI) of the piece (S) into a plurality of scanning areas;
- determining at least one describing parameter for each of said scanning areas, said describing parameter being preferably correlated with a colour that can be locally ascribed to a scanning line and/or with surface features to be, on average, ascribed to said scanning area; and
- determining an extension and/or localisation of at least one "flaw", through determination of at least one deviation between said average identifier parameter and said at least one describing parameter.
10. A method as claimed in claim 9, characterised in that said step of determining an extension and/or a localisation of a flaw comprises a sub-step of selectively choosing one or more processing algorithms, said step of selectively choosing being correlated at least with the average identifier parameter and preferably comprising a sub-step of establishing a single processing algorithm and/or establishing a sequence of two or more processing algorithms and/or establishing an execution in parallel of two or more processing algorithms, the method according to claims 8 and/or 9 being more preferably implemented through a machine for automatic detection of flaws on pieces of leather or similar material in accordance with anyone of the preceding claims 1 to 8.
PCT/IT2009/000407 2009-09-11 2009-09-11 Machine for automatic detection of flaws on pieces made of leather or similar material WO2011030360A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/IT2009/000407 WO2011030360A1 (en) 2009-09-11 2009-09-11 Machine for automatic detection of flaws on pieces made of leather or similar material
TW099129651A TW201115138A (en) 2009-09-11 2010-09-02 Machine for automatic detection of flaws on pieces made of leather or similar material

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/IT2009/000407 WO2011030360A1 (en) 2009-09-11 2009-09-11 Machine for automatic detection of flaws on pieces made of leather or similar material

Publications (1)

Publication Number Publication Date
WO2011030360A1 true WO2011030360A1 (en) 2011-03-17

Family

ID=42203445

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IT2009/000407 WO2011030360A1 (en) 2009-09-11 2009-09-11 Machine for automatic detection of flaws on pieces made of leather or similar material

Country Status (2)

Country Link
TW (1) TW201115138A (en)
WO (1) WO2011030360A1 (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016006007A1 (en) * 2014-07-10 2016-01-14 Camoga S.P.A. Device for detection, marking and cutting of leather
ITUA20163300A1 (en) * 2016-05-10 2017-11-10 Luxury Brands Control S R L METHOD OF CLASSIFICATION OF THE CHOICE OF A LEATHER.
WO2021240279A1 (en) * 2020-05-29 2021-12-02 Conceria Pasubio S.P.A. Method and apparatus for identifying possible surface defects of a leather hide
WO2022053906A1 (en) * 2020-09-10 2022-03-17 Brevetti Cea Spa Inspecting machine for leather and inspection method applicable to said machine
CN114395883A (en) * 2022-01-13 2022-04-26 江正荣 Automatic feeding system for textile bleaching
IT202200013669A1 (en) 2022-06-28 2023-12-28 Cea Brevetti Spa LEATHER INSPECTION MACHINE

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102519981A (en) * 2011-12-16 2012-06-27 湖南工业大学 Online detection system for PVC (polyvinyl chloride) building material surface quality
US10297018B2 (en) * 2017-07-14 2019-05-21 Lear Corporation Method of digitally grading leather break
CN116106319A (en) * 2023-02-14 2023-05-12 浙江迈沐智能科技有限公司 Automatic detection method and system for defects of synthetic leather

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE4216469A1 (en) * 1992-05-19 1993-11-25 Diehl Gmbh & Co Defect classification system for skin to be used to make leather - analyses infrared radiation distribution transmitted through tanned stretched skin in wet-blue state and passed between heat-radiating quartz lamps and CCD or line camera.
US6157730A (en) * 1996-02-09 2000-12-05 Roever; Detlef E. Visual inspection system for leather hide
EP1081235A2 (en) * 1999-09-02 2001-03-07 Enea - Ente Per Le Nuove Tecnologie, L'energia E L'ambiente Method and device for measuring the elasticity of hides

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE4216469A1 (en) * 1992-05-19 1993-11-25 Diehl Gmbh & Co Defect classification system for skin to be used to make leather - analyses infrared radiation distribution transmitted through tanned stretched skin in wet-blue state and passed between heat-radiating quartz lamps and CCD or line camera.
US6157730A (en) * 1996-02-09 2000-12-05 Roever; Detlef E. Visual inspection system for leather hide
EP1081235A2 (en) * 1999-09-02 2001-03-07 Enea - Ente Per Le Nuove Tecnologie, L'energia E L'ambiente Method and device for measuring the elasticity of hides

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016006007A1 (en) * 2014-07-10 2016-01-14 Camoga S.P.A. Device for detection, marking and cutting of leather
CN107075589A (en) * 2014-07-10 2017-08-18 凯莫加股份公司 The equipment of detection, mark and cutting for leather
ITUA20163300A1 (en) * 2016-05-10 2017-11-10 Luxury Brands Control S R L METHOD OF CLASSIFICATION OF THE CHOICE OF A LEATHER.
WO2021240279A1 (en) * 2020-05-29 2021-12-02 Conceria Pasubio S.P.A. Method and apparatus for identifying possible surface defects of a leather hide
WO2022053906A1 (en) * 2020-09-10 2022-03-17 Brevetti Cea Spa Inspecting machine for leather and inspection method applicable to said machine
CN114395883A (en) * 2022-01-13 2022-04-26 江正荣 Automatic feeding system for textile bleaching
CN114395883B (en) * 2022-01-13 2023-11-24 阿克苏嘉德纺织有限公司 Automatic feeding system for textile bleaching
IT202200013669A1 (en) 2022-06-28 2023-12-28 Cea Brevetti Spa LEATHER INSPECTION MACHINE

Also Published As

Publication number Publication date
TW201115138A (en) 2011-05-01

Similar Documents

Publication Publication Date Title
WO2011030360A1 (en) Machine for automatic detection of flaws on pieces made of leather or similar material
US6157730A (en) Visual inspection system for leather hide
EP1995553B1 (en) System and method for identifying a feature of a workpiece
US9008266B2 (en) Device for determining the quality of orthogonally pretensioned leather
JP6789292B2 (en) How and equipment to inspect tires
JP2020510814A (en) Equipment to detect and inspect tire defects at the end of the manufacturing process
RU2764644C1 (en) Method for detecting surface defects, device for detecting surface defects, method for producing steel materials, method for steel material quality control, steel materials production plant, method for generating models for determining surface defects and a model for determining surface defects
EP3537136A1 (en) Image inspecting apparatus, image inspecting method and image inspecting program
WO2020169456A1 (en) Automatic equipment for performing a perimeter finish cut of semi-finished leather sheets
CN110231352B (en) Image inspection apparatus, image inspection method, and image inspection recording medium
KR20060055337A (en) Method of inspecting wafer
EP2212909B1 (en) Patterned wafer defect inspection system and method
KR20120089541A (en) Method and system for wafer registration
CN115035092A (en) Image-based bottle detection method, device, equipment and storage medium
WO2008144717A1 (en) Automated detection of leather hide and flexible material defects
JPH09152322A (en) Method and device for surface quality inspection
JP2020012808A (en) Detection method of leather by artificial intelligence, leather detection equipment, and manufacturing method of leather product
US5353358A (en) Method for determining article orientation in an article handling system using a sampling of a CCD array
KR20050090168A (en) The auto-nesting apparatus with image processing
JP6049652B2 (en) Bottle bottom inspection device
CN111257422B (en) Wheel axle defect identification model construction method and defect identification method based on machine vision
JP2002365227A (en) Apparatus for inspecting pressed product
JP2009139299A (en) Emission analyzer
CN112147148A (en) Tire testing system
JP2002055057A (en) Inspection apparatus for cloth bar weaving

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: 09744487

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: 09744487

Country of ref document: EP

Kind code of ref document: A1