DE102016012371A1 - Method and system for determining the defect surface of at least one defect on at least one functional surface of a component or test specimen - Google Patents

Abstract

The invention relates to determining the defect surface (DF) of at least one defect (7) on at least one functional surface Z (x, y) of a component or test body (1), comprising the method steps: the component or the test body (1) is in at least one Measuring range (2) of the functional surface Z (x, y) with at least one with respect to radiation intensity, direction and position defined light source (3) sequentially lit from at least four different positions (a, b, c, d), of the functional surface Z ( x, y) of the component or specimen (1) is at least one with a control and computer system (5) connected camera (4) known position respectively a different real image acquisition (raw image) (RR, R, R) of the functional surface Z (x, y) in the at least one measuring range (2) created by the control and computer system (5) are based on the Shape-from-Shading (SfS) method for the four real images (raw images) (RR, R, R) calculates the x- or y-tilt images (N, N) of the functional surface Z (x, y) from the functional surface Z (x, y) of the component or specimen (1) during a dark field illumination creating a dark-field image acquisition (D) of the functional surface Z (x, y) in the measurement area (2), and

Description

The invention relates to a method and a system for determining the defect surface of at least one defect on at least one functional surface of a component or specimen.

The testing of functional surfaces for damage such as e.g. Pores, scratches, dents, etc. are usually also in largely fully automated manufacturing processes usually by a personal visual inspection according to specified test specifications at the end of the manufacturing process. In this case, the person to be tested must be in good order (i.O.) or not in order (n.i.O.) in the production cycle about the quality. decide on a component.

Since, in particular in the automotive industry, the quality specifications prescribed by the test regulations are in the range of tenths of millimeters, such as, for example, given error sizes i.O. <500μm and n.i.O. > 500 μm, the human eye is only able to dissolve and reproduce such a size range unreliably. In addition, the extensive test specifications such. extremely complex and varied geometry of the surfaces with different evaluation ranges with respect to error sizes and types of overtaxing the visually inspecting person. Decisions about OK or NOK judgments are therefore made on the basis of estimated error magnitudes, subjective perception and experience, so that irregular slip, ie, the examiner does not check accurately enough and / or pseudo, ie, the examiner examines too closely , which can result in manufacturing, resulting in delays in the production process and additional costs.

Optical methods, in particular laser-based, generally require a relatively high calibration effort. A special camera-based optical method is the stereometry, in which the surface of a component or specimen to be examined is recorded from slightly different viewing angles and the structures of the surface of the component or specimen are calculated from the evaluation of the minor stereoscopic deviations.

Changes in the inclination angle course or even flat structures such as low depressions and / or slight elevations with low inclination can be determined by the method of shape-from-shading (see in particular X. Jiang, H. Bunke, Three-dimensional Computer Vision, Springer Verlag, 1997 Berlin ). In this case, small changes in the reflected light intensity are evaluated in order to close the known inclination of the reflective areas with a known geometric arrangement between the camera, component or specimen and light source.

From the DE 10 2004 038 761 B4 is a method for camera-based examination of objects, in particular of cast and forged parts, based on the method of shape-from-shading known, wherein the object to be tested is illuminated by at least one light source. In this case, at least one part of the part of the object surface illuminated by the light source is detected by means of at least one camera and the image data resulting therefrom are subjected to image processing. An evaluation with respect to the entire object surface does not take place here, but partial sections are selected within the image data image regions (ROI, regions of interest) and fed to a further processing in which the method of shape-from-shading (SfS) is applied to the image regions. The determination of an approximately real geometry of the defect surface of defects of the examined object surface is not possible in this known method.

Furthermore, goes from the DE 20317095 U1 a device for detecting defects of a surface of an object is known as known, in which a light source for illuminating the object with an illumination beam which falls on the object at an angle of incidence to the surface, and a light sensor are provided, one of which from the surface of the Object on the exposure beam towards reflected beam is detected, which is reflected at a projection angle to the surface of the object of the latter. In this case, the angle of reflection is equal to the angle of incidence in order to obtain image data which are to be evaluated by an evaluation unit in order to detect possible errors. A determination of the real area of defects of the inspected object surface is also not provided in this known method and is not considered.

The present invention is therefore based on the object of providing a method and a plant of the type mentioned above, with which or a determination of the real geometric features of at least one defect on at least one functional surface of a component or Prüfköpers is possible.

• - das Bauteil oder der Prüfkörper wird in mindestens einem Messbereich der Funktionsoberfläche mit mindestens einer hinsichtlich Strahlungsintensität, -richtung sowie - position definierten Lichtquelle sequentiell aus mindestens vier unterschiedlichen Positionen beleuchtet,
• - von der Funktionsoberfläche des Bauteils oder Prüfkörpers wird mit mindestens einer mit einem Steuer-und Rechnersystem verbundenen Kamera bekannter Position jeweils eine aus unterschiedlicher Richtung beleuchtete reale Bildaufnahme (Roh-Bild) der einen Funktionsoberfläche in dem Messbereich erstellt,
• - von dem Steuer-und Rechnersystem werden auf der Basis der Shape-from-Shading-Methode für die vier unterschiedlich beleuchteten realen Bildaufnahmen (Roh-Bilder) die Neigungsbilder in der x- bzw. y-Ableitung der Funktionsoberfläche berechnet,
• - von der Funktionsoberfläche des Bauteils oder Prüfkörpers wird zusätzlich zu den Rohbildern zur Neigungsbildberechnung während einer Dunkelfeldbeleuchtung noch eine Dunkelfeld-Bildaufnahme der Funktionsoberfläche in dem Messbereich erstellt, und
• - aus den zwei Neigungsbild-Berechnungen der Funktionsoberfläche und den Informationen der Dunkelfeld-Bildaufnahme wird die reale Defektfläche der mindestens einen Fehlstelle in dem Messbereich der Funktionsoberfläche des Bauteils oder Prüfkörpers mittels des Steuer-und Rechnersystems bestimmt.

This object is achieved according to the invention by a method for determining the defect surface of at least one defect on at least one functional surface of a component or test specimen, comprising the method steps:

• - The component or the test specimen is in at least one measuring range of the functional surface with at least one with respect to radiation intensity, direction and position light source sequentially lit from at least four different positions,
• from the functional surface of the component or test specimen, a real image image (raw image) of a functional surface in the measuring region, which is illuminated from different directions, is created with at least one camera connected to a control and computer system,
• from the control and computer system, the inclination patterns in the x- or y-derivative of the functional surface are calculated on the basis of the shape-from-shading method for the four differently illuminated real image recordings (raw images),
• - In addition to the raw images for tilt image calculation during dark field illumination of the functional surface of the component or specimen is still created a dark field image recording of the functional surface in the measuring range, and
• the real defect surface of the at least one defect in the measuring region of the functional surface of the component or test specimen is determined by means of the control and computer system from the two inclination image calculations of the functional surface and the information of the dark field image acquisition.

Preferably, the calculated inclination images are binarized, and the at least one defect on the at least one functional surface is detected from the binarized inclination images, wherein for a defect size in the range of tenths of millimeters, its geometry in the binarized inclination images approximately matches the real defect surface of the defect However, in the case that the size of partial areas of the defect lies within a relatively large area, these partial areas, which are reproduced only incompletely in the inclination images, are binarized via the darkfield image acquisition of the functional surface and combined with the binarization of the inclination images. From the combined binarization of the tilt images and the dark field image acquisition, the real geometry of the defect surface of the defect of the functional surface is then determined approximately accurately.

Preferably, four individual, in series reflectors are used as light sources, of which the functional surface is illuminated for at least one image acquisition series with four real image recordings (raw images) successively each with a defined exposure time.

As dark field lighting LED strips, preferably four LED strips, used, which are all connected together.

Preferably, three image acquisition series with four real image recordings (raw images) are created by the functional surface by means of a CCD camera, wherein the functional surface in each of the three image acquisition series is illuminated successively by the series-connected reflectors with a defined exposure time.

For inhomogeneous reflection properties, i. In a little to highly reflective functional surface of the component or specimen, the three image acquisition series with the four real images (raw images) are created with different exposure times of the four reflectors by the component or the specimen in at least one measuring range of the functional surface with the Reflectors defined in terms of radiation intensity, direction and position is illuminated sequentially from the four different positions.

Thus, the first image acquisition series with real first four images (raw images) with an exposure time of 0.01 ms,
the second image acquisition series with real four other images with an exposure time of 0.025 ms,
the third image acquisition series with another four real images with an exposure time of 0.05 ms, and
after the first, the second and the third image acquisition series, respectively, a dark field image acquisition for the respective determination of the defect surface of the defect can be created.

Instead of the dark field illumination, bright field illumination can be generated, during which a bright field image acquisition of the at least one functional surface in the at least one measurement range is created. Subsequently, from the inclination image calculations of the functional surface and the information of the bright field image acquisition, the real defect surface of the at least one defect in the at least one measurement region of the functional surface of the component or specimen is determined by means of the control and computer system.

After the inclination images have been calculated from the real image recordings (raw images) of the at least one functional surface in the at least one measuring region, a grayscale equalization can take place via a Savitzky-Golay filter.

Preferably, the at least one measuring range of the functional surface is divided into sectors, and a separate parameterization per pitch Image recording and per darkfield image acquisition or per brightfield image acquisition made.

Masking may be performed in the processing of the inclination image recordings and the darkfield image recordings or the bright field image recordings, wherein error detection is performed with realistic geometric information about the combination of information of the respective inclination image recordings and dark field and bright field image recordings and the detected errors are classified, displayed and / or stored.

Preferably, at least one high-resolution CCD camera with 29 or more megapixels is used for imaging.

• ein kastenförmiges Gehäuse, in dem entweder eine feste Lagerung oder eine verfahrbare Lineareinheit mit definierter Lagerung und Auflagefläche zur Aufnahme des Bauteils oder Prüfkörpers vorgesehen ist,
• mindestens eine hinsichtlich Strahlungsintensität, -richtung und -position definierte Lichtquelle, von der mindestens ein Messbereich der Funktionsoberfläche des in dem kastenförmigen Gehäuse auf der Auflagefläche der Lineareinheit positionierten Bauteils oder Prüfkörpers sequentiell aus mindestens vier unterschiedlichen Positionen zu beleuchten ist,
• mindestens eine Kamera bekannter Position, mittels der von der Funktionsoberfläche des Bauteils oder Prüfkörpers in dem Messbereich jeweils eine unterschiedliche reale Bildaufnahme (Roh-Bild) zu erstellen ist,
• eine in dem kastenförmigen Gehäuse positionierte Dunkelfeld-Beleuchtungseinrichtung, von der in dem Gehäuse eine Dunkelfeldbeleuchtung zu erzeugen ist, wobei von der Funktionsoberfläche des Bauteils oder Prüfkörpers während der Dunkelfeldbeleuchtung mittels der mindestens einen Kamera eine Dunkelfeld-Bildaufnahme der mindestens einen Funktionsoberfläche in dem mindesten einen Messbereich zu erstellen ist, und
• ein mit der mindestens einen Lichtquelle, der Dunkelfeld-Beleuchtungseinrichtung und der mindestens einen Kamera verbundenes Steuer- und Rechnersystem mit einer automatischen Steuer- / Auswertesoftware, von dem auf der Basis der Shape-from-Shading Methode für die vier unterschiedlichen realen Bildaufnahmen (Roh-Bilder) die x- bzw. y- Neigungsbild-Berechnungen der mindestens einen Funktionsoberfläche zu erstellen sowie aus den Neigungsbild-Berechnungen der Funktionsoberfläche und den Informationen der Dunkelfeld-Bildaufnahme die reale Defektfläche der mindestens einen Fehlstelle in dem mindestens einen Messbereich der Funktionsoberfläche des Bauteils oder Prüfkörpers zu bestimmen ist.

The object of the invention is also achieved by a system for determining the defect surface of at least one defect on at least one functional surface of a component or test specimen, comprising:

• a box-shaped housing, in which either a fixed bearing or a movable linear unit with a defined bearing and support surface is provided for receiving the component or specimen,
• at least one light source defined with regard to radiation intensity, direction and position, from which at least one measuring region of the functional surface of the component or test body positioned in the box-shaped housing on the support surface of the linear unit is to be illuminated sequentially from at least four different positions,
• at least one camera of known position, by means of which in each case a different real image acquisition (raw image) is to be created by the functional surface of the component or specimen in the measurement area,
• a dark field illumination device positioned in the box-shaped housing, of which dark field illumination is to be generated in the housing, wherein from the functional surface of the component or specimen during the dark field illumination by means of the at least one camera a dark field image acquisition of the at least one functional surface in the at least one measurement range to create, and
• a control and computer system connected to the at least one light source, the dark field illumination device and the at least one camera with automatic control / evaluation software, on the basis of the shape-from-shading method for the four different real image recordings (raw material). Images) to produce the x- or y-tilt image calculations of the at least one functional surface and from the tilt image calculations of the functional surface and the dark field image acquisition information the real defect area of the at least one defect in the at least one measurement area of the functional surface of the component or Specimen is to be determined.

The dark field illumination device is preferably constructed of LED strips. Preferably, four LED strips are provided, which are positioned in the box-shaped housing in each case parallel to one of the edges of the receiving surface of the fixed bearing or the linear unit on correctly aligned height of the functional surface to be tested and interconnected.

The box-shaped housing can, depending on the structure (fixed storage or displaceable linear unit), be installed in-line in a production or set up close to the production offline.

As light sources, preferably four individual reflectors connected in series are arranged in the box-shaped housing in its ceiling region, of which the functional surface of the component or test body positioned on the bearing (fixed support or linear unit) for at least one image acquisition series with four real image recordings (raw material). Pictures) one after the other with each defined exposure time to illuminate.

Preferably, at least one high-resolution CCD camera with at least 29 megapixels is provided for imaging in the box-shaped housing, from which e.g. three image acquisition series with four real images (raw images) are to be created with different exposure times.

Among other things, the method according to the invention makes it possible to use test systems for the optical control of functional surfaces of components of which not only defects as such can be detected, but also a realistic detection and specification of geometric properties such as length, width and area of the detected defect is guaranteed. In this way, a correct and reproducible transfer of the requirements of quality assurance in the test regulations in the series process is possible. In addition, an adaptation of the test specification is possible at any time, since it can be used directly with length specifications.

The present invention will now be explained with reference to the figures of the drawings.

• 1 eine Draufsicht auf eine schematisch dargestellte Anlage zum Ermitteln der Defektfläche mindestens einer Fehlstelle auf mindestens einer Funktionsoberfläche eines Bauteils oder Prüfkörpers.
• 2 eine schematische Ansicht des kastenförmigen Gehäuses der Anlage, das frontseitig geöffnet ist.
• 3 eine diagrammartige Darstellung einer vollständigen Bildaufnahme-Serie mit vier realen Bildaufnahmen (Roh-Bilder) bei jeweils unterschiedlicher Beleuchtungsrichtung für die Neigungs-Bildberechnung nach der Shape-from-Shading-Methode und einer fünften Bildaufnahme bei Dunkelfeldbeleuchtung.
• 4 eine diagrammartige Darstellung von drei vollständigen Bildaufnahme-Serien jeweils entsprechend 3, jedoch bei jeweils unterschiedlichen Belichtungszeiten.
• 5 ein blockdiagrammartiges Fließbild eines schematischen Ablaufs der Ermittlung der Defektfläche mindestens einer Fehlstelle einer Funktionsoberfläche eines Bauteils oder Prüfkörpers im Produktionsprozess.

In these are:

• 1 a plan view of a schematically illustrated system for determining the defect surface of at least one defect on at least one functional surface of a component or specimen.
• 2 a schematic view of the box-shaped housing of the system, which is open at the front.
• 3 a diagrammatic representation of a complete image acquisition series with four real images (raw images) in each case different illumination direction for tilt image calculation according to the shape-from-shading method and a fifth image recording in dark field illumination.
• 4 a diagrammatic representation of three complete image acquisition series respectively 3 , but with different exposure times.
• 5 a block diagram-like flow chart of a schematic sequence of the determination of the defect surface of at least one defect of a functional surface of a component or specimen in the production process.

Like from the 1 and 2 shows, the inventive system for determining the defect area DF at least one defect 7 at least one functional interface Z (x . y ) of a component or specimen 1 a box-shaped housing 8th on, which can be moved, like 2 shows.

In the lower part of the box-shaped housing 8th is either a fixed bearing or a movable linear unit 9 with defined storage and rubberized support surface 10 for receiving the component or specimen to be tested 1 intended. Furthermore, in the box-shaped housing 8th in the ceiling area 13 in four positions a . b . c . d as light sources four individual reflectors 3 _a . 3 _b . 3 _c . 3 _d provided which defines radiation intensity, direction and position, electrically connected in series and having a control and computer system 5 connected, which includes an automatic control / evaluation software.

The functional interface Z (x . y ) of the rubberized bearing surface 10 the solid storage or the linear unit 9 positioned component or specimen 1 is how the 1 and 2 make clear from the four reflectors in series 3 _a . 3 _b , 3 _c , 3 _d sequentially from the four different positions a . b . c . d each with a defined exposure time to illuminate.

As in 2 recognizable, is within the box-shaped housing 8th above the support surface 10 storage 9 at least one CCD camera 4 , preferably provided with 29 or more megapixels in a defined position and with the control and computer system 5 connected.

During the sequential lighting of the functional surface Z (x . y ) of the bearing surface 10 storage 9 positioned component or specimen 1 by means of the four reflectors 3 _a . 3 _b . 3 _c . 3 _d from the four different positions a . b . c . d in the ceiling area 13 the box-shaped housing 8th will be at least one of 3 apparent image acquisition - series S _l with four different real images (raw images) R _a, R _b . R _c . R _d of at least one measuring range 2 the functional interface Z (x . y ) created. By means of the automatic control / evaluation unit of the control and computer system 5 then takes place on the basis of the Shape-from-Shading (SfS) method for the four real images (raw images) R _a, R _b . R _c . R _d a tilt image calculation in the x or y direction N _x , N _{y of} the functional surface Z (x . y ).

In the box-shaped housing 8th is still one with the control and computer system 5 connected dark field illumination device 11 provided, as in 1 is apparent, preferably four LED strips 12 is formed, which are interconnected. According to 1 are the four LED strips 12 each parallel to an edge of the support surface 10 storage 9 arranged on which the component or the specimen 1 is positioned.

During a dark field illumination by means of the four interconnected LED strips 12 is from the at least one CCD camera 4 additionally a dark field image acquisition D the functional interface Z (x . y ) in the at least one measuring range 2 created, like out 3 evident.

Subsequently, by means of the automatic control / evaluation software unit of the control and computer system 5 from the tilt image calculations based on the Shape-from-Shading (SfS) method N _x and N _y the functional interface Z (x . y ) and the dark field image acquisition information D the functional interface Z (x . y ) the real defect area DF the at least one defect 7 in the at least one measuring range 2 the functional interface Z (x . y ) of the component or specimen 1 certainly.

Out 4 is a schematic diagrammatic representation of three complete image acquisition series S ₁ . S ₂ . S ₃ respectively according to the image-taking series S _l according to 3 but at different exposure times, as for measuring the defect area DF a defect 7 a functional interface Z (x . y ) with inhomogeneous ones Reflection properties is necessary. Here are
the first image acquisition series S _l with real first four images (raw images) R _a1 - R _d4 with an exposure time of 0.01 ms,
the second picture-taking series S ₂ with real four other images R _a6 - R _d9 with an exposure time of 0.025 ms,
the third picture-taking series S ₃ with a further four real image recordings R _a11- R _d14 with an exposure time of 0.05 ms, and
after the first, second and third image acquisition series S ₁ respectively. S ₂ respectively. S ₃ in each case a dark-field image recording D ₅ or D ₁₀ or D ₁₅ for the respective determination of the defect surface DF the defect 7 in the at least one measuring range 2 the functional interface Z (x . y ) of the component or specimen 1 created.

• A) Erkennen eines zu messenden Bauteils oder Prüfkörpers 1 durch einen z.B. induktiven Sensor.
• B) 1. Start eines neuen Prüfprozesses bestehend aus:
  1. a) Erstellen von drei vollständigen realen Bildaufnahme-Serien S₁ , S₂ , S₃ mit jeweils vier realen Bildaufnahmen (Roh-Bildern) R_a, R_b , R_c , R_d mit unterschiedlichen Belichtungsstufen als Basis für eine SfS-Berechnung für den Fall inhomogener Reflexionseigenschaften der Funktionsoberfläche Z (x, y) des Bauteils oder Prüfkörpers 1.
  2. b) Ansonsten Erstellen einer vollständigen Bildaufnahme-Serie S₁ mit jeweils vier realen Bildaufnahmen (Roh-Bildern) R_a, R_b , R_c , R_d mit definierter Belichtungszeit als Basis für die SfS- Berechnung.
• B) 2. Erfassen einer Bauteil- oder Prüfkörper-ID
• B) 3. SfS-Berechnung gemäß Punkt B) 1. a)
• B) 4. Neigungsbild-Berechnung N_x , N_y
• B) 5. Grauwertausgleich über Savitzky-Golay-Filter
• B) 6. Bildverarbeitung (Maskierung/Fehlererkennung mit realen geometrischen Informationen über eine Kombination von Neigungsbild-Berechnungen N_x , N_y und Informationen der Dunkelfeld-Bildaufnahmen D_5, D₁₀ , D₁₅ / Fehlerklassifizierung)
• B) 7. Fehleranzeige
• B) 8. Speichern der Ergebnisse

5 shows a block diagram-like flow chart of a schematic sequence of the determination of the defect area DF at least one defect 7 the functional interface Z (x . y ) of a component or specimen 1 in the production process with the following process steps:

• A) Detecting a component or test piece to be measured 1 by an eg inductive sensor.
• B) 1. Start of a new test process consisting of:
  1. a) Create three complete real image acquisition series S ₁ . S ₂ . S ₃ with four real images (raw images) R _a, R _b . R _c . R _d with different exposure levels as the basis for an SfS calculation in the case of inhomogeneous reflection properties of the functional surface Z (x . y ) of the component or specimen 1 ,
  2. b) Otherwise, create a complete image capture series S ₁ with four real images (raw images) R _a, R _b . R _c . R _d with defined exposure time as the basis for the SfS calculation.
• B) 2. Detecting a part or specimen ID
• B) 3. SfS calculation according to point B) 1. a)
• B) 4. Slope image calculation N _x . N _y
• B) 5. Gray balance over Savitzky-Golay filter
• B) 6. Image processing (masking / error detection with real geometric information about a combination of tilt image calculations N _x . N _y and information of the dark field image recordings D _5, D ₁₀ . D ₁₅ / Error classification)
• B) 7. Error display
• B) 8. Save the results

It should be understood that the embodiments of the present invention are not limited to the particular structures, process steps, or materials disclosed herein, but their equivalents may be extended to those of ordinary skill in the relevant arts. It should be understood that the terminology used herein is used merely to describe particular embodiments and is not to be construed as limiting. The described features, structures or properties may be combined in any suitable manner in one or more embodiments.

LIST OF REFERENCE NUMBERS

1

Component, test specimen

2

measuring range

3

light source

3 _a , 3 _b , 3 _c , 3 _d

collectors

4

Camera, CCD camera with 29 or more megapixels

5

Control and computer system

6

Plant for determining the defect area of a defect

7

void

8th

box-shaped housing

9

fixed storage or movable linear unit

10

rubberized support surface

11

Dark field illumination means

12

LED bars

13

Ceiling area of the box-shaped housing

Z (x, y)

Working surface

a, b, c, d

Positions of the light source or collectors

R _a, R _b , R _c , R _d

real pictures (raw pictures)

R _a1 , R _b2 , R _c3 , R _d4

Real image images (raw images) of the first image acquisition series S ₁

R _a6 , R _b7 , K _c8 , R _d9

real images (raw images) of the second image acquisition series S ₂

R _a11 , R _b12 R _c13 , R _d14

real images (raw images) of the third image acquisition series S ₃

N _x , N _y

Slope image calculations in x- or y-derivative

S ₁ S ₂ , S ₃

Image capture series

SfS method

Shape-from-shading method

D

Dark field imaging

D _5, D ₁₀ , D ₁₅

Darkfield image acquisition after the first, second or third image acquisition series

H

Bright field imaging

DF

Defect area of a defect

F

Arrow symbolizing the direction of the manufacturing process

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102004038761 B4 [0006]
• DE 20317095 U1 [0007]

Claims (21)

Method for determining the defect area (DF) of at least one defect (7) on at least one functional surface Z (x, y) of a component or test body (1), comprising the method steps: - the component or the test body (1) is in at least one measuring range (2) the functional surface Z (x, y) is illuminated sequentially from at least four different positions (a, b, c, d) with at least one light source (3) defined with respect to radiation intensity, direction and position; x, y) of the component or specimen (1) is at least one with a control and computer system (5) connected camera (4) known position respectively a different real image acquisition (raw image) (R _a, R _b , R _c , R _d ) of the functional surface Z (x, y) in the at least one measuring range (2) created by the control and computer system (5) are based on the Shape-from-Shading (SfS) method for the four real images (Ro h-images) (R _a, R _b , R _c , R _d ) calculates the x or y inclination images (N _x , N _y ) of the functional surface Z (x, y), from the functional surface Z (x, y) of the component or specimen (1) a dark field image acquisition (D) of the functional surface Z (x, y) in the measurement region (2) is created during dark field illumination, and - from the x or y inclination image computations ( N _x , N _y ) of the functional surface Z (x, y) and the information of the dark-field image acquisition (D), the real defect surface (DF) of the at least one defect (7) in the at least one measuring region (2) of the functional surface Z (FIG. x, y) of the component or test specimen (1) by means of the control and computer system (5). Method according to Claim 1 , characterized in that the calculated x- or y-slope images (N _x , N _y ) are binarized and the at least one defect (7) on the at least one functional surface Z (x, y) based on the calculated and binarized slope images (N _x , N _y ) is detected, with a size of the defect (7) in the range of tenths of millimeters whose geometry in the binarized inclination images (N _x , N _y ) approximately coincides with the real defect surface (DF) of the defect (7) the case, however, that the size of the defect (7) lies in a larger area, its partial areas, which are reproduced only incompletely in the inclination images (N _x , N _y ), on the dark field image acquisition (D) of the functional surface Z (x , y) are binarized and combined with the binarization of the inclination images (N _x , N _y ) and from the combined binarization of the inclination images (N _x , N _y ) and the dark field image acquisition (D) the real geometry of the defect surface Area (DF) of the defect (7) of the functional surface Z (x, y) is approximately determined. Method according to one of the preceding claims, characterized in that four individual, in series reflectors (3 _a , 3 _b , 3 _c , 3 _d ) are used as light sources, of which the functional surface Z (x, y) successively with each defined Exposure time is illuminated. Method according to one of the preceding claims, characterized in that as dark field illumination LED lights (12), preferably four LED strips 12 are used, which are connected together. Method according to one of the preceding claims, characterized in that from the functional surface Z (x, y) by means of a CCD camera (4) three image recording series (S ₁ , S ₂ , S ₃ ) each having four real image recordings (raw material). Images) (R _a, R _b , R _c , R _d ), wherein the functional surfaces Z (x, y) in each of the three image acquisition series (S ₁ , S ₂ , S ₃ ) of the series-connected reflectors (3a, 3 _b , 3 _c , 3 _d ) is illuminated successively with a defined exposure time. Method according to one of the preceding claims, characterized in that in the case of inhomogeneous reflection properties of the functional surface Z (x, y) of the component or specimen (1), the three image acquisition series (S ₁ , S ₂ , S ₃ ) are each provided with four real image recordings ( Raw images) (R _a, R _b , R _c , R _d ) with different exposure times of the four reflectors (3 _a , 3 _b , 3 _c , 3 _d ) are created by the component or the test specimen (1) in at least a measuring range (2) of the functional surface Z (x, y) with the reflectors (3 _a , 3 _b , 3 _c , 3 _d ) defined in terms of radiation intensity, direction and position sequentially from the four different positions (a, b, c , d) is illuminated. Method according to one of the preceding claims, characterized in that the real image recordings (raw images) (R _a, R _b , R _c , R _d ) of each of the three image acquisition series (S ₁ , S ₂ , S ₃ ) with different Exposure times are created. Method according to Claim 7 , characterized in that the first image recording series (S ₁ ) with real image recordings (R _a1 -R _d4 ) with an exposure time of 0.01 ms, the second image recording series (S ₂ ) with real image recordings (R _a6 -R _d9 ) with an exposure time of 0.025 ms, the third image acquisition series (S ₃ ) with real image recordings (R _a11 -R _d14 ) with an exposure time of 0.05 ms, and after the first, the second and the third image acquisition series (S ₁ or S ₂ or S ₃ ) in each case a dark field image acquisition (D ₅ or D ₁₀ or D ₁₅ ) for the respective determination of the defect surface (DF) of the at least one defect (7) is created. Method according to one of the preceding claims, characterized in that a bright field illumination is generated instead of the dark field illumination, during which a bright field image acquisition (H) of the at least one functional surface Z (x, y) in the at least one measurement range (2) is created, and from the inclination image calculations (N _x , N _y ) of the functional surface Z (x, y) and the information of the bright field image acquisition (H), the real defect surface (DF) of the at least one defect (7) in the at least one measurement region (2 ) of the functional surface Z (x, y) of the component or specimen (1) by means of the control and computer system (5) is determined. Method according to one of the preceding claims, characterized in that in the creation of the respective different inclination images (N _x , N _y ) of the at least one functional surface Z (x, y) in the at least one measuring range (2) a grayscale compensation over Savitzky-Golay- Filter takes place. Method according to any one of the preceding claims, characterized in that the at least one measuring range (2) of the functional surface Z (x, y) is subdivided into sectors, and a separate parameterization per inclination image recording (N _x , N _y ) and per dark field Image acquisition (D) or per bright field image acquisition (H) takes place. Method according to any one of the preceding claims, characterized in that in the processing of the inclination image recordings (N _x , N _y ) according to the shape-from-shading (SfS) method and the dark-field image recordings (D) or the bright field Masking takes place (H) a masking, wherein an error detection with realistic geometric information on the combination of information of the respective inclination image recordings (N _x , N _y ) and dark field image recordings (D) and the bright field image recordings (H) performed and the detected errors are classified, displayed and / or stored. Method according to one of the preceding claims, characterized in that at least one high-resolution CCD camera (4) with at least 29 megapixels is used for image recording. Installation for determining the defect area (DF) of at least one defect (7) in at least one measuring area (2) of at least one functional surface Z (x, y) of a component or test body (1), comprising: a box-shaped housing (8) in which a fixed bearing or a movable linear unit (9) with a defined bearing and support surface (10) for receiving the component or specimen (1) is provided, at least one with respect to radiation intensity, direction and position defined light source (3) from which the measuring range ( 2) of the functional surface Z (x, y) of the component or test body (1) positioned in the box-shaped housing (8) on the bearing surface (10) of the bearing (9), sequentially from at least four different positions (a, b, c, d) ), at least one camera (4) known position, by means of the function of the surface Z (x, y) of the component or specimen (1) each have a different real image acquisition (raw Bi (R _a, R _b , R _c , R _d ) in the at least one measuring range (2) is to be created, a dark field illumination device (11) positioned in the box-shaped housing (8), in which the housing (8 a dark-field illumination is to be generated, wherein from the at least one functional surface Z (x, y) of the component or specimen (1) during the dark-field illumination by means of the at least one camera (4) a dark-field image acquisition (D) of the functional surface Z (x, y) in which at least one measuring range (2) is to be created, and a control and computer system (5) connected to the at least one light source (3), the dark field illumination device (11) and the at least one camera (4) automatic control / evaluation software, on the basis of the Shape-from-Shading (SfS) method for each real image images (raw images) (R _a, R _b , R _c , R _d ) an x or y Slope calculation (N _x , N _y ) of the functional surface Z (x, y) as well as from the various inclination-image calculations (N _x , N _y ) of the functional surface Z (x, y) and the information of the dark-field image acquisition (D), the real defect surface (DF) of the at least one defect (7) in the at least one measurement region (2) the functional surface Z (x, y) of the component or specimen (1) is to be determined. Plant after Claim 14 , characterized in that the dark field illumination device (11) of LED strips (12), preferably formed of four parallel LED strips (12). Plant after Claim 14 and 15 , characterized in that each one of the four parallel LED strips (12) in the box-shaped housing (8) in each case parallel to one of the edges of the component or the test specimen (1) receiving defined bearing surface (10) is arranged. Plant according to one of the Claims 14 to 16 , characterized in that in the box-shaped housing (8) in its ceiling region (13) at four positions (a, b, c, d) as light sources four individual, in series reflectors (3 _a , 3 _b , 3 _c , 3 _d ), of which the functional surface Z (x, y) of the component or specimen (1) positioned on the support surface (10) of the bearing (9) for at least one image acquisition series (S ₁ ) with four real image recordings (R _a, R _b , R _c , R _d ) is to be illuminated in succession, each time with a defined exposure time. Plant according to one of the Claims 14 to 17 , characterized in that at least one high-resolution CCD camera (4) is provided with at least 29 megapixels, of the three image acquisition series (S ₁ , S ₂ , S ₃ ), each with four real image recordings (raw images) (R _{a ,} R _b, R _c, R _d) are to be compiled. Plant according to one of the Claims 14 to 18 , characterized in that the real image recordings (raw images) (R _a, R _b , R _c , R _d ) of each of the three image acquisition series (S ₁ , S ₂ , S ₃ ) are to be created with different exposure times. Plant after Claim 19 , characterized in that the first image recording series (S ₁ ) with image recordings (raw images) (R _a1- R _d4 ) with an exposure time of 0.01 ms, the second image recording series (S ₂ ) with image recordings (raw material) Images) (R _a6- R _d9 ) with an exposure time of 0.025 ms, the third image acquisition series (S ₃ ) with image recordings (raw images) (R _a11- R _d14 ) with an exposure time of 0.05 ms and after the first, the second and the third image recording series (S ₁ or S ₂ or S ₃ ) each have a dark field image recording (D ₅ or D ₁₀ or D ₁₅ ) for the respective determination of the real defect surface (DF) of the at least one Defect (7) is to create. Plant according to one of the Claims 14 to 20 , characterized in that instead of the dark field illumination device (11) a bright field illumination device is provided, wherein a bright field image capture (H) of the at least one functional surface Z (x, y) in the at least one measuring range (2) to create and off the tilt image calculations (N _x , N _y ) of the functional surface Z (x, y) and the information of the bright field image acquisition (H) the real defect surface (DF) of the at least one defect (7) in the at least one measurement region (2) the functional surface Z (x, y) of the component or specimen (1) is to be determined by means of the control and computer system (5).

Images