Method and device for inspection of a geometry, the device comprising image capturing and shape scanning means
Method for inspecting a geometry ( 10 ) having a surface ( 11 ), which method uses an inspection device ( 100 ) having a digital image capturing means ( 120 ) and a shape scanning means ( 130 ), comprising the steps of a) orienting the inspection device ( 100 ) in a first orientation; b) depicting the surface ( 11 ) to produce a first image; c) measuring a shape of a first geometry part to produce a first shape; d) moving the inspection device ( 100 ) to a second orientation; e) depicting the surface ( 11 ) to produce a second image; f) measuring a second part of said geometry ( 10 ) to produce a second shape; g) using digital image processing based on said first and second images, determining a geometric orientation difference between said first and second orientations; h) determining a geometric relation between the first and second shapes; and i) producing a data representation of said geometry based on said first shape, said second shape and said geometric relation. The invention also relates to a device.
1. A method for inspecting a geometry ( 10 ), the geometry ( 10 ) being a weld joint having a surface ( 11 ), by using a handheld, mobile inspection device ( 100 ) having a digital image capturing means ( 120 ) and a shape scanning means ( 130 ), wherein the method comprises the steps of:
a) freely orienting the inspection device ( 100 ) in a first orientation in relation to the surface ( 11 ) so that the inspection device ( 100 ) can perform an image capturing and a shape measuring of a certain part of the geometry ( 10 );
b) imaging the surface ( 11 ) using the digital image capturing means ( 120 ) to produce a first image;
c) using the shape scanning means ( 130 ) to perform a laser-based distance measurement from the shape scanning means ( 130 ) to the surface ( 11 ) to measure a shape of a first part of the geometry ( 10 ) to produce a first shape, the shape scanning means ( 130 ) being separate from the digital image capturing means ( 120 );
d) freely moving the inspection device ( 100 ) to a second orientation in relation to the surface ( 11 ), wherein the second orientation is not the same as the first orientation, wherein the moving comprises a translation movement of the inspection device ( 100 ) along the surface ( 11 ) in a longitudinal direction (L) along a length of the geometry ( 10 );
e) imaging the surface ( 11 ) using the digital image capturing means ( 120 ) to produce a second image;
f) using the shape scanning means ( 130 ) to perform a laser-based distance measurement from the shape scanning means ( 130 ) to the surface ( 11 ) to measure a second part of the geometry ( 10 ) to produce a second shape;
g) determining, by using digital image processing based on the first and second images, a geometric orientation difference between the first and second orientations, wherein the determining comprises calculating a maximum value for a correlation between the first and second images when varying a translational relative position of the first and second images;
h) based on the geometric orientation difference, determining a geometric relation between the first and second shapes; and
i) producing a data representation of the geometry ( 10 ) based on the first shape, the second shape and the geometric relation,
wherein the first shape is measured at a time between capturing the first and second images, and wherein a location of the first shape is determined by interpolating between the first and second images.
2. The method according to claim 1 , wherein the method furthermore comprises geometrically relating the first image and the second image to the data representation and presenting to a user an interactive user interface which a user can use to select a particular part of the geometry ( 10 ) and as a result thereof be presented with both a representation of the particular part and also with image material visually depicting the particular part.
3. The method according to claim 1 , wherein the correlation is performed in the frequency domain using a respective Fourier transformation of each of the first and second images.
4. The method according to claim 1 , wherein the correlation is performed at least partly in a log-polar coordinate system.
5. The method according to claim 1 , wherein the method is performed iteratively whereby steps d) to h) are iterated whereby third and subsequent images and shapes are produced.
6. The method according to claim 5 , wherein the determining in step g) comprises applying a Kalman filter designed to assume a predetermined translational movement direction of the inspection device across the surface ( 11 ).
7. The method according to claim 6 , wherein step g) comprises providing to the Kalman filter an estimated measurement quality of a determined geometric orientation difference for a particular produced image, and wherein the determination in step g) further takes into consideration different produced images to different degrees, in turn, based on a determined estimated measurement uncertainty for the particular produced image.
8. The method according to claim 5 , wherein the determination in step g) is based on at least three different produced images.
9. The method according to claim 1 , wherein the geometry ( 10 ) has a substantially constant cross-section across the longitudinal direction (L).
10. The method according to claim 1 , wherein the inspection device ( 100 ) comprises a light source ( 140 ) arranged to project a guide light ( 141 ) of predetermined wavelengths useful for guiding a user of the inspection device ( 100 ) when moving the inspection device ( 100 ) across the surface ( 11 ), wherein at least one of the first and second images cover an area comprising at least part of the guide light ( 141 ) as reflected by the geometry ( 10 ), and wherein the imaging in steps b) and e) comprises a filtering sub step, wherein the filtering sub step is designed to dampen the predetermined wavelengths.
11. The method according to claim 10 , wherein the filtering sub step is implemented using a physical light filter belonging to the digital image capturing means.
12. The method according to claim 10 , wherein the filtering sub step is implemented as a part of the digital image processing.
13. A handheld, mobile inspection device ( 100 ) for inspecting a geometry ( 10 ), the geometry ( 10 ) being a weld joint having a surface ( 11 ), wherein the inspection device ( 100 ) comprises a computer device, a digital image capturing means ( 120 ) and a shape scanning means ( 130 ), the shape scanning means ( 130 ) being separate from the digital image capturing means ( 120 ), wherein the inspection device ( 100 ) is arranged to, when freely oriented in a first orientation in relation to the surface ( 11 ) so that the inspection device ( 100 ) can perform an image capturing and a shape measuring of a certain part of the geometry ( 10 ):
image the surface ( 11 ) using the digital image capturing means ( 120 ) to produce a first image;
use the shape scanning means ( 130 ) to perform a laser-based distance measurement from the shape scanning means ( 130 ) to the surface ( 11 ) to measure a shape of a first part of the geometry ( 10 ) to produce a first shape;
freely move the inspection device ( 100 ) to a second orientation in relation to the surface ( 11 ), wherein the second orientation is not the same as the first orientation, wherein the moving comprises a translation movement of the inspection device ( 100 ) along the surface ( 11 ) in a longitudinal direction (L) along a length of the geometry ( 10 );
image the surface ( 11 ) using the digital image capturing means ( 120 ) to produce a second image;
use the shape scanning means ( 130 ) to perform a laser-based distance measurement from the shape scanning means ( 130 ) to the surface ( 11 ) to measure a second part of the geometry ( 10 ) to produce a second shape;
determine, by using digital image processing based on the first and second images, a geometric orientation difference between the first and second orientations, wherein the determination comprises calculating a maximum value for a correlation between the first and second images when varying a translational relative position of the first and second images;
determine a geometric relation between the first and second shapes based on the geometric orientation difference; and
produce a data representation of the geometry ( 10 ) based on the first shape, the second shape and the geometric relation,
wherein the first shape is measured at a time between capturing the first and second images, and wherein a location of the first shape is determined by interpolating between the first and second images.