Reconstruction of virtual raster
View Patent ↗A method aims at identifying, starting from objects in a digital image, a virtual raster included in a coding pattern. The objects in the image at least partially reproduce marks on a base, each mark being associated with a respective point of intersection of raster lines belonging to the virtual raster. The method comprises the steps of matching sets of objects against a cell unit, which corresponds to a recurring known basic element of said raster; when a set corresponds with the cell unit, identifying the objects in the set as approved; and reconstructing the virtual raster on the basis of the mutual arrangement of the approved objects. A computer program, a storage medium and a device for position determination are also described.
1. A method of correcting for geometrical perspective distortion in a sequence of digital representations of a coding pattern, said method comprising:
deriving current correction data for a digital representation in said sequence;
using the current correction data to identify a correspondence between objects in the digital representation and elements of the coding pattern;
generating, from the digital representation and based upon said correspondence, a distortion-corrected digital representation; and
deriving subsequent correction data for a subsequent digital representation in said sequence based upon the current correction data.
2. The method of claim 1 , wherein said generating comprises: computing homogeneous transformation data based upon said correspondence.
3. The method of claim 2 , wherein said generating further comprises: operating the homogeneous transformation data on the digital representation to generate the distortion-corrected digital representation.
4. The method of claim 2 , wherein the subsequent correction data is derived from the homogeneous transformation data.
5. The method of claim 4 , wherein the subsequent correction data comprises linear transformation data extracted from the homogeneous transformation data.
6. The method of claim 1 , wherein the current correction data comprises linear transformation data, and wherein said using comprises: operating the linear transformation data on said objects.
7. The method of claim 1 , wherein said elements comprise at least one of: code marks of the coding pattern and points of intersection of raster lines included in the coding pattern.
8. The method of claim 1 , wherein said correspondence relates a number of object locations in the digital representation to a corresponding number of intersections of raster lines included in the coding pattern.
9. The method of claim 8 , wherein said number is at least four.
10. The method of claim 2 , wherein said computing comprises computing eight homogeneous transformation parameters.
11. The method of claim 1 , wherein the current correction data comprises linear transformation data.
12. The method of claim 11 , wherein said deriving current correction data comprises:
identifying at least two main vectors of the current digital representation; and
deriving the linear transformation data that transforms the main vectors to essentially coincide with corresponding reference vectors of the coding pattern.
13. The method of claim 12 , wherein the linear transformation data is derived to adapt the magnitude of the main vectors to the magnitude of the corresponding reference vectors.
14. The method of claim 12 , wherein the reference vectors are mutually orthogonal.
15. The method of claim 1 , wherein each digital representation corresponds to an image in a sequence of two-dimensional digital images.
16. The method of claim 15 , wherein each digital representation is binarized version of said image.
17. The method of claim 15 , wherein each digital representation comprises a data structure containing location indications for objects in said image.
18. The method of claim 1 , wherein the coding pattern comprises a raster formation comprising a first plurality of raster line intersections, and a second plurality of code marks, each raster line intersection being assigned at least one code mark.
19. The method of claim 18 , wherein each raster line intersection represents a value by way of a relative location of said at least one code mark.
20. The method of claim 19 , wherein said value is given by a displacement of the center of gravity of the code mark relative to the raster line intersection.
21. A computer-readable medium comprising program code which, when executed in a computer, causes the computer to implement a method of correcting for geometrical perspective distortion in a sequence of digital representations of a coding pattern, said method comprising:
deriving current correction data for a digital representation in said sequence;
using the current correction data to identify a correspondence between objects in the digital representation and elements of the coding pattern;
generating, from the digital representation and based upon the current correction datasaid correspondence, a distortion-corrected digital representation; and
deriving subsequent correction data for a subsequent digital representation in said sequence based upon the current correction data.
22. A device for correcting for geometrical perspective distortion in a sequence of digital representations of a coding pattern, said device comprising:
a first calculator which derives current correction data for a digital representation in said sequence;
a corrector which uses the current correction data to identify a correspondence between objects in the digital representation and elements of the coding pattern and which generates, from the digital representation and based upon said correspondence, a distortion-corrected digital representation; and
a second calculator which derives subsequent correction data for a subsequent digital representation in said sequence based upon the current correction data.
23. A handheld apparatus, comprising:
a reader for generating a sequence of digital representations of a position-coding pattern,
a device for generating a sequence of distortion-corrected digital representations, said device comprising a first calculator which derives current correction data for a digital representation in said sequence of digital representations, a corrector which generates, from the digital representation and based upon the current correction data, a distortion-corrected digital representation, and a second calculator which derives subsequent correction data for a subsequent digital representation in said sequence based upon the current correction data; and
a position decoder for generating position data based upon said sequence of distortion-corrected digital representations.