Data processing for structural scanners

Abstract

A structural scanning and analysis apparatus 1, which may be used in the packaging or battery production industries for example, comprises a solid state light-based scanner 100, a processing module 25

Claims

1. A structural scanning and analysis apparatus, comprising:a solid state light-based scanner arranged to capture image data associated with at least one image of a target surface which is substantially flat;a processing module arranged to receive image data from the scanner, wherein the processing module comprises an illumination correction module arranged to normalise the intensity of the image data to correct for non-uniform illumination and / or non-uniform sensor response across the at least one image; anda depth-map generation module arranged to generate a depth map of the target surface based on image data generated by the image processing module.

2. The structural scanning and analysis apparatus of claim 1, wherein:the illumination correction module is arranged to perform an illumination correction process which comprises multiplying the intensity of each pixel of the image data being processed by the processing module by a correction factor to normalise the intensity of the image data;the correction factor is determined by dividing the mean intensity of a respective calibration image by the intensity of each pixel in the calibration image; andthe respective calibration image is an image of a flat, uniformly reflective surface under the same illumination conditions as those in which the image data are captured.

3. The structural scanning and analysis apparatus of claim 2, wherein:the light-based scanner comprises one or more cameras and a plurality of lights;capturing image data associated with at least one image of the target surface comprising obtaining image data using the one or more cameras and each one of the lights; andperforming the illumination correction process comprises multiplying the image data captured using each light by an associated correction factor, the associated correction factor being determined using a calibration image obtained using the same light and camera combination.

4. The structural scanning and analysis apparatus of any preceding claim, wherein the image data associated with at least one image of the target surface includes image data obtained using at least three linearly independent lighting directions.

5. The structural scanning and analysis apparatus of any preceding claim, wherein:the processing module further comprises an image stacking module arranged to combine pixel data from an image stack, the image stack comprising a plurality of images captured by the scanner using the same light.

6. The structural scanning and analysis apparatus of any preceding claim, wherein the processing module further comprises an image distortion correction module arranged to adjust the image data for alignment with a known physical coordinate space.

7. The structural scanning and analysis apparatus of claim 6, wherein the image distortion correction module comprises a grid warping module arranged to adjust image data being processed by the processing module.

8. The structural scanning and analysis apparatus of claim 7. wherein the image data is adjusted using a thin-plate-spline algorithm.

9. The structural scanning and analysis apparatus of any of claims 6 to 8, wherein the image distortion correction module comprises a lens intrinsics module arranged to apply an intrinsics model of the optics used in the scanner to adjust the image data being processed by the processing module.

10. The structural scanning and analysis apparatus of any of claims 6 to 9, wherein the image distortion correction module comprises a homography module arranged to apply a homography matrix to adjust the image data being processed by the processing module.

11. The structural scanning and analysis apparatus of any preceding claim, wherein the scanner comprises a plurality of lights and / or a plurality of cameras arranged to illuminate the target surface and is arranged to capture image data associated with the target surface using different combinations of the lights and / or cameras, the depth-map generation module being arranged to generate the depth map by:computing a normal map using image data associated with different light and / or camera combinations to generate a plurality of normal maps;generating a refined normal map by determining an optimum pixel value from amongst a set of pixels, each member of the set of pixels corresponding to the same pixel position in the plurality of normal maps, wherein the optimum pixel value is determined from the pixel or pixels which are least affected by imaging artifacts caused by interactions between the lights and target surface.

12. The structural scanning and analysis apparatus of claim 11, wherein the imaging artifacts are caused by any one or more of specular reflection, reflective spots, and shadows.

13. The structural scanning and analysis apparatus of any preceding claim, further comprising an analysis module configured to identify one or more 3D topological features ofthe target surface based on the depth map, wherein the one or more 3D topological features comprise:i) braille provided on the target surface; and / orii) a crease on the target surface.

14. A method of generating a depth map using a structural scanning and analysis apparatus, the method comprising:capturing image data associated with at least one image of a target surface which is substantially flat using a solid state light-based scanner;processing the image data by normalising the intensity of the image data to correct for non-uniform illumination and / or sensor response across the at least one image; andgenerating a depth map of the target surface based on the processed image data.

15. The method of claim 14, wherein:normalising the image intensity comprises performing an illumination correction process which comprises multiplying the intensity of each pixel of the image data being processed by a correction factor to normalise the intensity of the image data;the correction factor is determined by dividing the mean intensity of a respective calibration image by the intensity of each pixel in the calibration image; andthe calibration image is an image of a flat, uniformly reflective surface under the same illumination conditions as those in which the image data are captured.

16. The method of claim 15, wherein:the light-based scanner comprises one or more cameras and a plurality of lights;capturing image data associated with at least one image of the target surface comprising obtaining image data using the one or more cameras and each one of the lights; andperforming the illumination correction process comprises multiplying the image data captured using each light by an associated correction factor, the associated correction factor being determined using a calibration image obtained using the same combination of light and camera.

17. The method of any of claims 14 to 16, wherein the image data associated with the at least one image of the target surface includes image data obtained using at least three linearly independent lighting directions.

18. The method of any of claims 14 to 17, wherein processing the image data further comprises combining pixel data from an image stack, the image stack comprising a plurality of images captured by the scanner using the same light.

19. The method of any of claims 14 to 18, wherein processing the image data further comprises adjusting the image data for alignment with a known physical coordinate space.

20. The method of claim 19, wherein adjusting the image data comprises applying a grid warping algorithm, preferably a thin-plate-spline algorithm.

21. The method of claim 19 or claim 20, wherein adjusting the image data comprises applying an intrinsics model of the optics used in the scanner to adjust the image data.

22. The method of claim 19 or claim 20 or claim 21, wherein adjusting the image data comprises applying a homography matrix to adjust the image data.

23. The method of any of claims 14 to 22, wherein the scanner comprises a plurality of lights and / or a plurality of cameras arranged to illuminate the target surface and is arranged to capture image data associated with the target surface using different combinations of the lights and / or cameras, the method further comprising:computing a normal map using image data associated with different light and / or camera combinations to generate a plurality of normal maps;generating a refined normal map by determining an optimum pixel value from amongst a set of pixels, each member of the set of pixels corresponding to the same pixel position in the plurality of normal maps, wherein the optimum pixel value is determined from the pixel or pixels which are least affected by imaging artifacts caused by interactions between the lights and target surface.

24. The method of claim 23, wherein the imaging artifacts are caused by any one or more of specular reflection, reflective spots, and shadows.

25. The method of any of claims 14 to 24, further comprising identifying one or more 3D topological features of the target surface based on the depth map, wherein the one or more 3D topological features comprise:i) braille provided on the target surface; and / orii) a crease on the target surface.

Images