Augmented image correlation
View Patent ↗Systems and methods search pixels of source images and compare them to pixels of templates. Best matches correlate to objects in the image. That environmental conditions impact the appearance of objects, best matching scores suffer under poor lighting and other adverse conditions. Improving scores includes augmenting traditional correlation techniques with object features extracted from the source image and matching them to templates corresponding to the features. Certain embodiments contemplate corrupting pixels of image templates corresponding to objects with pixels extracted from the source image corresponding to features. Representative features include corners, edges, ridges, points/regions of interest, etc. Other embodiments note augmented correlation as a computing application and computing devices therefore, including cameras for capturing images and displaying results to users.
1. In a computing system environment, a method for recognizing objects in images defined by pixels, comprising:
searching pixels of a source image;
correlating the searched pixels to a plurality of predefined image templates of pixels and selecting a best matching image template therefore;
extracting certain pixels from the source image that delimit features of the objects;
correlating the extracted certain pixels to a plurality of predefined feature templates of pixels and selecting a best matching feature template therefore; and
combining together results of said correlating; and display to a user.
2. The method of claim 1 , further including digitally capturing the source image with a camera.
3. The method of claim 1 , wherein the searching pixels of the source image further includes advancing a window through the pixels of the source image, the window having an m×n array of pixels corresponding to an m×n array of the pixels in the plurality of predefined image templates.
4. The method of claim 1 , where the extracting certain pixels from the source image includes extracting feature corners of the objects in the source image.
5. The method of claim 1 , wherein the plurality of predefined image templates is the same as the predefined features templates.
6. The method of claim 1 , wherein the plurality of image templates include pixel values obtained from the certain pixels said extracted from the source image.
7. The method of claim 1 , wherein the plurality of image templates are obtained by corrupting original pixel values of a plurality of source templates with the certain pixels extracted from the source image and changing one or more pixel values therefore.
8. The method of claim 1 , wherein the correlating the extracted certain pixels to the plurality of predefined feature templates of pixels further includes determining numbers of simultaneously on pixels there between.
9. The method of claim 1 , wherein either the predefined image templates or the predefined features templates are inverted versions of templates defining the objects.
10. In a computing system environment, a method for recognizing objects in images defined by pixels, comprising:
searching pixels of a source image;
extracting certain pixels from the source image that delimit features of the objects;
correlating the extracted certain pixels to a plurality of predefined feature templates of pixels;
modifying one or more of a plurality of image templates of pixels with the certain pixels said extracted from the source image to obtain a plurality of modified templates, including corrupting the pixels of the one or more plurality of image templates by changing one or more pixel values therefore;
correlating the searched pixels of the source image to the plurality of modified templates of pixels; and
combining together results of said correlating; and display to a user.
11. The method of claim 10 , further including digitally capturing the source image with a camera.
12. The method of claim 10 , wherein the searching pixels of the source image further includes advancing a window through the pixels of the source image, the window having an m×n array of pixels corresponding to an m×n array of the pixels in the plurality of modified templates.
13. The method of claim 10 , wherein the extracting certain pixels from the source image includes extracting feature corners of the objects in the source image.
14. A computing device that recognizes objects in images defined by pixels, comprising:
a housing;
a camera activated by a user manipulating the housing; and
a controller in the housing hosting a computing application with executable instructions operative for
receiving an image captured by the camera;
searching pixels of the image;
correlating the searched pixels to a plurality of image templates stored in memory for access by the controller;
extracting certain pixels from the image that delimit features of the objects; and
correlating the extracted certain pixels to a plurality of feature templates stored in memory for access by the controller; and
selecting a best object from the correlated searched pixels and the correlated extracted certain pixels and combine together results of the correlating;
and a display for showing the user the best object.
15. The computing device of claim 14 , wherein the housing defines a computing tablet, a smart phone, or an imaging device.
16. The computing device of claim 14 , wherein the objects include numerals on a user banking card and the display shows the both the banking card as captured by the camera and the objects after execution of the computing application hosted on the controller.
17. An augmented correlation application available on a computer readable medium or hosted on a computing device having memory for download onto a controller of an imaging device or a mobile computing device comprising executable code operative for extracting objects and delimiting features thereof from a source image for comparison to pluralities of templates of same and selecting best matches, and combining together results thereof.
18. The augmented correlation application of claim 17 , further operative for searching pixels of the source image and correlating the searched pixels to pixels of the pluralities of templates.
19. The augmented correlation application of claim 18 , further operative for modifying one or more of the pluralities of templates with the extracted delimiting features, including corrupting pixels of the one or more pluralities of templates by changing one or more pixel values therefore.
20. The augmented correlation application of claim 17 , further operative for causing display to a user of the imaging device of mobile computing device the extracted objects upon execution of the application by the controller.