Method of graphical objects recognition using the integrity principle
View Patent ↗The present invention discloses a method to increase reliability, correctness of objects recognition processing by performing a recognized object description as a set of special standard elements along with the spatial and parametrical correlation thereof. The said standard elements are preliminarily assigned graphic structures of elementary form and of easy identification and recognition. They may be provided with spatial and/or parametric details and thus may describe any object on the image including characters of text.
1. A method of object recognition using a classifier on a bit-mapped image, comprising:
identifying objects to be recognized on the bit-mapped image;
preliminarily assigning at least one graphical structure comprising more than one primary graphical unit to be used as a standard element constituent of each object to be recognized;
preliminarily describing each object to be recognized as a set of said standard elements of at least one type along with spatially parametrical correlations thereof;
performing the following steps:
search and identification of at least one standard element on the said-bit-mapped image;
selection of at least one standard element image for testing on as belonging to the object to be recognized; and
setting up and testing a hypothesis about the object to be recognized on the basis of an image formed by aggregating all each selected standard element image taking into account spatially parametrical correlations thereof.
2. The method as recited in claim 1 , wherein at least one standard element composing the recognized object is specified with alternative variants.
3. The method as recited in claim 1 , wherein the set of standard elements composing the recognized object is specified with alternative variants.
4. The method as recited in claim 1 , wherein at least one standard element composing the recognized object comprises an interval for at least one spatially parametrical correlation value.
5. The method as recited in claim 1 , wherein the image at least partly contains standard elements connected by relations of mathematical logic.
6. The method as recited in claim 1 , wherein aggregating each selected standard element image additionally comprises analysis of standard elements connected by a relation of “AND” type, analysis of standard elements connected by a relation of “OR” type, and analysis of standard elements connected by a relation of “NOT” type.
7. The method as recited in claim 1 , wherein said standard element correlations in the recognized object are expressed in the form of more than single-level structure.
8. The method as recited in claim 1 , wherein said standard elements at least partly contain portions of white color.
9. The method as recited in claim 1 , wherein said standard elements at least partly contain transparent portions.
10. The method as recited in claim 1 , wherein in the case of an ambiguous result of hypotheses testing setting up and testing a supplementary information.
11. The method as recited in claim 1 , wherein in the case of an ambiguous result of hypotheses testing setting up and testing supplementary recognition methods.
12. The method as recited in claim 1 , wherein the said standard element is composed of more prime standard elements of at least one type.
13. The method as recited in claim 1 , wherein the description of an object to be recognized is specified as a set of standard elements and spatially parametrical correlation thereof and is placed into a special means for storage and search.
14. The method of claim 1 , wherein the object to be recognized is a graphical object.
15. The method of claim 1 , wherein the object to be recognized is a character.
16. The method of claim 1 , wherein said selection is of a group of standard elements.