Bit-mapped image multi-stage analysis method
View Patent ↗A method is described of bit-mapped image analysis comprising division of all analysis means at one's disposal into several groups differing in accuracy and further processing multi-stage analysis. The analysis comprises a primary analysis stage and at least one profound analysis stage, with supplemental data collected at both stages. The primary analysis, includes preliminary recognition of objects with distortion and detection of objects that require more precise analysis means to overcome the distortion. At the primary analysis stage, the analysis means from the group of the most inaccurate group are used. The profound analysis stage includes repeating recognition of objects with a distortion taking into account the supplemental data obtained at the previous stage, detecting objects that require more precise analysis means to overcome the distortion, and collection of newly appeared supplemental data. Each subsequent profound analysis stage uses analysis means from the group of more accurate means.
1. A method of bit-mapped image analysis comprising
preliminary division of all analysis means at the disposal into two or more groups differing in accuracy of results;
performing a multi-stage analysis, said analysis further comprising at least
a primary analysis stage along with collection of supplemental data, said primary analysis further comprising at least parsing the image into objects, detecting among said objects those with distortion that embarrasses character recognition, preliminary recognition of said objects with the distortion, detecting among said objects those requiring more accurate analysis means to overcome the influence of distortion, said primary analysis stage using the analysis means from group of most inaccurate means;
at least one profound analysis stage along with collection of newly appeared supplemental data, said profound analysis stage further comprising repeating of recognition of said objects with distortion that were not recognized at the previous stage taking into account the supplemental data obtained at the previous stage, detecting among said objects those requiring more accurate analysis means to overcome the influence of distortion, collection of newly appeared supplemental data, making a decision about the necessity of repeating of the profound analysis stage taking into account the whole supplemental data collected at the previous stages or finishing the analysis process, each subsequent profound analysis stage using the analysis means from the group of more accurate means.
2. The method of bit-mapped image analysis as recited in claim 1 , further comprising outputting results to a subsequent process.
3. The method of bit-mapped image analysis as recited in claim 2 , wherein the subsequent process is a character recognition process.
4. The method of bit-mapped image analysis as recited in claim 1 , further comprising additional correction of objects parsing at the profound analysis stage.
5. A method of bit-mapped image analysis as recited in claim 1 , wherein the decision about whether or not to perform repeating of the profound analysis stage is made taking into account a predetermined recognition accuracy level.
6. A method of bit-mapped image analysis as recited in claim 1 , wherein the decision not to process the repeating of the profound analysis stage is made after exhausting all analysis means at the disposal.
7. A method of bit-mapped image analysis as recited in claim 1 , wherein the distortion may appear as cursive character matter.
8. A method of bit-mapped image analysis as recited in claim 1 , wherein the distortion may appear as inverted character matter.
9. A method of bit-mapped image analysis as recited in claim 1 , wherein the said distortion may appear as extra dots in the image.
10. A method of bit-mapped image analysis as recited in claim 1 , wherein the said distortion may appear as missing dots in the image.
11. A method of bit-mapped image analysis as recited in claim 1 , wherein the distortion may appear as a skew of the image.
12. A method of bit-mapped image analysis as recited in claim 1 , further comprising an additional step of removing distortion from the image.
13. A method of bit-mapped image analysis as recited in claim 1 , wherein supplemental data may be represented as a correlation among the parameters of the characters within the object.
14. A method of bit-mapped image analysis as recited in claim 1 , wherein supplemental data may be represented as correlation of parameters within the line.
15. A method of bit-mapped image analysis as recited in claim 1 , wherein supplemental data may be represented as correlation of parameters between plurality of objects.
16. A method of bit-mapped image analysis as recited in claim 1 , wherein at least one supplemental image analysis means of other kind may be used before the character recognition process.