Method for determining width of lines in hand drawn table

Patent metadata

Field	Value
Publication number	US-9898653-B2
Application number	US-201615164302-A
Country	US
Kind code	B2
Filing date	May 25, 2016
Priority date	May 25, 2016
Publication date	Feb 20, 2018
Grant date	Feb 20, 2018

Abstract

A method for image processing includes obtaining a mask of a stroke from an image; determining a plurality of cross edges for the stroke based on the mask; generating a histogram comprising a plurality of widths of the cross edges and a plurality of frequencies of the plurality of widths from the cross edges; estimating a lower bound of a width range for the stroke based on a mode width of the plurality of widths, a first subset of the plurality of widths below the mode width, and a first plurality of weights assigned to the first subset of the plurality of widths; and estimating an upper bound of the width range for the stroke based on the mode width, a second subset of the plurality of widths above the mode width, and a second plurality of weights assigned to the second subset of the widths.

First claim

What is claimed is:

1. A method for image processing, comprising:
obtaining a mask of a stroke from an image;
determining a plurality of cross edges for the stroke based on the mask;
generating a histogram comprising a plurality of widths of the cross edges and a plurality of frequencies of the plurality of widths from the cross edges;
estimating a lower bound of a width range for the stroke based on
a mode width of the plurality of widths,
a first subset of the plurality of widths below the mode width, and
a first plurality of weights assigned to the first subset of the plurality of widths; and

estimating an upper bound of the width range for the stroke based on the mode width,
a second subset of the plurality of widths above the mode width, and
a second plurality of weights assigned to the second subset of the plurality of widths.

2. The method of claim 1 , wherein estimating the lower bound comprises:
calculating a first weight for a first width of the first subset based on a distance between the lower bound and the first width, wherein the lower bound is initialized by the mode width;
generating a first product by multiplying the first weight and a frequency corresponding to the first width;
adding the first product to a first sum initialized by zero;
comparing the first sum with a width threshold; and
decrementing the lower bound when the first sum is greater than the width threshold.

3. The method of claim 2 , wherein when the first sum is smaller than or equal to the width threshold, estimating the lower bound further comprises:
generating a second weight for a second width of the first subset based on a distance between the lower bound and the second width;
generating a second product by multiplying the second weight and a frequency corresponding to the second width;
adding the second product to the first sum;
comparing the first sum added by the second product with the width threshold; and
decrementing the lower bound when the first sum added by the second product is greater than the width threshold.

4. The method of claim 1 , wherein estimating the upper bound comprises:
calculating a third weight for a third width of the second subset based on a distance between the upper bound and the third width, wherein the upper bound is initialized by the mode width;
generating a third product by multiplying the third weight and a frequency corresponding to the third width;
adding the third product to a second sum initialized by zero;
comparing the second sum with a width threshold; and
incrementing the upper bound when the second sum is greater than the width threshold.

5. The method of claim 4 , wherein when the second sum is smaller than or equal to the width threshold, estimating the upper bound further comprises:
generating a fourth weight for a fourth width of the second subset based on a distance between the upper bound and the fourth width;
generating a fourth product by multiplying the fourth weight and a frequency corresponding to the fourth width;
adding the fourth product to the second sum;
comparing the second sum added by the fourth product with the width threshold; and
incrementing the upper bound when the second sum added by the fourth product is greater than the width threshold.

6. The method of claim 2 , wherein the width threshold is initialized by a frequency of the mode width, multiplied by a predetermined percentage.

7. The method of claim 6 , wherein
the predetermined percentage is 5%.

8. The method of claim 2 , further comprising updating the width threshold based on the first sum and a predetermined percentage, wherein estimating the lower bound further comprises:
calculating a fifth weight for a fifth width of the first subset based on a distance between the lower bound and the fifth width;
generating a fifth product by multiplying the fifth weight and a frequency corresponding to the fifth width;
adding the fifth product to the first sum reinitialized by zero;
comparing the first sum added by the fifth product with the updated width threshold; and
decrementing the lower bound when the first sum added by the fifth product is greater than the updated width threshold.

9. The method of claim 4 , further comprising updating the width threshold based on the second sum and a predetermined percentage, wherein estimating the upper bound further comprises:
calculating a sixth weight for a sixth width of the second subset based on a distance between the upper bound and the sixth width;
generating a sixth product by multiplying the sixth weight and a frequency corresponding to the sixth width;
adding the sixth product to the second sum reinitialized by zero;
comparing the second sum added by the sixth product with the updated width threshold; and
incrementing the upper bound when the second sum added by the sixth product is greater than the updated width threshold.

10. The method of claim 2 , wherein estimating the upper bound comprises:
calculating a third weight for a third width of the second subset based on a distance between the upper bound and the third width;
generating a third product by multiplying the third weight and a frequency corresponding to the third width;
adding the third product to a second sum initialized by zero;
comparing the second sum with the width threshold; and
incrementing the upper bound when the second sum is greater than the width threshold, and
the upper bound is initialized by the mode width.

11. The method of claim 10 , further comprising updating the width threshold with a sum of the first sum and the second sum, multiplied by a predetermined percentage.

12. A non-transitory computer readable medium (CRM) storing computer readable program code embodied therein that:
obtains a mask of a stroke from an image;
determines a plurality of cross edges for the stroke based on the mask;
generates a histogram comprising a plurality of widths of the cross edges and a plurality of frequencies of the plurality of widths from the cross edges;
estimates a lower bound of a width range for the stroke based on
a mode width of the plurality of widths,
a first subset of the plurality of widths below the mode width, and
a first plurality of weights assigned to the first subset of the plurality of widths; and

estimates an upper bound of the width range for the stroke based on the mode width,
a second subset of the plurality of widths above the mode width, and
a second plurality of weights assigned to the second subset of the plurality of widths.

13. The non-transitory CRM of claim 12 , wherein estimating the lower bound comprises:
calculating a first weight for a first width of the first subset based on a distance between the lower bound and the first width, wherein the lower bound is initialized by the mode width;
generating a first product by multiplying the first weight and a frequency corresponding to the first width;
adding the first product to a first sum initialized by zero;
comparing the first sum with a width threshold; and
decrementing the lower bound when the first sum is greater than the width threshold.

14. The non-transitory CRM of claim 13 , wherein when the first sum is smaller than or equal to the width threshold, estimating the lower bound further comprises:
generating a second weight for a second width of the first subset based on a distance between the lower bound and the second width;
generating a second product by multiplying the second weight and a frequency corresponding to the second width;
adding the second product to the first sum;
comparing the first sum added by the sec

Assignees

• Konica Minolta Laboratory Usa Inc
• Konica Minolta Laboratory Usa Inc

Inventors

• Bellert Darrell Eugene

Classifications

• G06V10/44

  Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components · CPC title

• G06V10/50

  by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis · CPC title

• G06V10/758

  Involving statistics of pixels or of feature values, e.g. histogram matching · CPC title

• G06K9/6212

  Physics · mapped topic

• G06K9/00416 Primary

  Physics · mapped topic