Merging non-overlapping broken lines of a table
A method for image processing includes obtaining a mask of an image; identifying, in the mask, a first stroke segment and a second stroke segment for possible merging, wherein the first stroke segment and the second stroke segment do not overlap; calculating a plurality of merger eligibility scores for the first stroke segment and the second stroke segment; calculating an aggregate merger eligibility score for the first stroke segment and the second stroke segment based on the plurality of merger eligibility scores; and merging the first stroke segment and the second stroke segment based on the aggregate merger eligibility score.
1. A method for image processing, comprising:
obtaining a mask of an image;
identifying, in the mask, a first stroke segment and a second stroke segment for possible merging, wherein the first stroke segment and the second stroke segment do not overlap;
calculating a plurality of merger eligibility scores for the first stroke segment and the second stroke segment;
calculating an aggregate merger eligibility score for the first stroke segment and the second stroke segment based on the plurality of merger eligibility scores; and
merging the first stroke segment and the second stroke segment based on the aggregate merger eligibility score,
wherein calculating the plurality of merger eligibility scores comprises:
calculating a length of a gap between the first stroke segment and the second stroke segment;
comparing the length of the gap with a length of the second stroke segment, wherein the second stroke segment is shorter than the first stroke segment; and
setting a closeness score to the smaller of:
an upper bound; and
a ratio of the length of the second stroke segment to the length of the gap, wherein the plurality of merger eligibility scores comprises the closeness score.
2. The method of claim 1 , wherein calculating the plurality of merger eligibility scores comprises:
calculating a width difference between the first stroke segment and the second stroke segment by subtracting a maximum probable width of the first stroke segment from a maximum probable width of the second stroke segment,
wherein the second stroke segment is shorter than the first stroke segment, and
wherein the width difference equals or exceeds zero;
setting a width score to the smaller of:
an upper bound; and
a ratio of the maximum probable width of the first stroke segment to the width difference,
wherein the plurality of merger eligibility scores comprises the width score.
3. The method of claim 1 , wherein calculating the plurality of merger eligibility scores comprises:
calculating a width difference between the first stroke segment and the second stroke segment by subtracting a maximum probable width of the first stroke segment from a maximum probable width of the second stroke segment,
wherein the second stroke segment is shorter than the first stroke segment, and
wherein the width difference is less than zero;
setting a width score to the smaller of:
an upper bound; and
a ratio of one-half of the maximum probable width of the first stroke segment to the absolute value of the width difference,
wherein the plurality of merger eligibility scores comprises the width score.
4. The method of claim 1 , wherein calculating the plurality of merger eligibility scores comprises:
calculating an absolute value of an angular difference between a first vector that represents a momentum of the first stroke segment and a second vector from the first stroke segment to the second stroke segment; and
setting an alignment score to the greater of:
a lower bound; and
a ratio of the absolute value of the angular difference subtracted from a predetermined angle to the predetermined angle,
wherein the plurality of merger eligibility scores comprises the alignment score.
5. The method of claim 4 , wherein
the first vector is calculated based on:
a midpoint of a trailing cross edge in the first stroke segment; and
a midpoint of a terminal cross edge in the first stroke segment; and
the second vector is calculated based on:
the midpoint of the terminal cross edge in the first stroke segment; and
a midpoint of a commencing cross edge in the second stroke segment.
6. The method according to claim 4 , wherein the predetermined angle is 45 degrees.
7. The method according to claim 1 , wherein calculating the plurality of merger eligibility scores comprises:
calculating an absolute value of an angular difference between a first vector that represents a direction of the first stroke segment and a second vector that represents a direction of the second stroke segment; and
setting an angular score to the greater of:
a lower bound; and
a ratio of the absolute value of the angular difference subtracted from a predetermined angle to the predetermined angle,
wherein the plurality of merger eligibility scores comprises the angular score.
8. The method of claim 7 , wherein
the first vector is calculated based on:
a midpoint of a commencing cross edge in the first stroke segment; and
a midpoint of a terminal cross edge in the first stroke segment; and
the second vector is calculated based on:
a midpoint of a commencing cross edge in the second stroke segment; and
a midpoint of a terminal cross edge in the second stroke segment.
9. The method according to claim 1 , wherein calculating the aggregate merger eligibility score comprises calculating a weighted average of the plurality of merger eligibility scores.
10. A non-transitory computer readable medium (CRM) storing computer readable program code embodied therein that:
obtains a mask of an image;
identifies, in the mask, a first stroke segment and a second stroke segment for possible merging, wherein the first stroke segment and the second stroke segment do not overlap;
calculates a plurality of merger eligibility scores for the first stroke segment and the second stroke segment;
calculates an aggregate merger eligibility score for the first stroke segment and the second stroke segment based on the plurality of merger eligibility scores; and
merges the first stroke segment and the second stroke segment based on the aggregate merger eligibility score,
wherein calculating the plurality of merger eligibility scores comprises:
calculating a length of a gap between the first stroke segment and the second stroke segment;
comparing the length of the gap with a length of the second stroke segment,
wherein the second stroke segment is shorter than the first stroke segment; and
setting a closeness score to the smaller of:
an upper bound; and
a ratio of the length of the second stroke segment to the length of the gap,
wherein the plurality of merger eligibility scores comprises the closeness score.
11. The non-transitory CRM of claim 10 , wherein calculating the plurality of merger eligibility scores comprises:
calculating a width difference between the first stroke segment and the second stroke segment by subtracting a maximum probable width of the first stroke segment from a maximum probable width of the second stroke segment,
wherein the second stroke segment is shorter than the first stroke segment, and
wherein the width difference equals or exceeds zero;
setting a width score to the smaller of:
an upper bound; and
a ratio of the maximum probable width of the first stroke segment to the width difference,
wherein the plurality of merger eligibility scores comprises the width score.
12. The non-transitory CRM of claim 10 , wherein calculating the plurality of merger eligibility scores comprises:
calculating a width difference between the first stroke segment and the second stroke segment by subtracting a maximum probable width of the first stroke segment from a maximum probable width of the second stroke segment,
wherein the second stroke segment is shorter than the first stroke segment, and
wherein the width difference is less than zero;
setting a width score to the smaller of:
an upper bound; and
a ratio of one-half of the maximum probable width of the first stroke segment to the absolute value of the width difference,
wherein the plurality of merger eligibility scores comprises the width score.
13. The non-transitory CRM of claim 10 , wherein calculating the plurality of merger eligibility scores comprises:
calculating an absolute value of an angular difference between a first vector that represents a momentum of the first stroke segment and a second vector from the first stroke segment to the second stroke segment; and
setting an alignment score to the greater of:
a lower bound; and
a ratio of the absolute value of the angular difference subtracted from a predetermined angle to the predetermined angle,
wherein the plurality of merger eligibility scores comprises the alignment score.
14. The non-transitory CRM of claim 10 , wherein calculating the plurality of merger eligibility scores comprises:
calculating an absolute value of an angular difference between a first vector that represents a direction of the first stroke segment and a second vector that represents a direction of the second stroke segment; and
setting an angular score to the greater of:
a lower bound; and
a ratio of the absolute value of the angular difference subtracted from a predetermined angle to the predetermined angle,
wherein the plurality of merger eligibility scores comprises the angular score.
15. An image processing system comprising:
a memory; and
a processor that:
obtains a mask of an image;
identifies, in the mask, a first stroke segment and a second stroke segment for possible merging, wherein the first stroke segment and the second stroke segment do not overlap;
calculates a plurality of merger eligibility scores for the first stroke segment and the second stroke segment;
calculates an aggregate merger eligibility score for the first stroke segment and the second stroke segment based on the plurality of merger eligibility scores; and
merges the first stroke segment and the second stroke segment based on the aggregate merger eligibility score,
wherein calculating the plurality of merger eligibility scores comprises:
calculating a length of a gap between the first stroke segment and the second stroke segment;
comparing the length of the gap with a length of the second stroke segment,
wherein the second stroke segment is shorter than the first stroke segment; and
setting a closeness score to the smaller of:
an upper bound; and
a ratio of the length of the second stroke segment to the length of the gap,
wherein the plurality of merger eligibility scores comprises the closeness score.
16. The image processing system of claim 15 , wherein calculating the plurality of merger eligibility scores comprises:
calculating an absolute value of an angular difference between a first vector that represents a momentum of the first stroke segment and a second vector from the first stroke segment to the second stroke segment; and
setting an alignment score to the greater of:
a lower bound; and
a ratio of the absolute value of the angular difference subtracted from a predetermined angle to the predetermined angle,
wherein the plurality of merger eligibility scores comprises the alignment score.
17. The image processing system of claim 15 , wherein calculating the plurality of merger eligibility scores comprises:
calculating an absolute value of an angular difference between a first vector that represents a direction of the first stroke segment and a second vector that represents a direction of the second stroke segment; and
setting an angular score to the greater of:
a lower bound; and
a ratio of the absolute value of the angular difference subtracted from a predetermined angle to the predetermined angle,
wherein the plurality of merger eligibility scores comprises the angular score.