Methods and apparatus to determine the dimensions of a region of interest of a target object from an image using target object landmarks
Methods and apparatus to determine the dimensions of a region of interest of a target object and a class of the target object from an image using target object landmarks are disclosed herein. An example method includes identifying a landmark of a target object in an image based on a match between the landmark and a template landmark; classifying a target object based on the identified landmark; projecting dimensions of the template landmark based on a location of the landmark in the image; and determining a region of interest based on the projected dimensions, the region of interest corresponding to text printed on the target object.
1. An apparatus comprising:
a landmark detector implemented by at least one processor to identify a landmark in an image based on a match between the landmark and a template landmark; and
a corner determiner implemented by at least one processor to:
generate a first side search area based on a first template edge of the template landmark, the first side search area being a first region including a first landmark edge of the landmark in the image;
identify the first landmark edge by analyzing first transition points in the first side search area;
generate a second side search area based on a second template edge of the template landmark, the second side search area being a second region including a second landmark edge of the landmark; and
identify the second landmark edge by analyzing second transition points in the second side search area; and
identify a region of interest of the image based on the first landmark edge and the second landmark edge, the region of interest corresponding to text printed in the image.
2. The apparatus of claim 1 , wherein the landmark detector is to match the landmark to the template landmark by comparing the landmark to a plurality of template landmarks.
3. The apparatus of claim 1 , wherein the corner determiner is to transform a template landmark size and a template landmark orientation of the template landmark to match a landmark size and a landmark orientation of the landmark.
4. The apparatus of claim 1 , wherein the corner determiner is to:
determine a first corner of the region of interest based on a first intersection between a third template edge of the template landmark and the first landmark edge; and
determine a second corner of the region of interest based on a second intersection between the third template edge of the template landmark and the second landmark edge; and
determine a first side edge of the region of interest and a second side edge of the region of interest, the first side edge corresponding to the first landmark edge and the second side edge corresponding to the second landmark edge.
5. The apparatus of claim 4 , wherein the first side search area and the second side search area are limited in direction and area.
6. The apparatus of claim 4 , wherein the corner determiner is to determine text lines within a third region, the third region corresponding to the third template edge of the landmark, the first side edge, the second side edge, and a bottom image line corresponding to a bottom of the image.
7. The apparatus of claim 6 , wherein the corner determiner is to:
determine a third corner of the region of interest based on a third intersection between the first side edge and a text line closest to the bottom image line; and
determine a fourth corner of the region of interest based on a fourth intersection between the second side edge and the text line closest to the bottom image line.
8. The apparatus of claim 7 , wherein the first, second, third, and fourth corners of the region of interest improve an accuracy of text recognition.
9. The apparatus of claim 1 , wherein the landmark detector is to determine a class of an object included in the image based on the matched template landmark, the region of interest included in the object, the class including at least one of a company or a store.
10. An apparatus comprising:
first means for identifying a landmark in an image based on a match between the landmark and a template landmark; and
second means for:
generating a first side search area based on a first template edge of the template landmark, the first side search area being a first region including a first landmark edge of the landmark in the image;
identifying the first landmark edge by analyzing first transition points in the first side search area;
generating a second side search area based on a second template edge of the template landmark, the second side search area being a second region including a second landmark edge of the landmark; and
identifying the second landmark edge by analyzing second transition points in the second side search area; and
identifying a region of interest of the image based on the first landmark edge and the second landmark edge, the region of interest corresponding to text printed in the image.
11. The apparatus of claim 10 , wherein the first means is to match the landmark to the template landmark by comparing the landmark to a plurality of template landmarks.
12. The apparatus of claim 10 , wherein the second means is to transform a template landmark size and a template landmark orientation of the template landmark to match a landmark size and a landmark orientation of the landmark.
13. The apparatus of claim 10 , wherein the second means is to:
determine a first corner of the region of interest based on a first intersection between a third template edge of the template landmark and the first landmark edge; and
determine a second corner of the region of interest based on a second intersection between the third template edge of the template landmark and the second landmark edge; and
determine a first side edge of the region of interest and a second side edge of the region of interest, the first side edge corresponding to the first landmark edge and the second side edge corresponding to the second landmark edge.
14. The apparatus of claim 13 , wherein the second means is to determine text lines within a third region, the third region corresponding to the third template edge of the landmark, the first side edge, the second side edge, and a bottom image line corresponding to a bottom of the image.
15. The apparatus of claim 14 , wherein the second means is to:
determine a third corner of the region of interest based on a third intersection between the first side edge and a text line closest to the bottom image line; and
determine a fourth corner of the region of interest based on a fourth intersection between the second side edge and the text line closest to the bottom image line.
16. The apparatus of claim 10 , wherein the first means is to determine a class of an object included in the image based on the matched template landmark, the object including the region of interest, the class including at least one of a company or store.
17. A tangible computer readable storage medium comprising instructions which, when executed, cause a processor to at least:
identify a landmark in an image based on a match between the landmark and a template landmark;
determine a first landmark edge by analyzing transition points in a first side search area corresponding to a first edge of the template landmark;
determine a second landmark edge by analyzing second transition points in a second side search area corresponding to a second edge of the template landmark;
determine a first side edge corresponding to the first landmark edge and a second side edge corresponding to the second landmark edge;
identify text lines within a region defined by the first side edge and the second side edge; and
identify a region of interest based on the text lines, the region of interest corresponding to text printed in the image.
18. The tangible computer readable storage medium of claim 17 , wherein the instructions cause the processor to:
determine a first corner of the region of interest based on a first intersection between a third edge of the template landmark and the first landmark edge; and
determine a second corner of the region of interest based on a second intersection between the third edge of the template landmark and the second landmark edge.
19. The tangible computer readable storage medium of claim 18 , wherein the first edge is an edge of an object included in the image, the object including the region of interest, and wherein the instructions cause the processor to determine the first side edge of the object by:
extending the first side edge to an image line corresponding to an end of the image;
generating a plurality of lines between the first corner and a bottom image line within a threshold range of intersection; and
selecting the first side edge based on transition point counts for the plurality of lines.
20. The tangible computer readable storage medium of claim 18 , wherein the instructions cause the processor to identify the text lines based on dark-to-bright transition points.