Image identification apparatus and non-transitory computer readable medium

1. An image identification apparatus comprising:

a processor programmed to:

extract lines from an image;

exclude from objects to be identified a boundary delimiting an entire area of the image among the extracted lines; and

identify as an object a plurality of lines that are among the extracted lines and that are not excluded from the objects to be identified if the plurality of lines are connected to each other.

2. The image identification apparatus according to claim 1 , wherein the processor is further programmed to:

exclude, as a region that corresponds to the boundary delimiting the entire area of the image, a region that is largest among regions that are included in the image and in which pixels of the extracted lines are aligned continuously.

3. The image identification apparatus according to claim 1 , wherein the processor is further programmed to:

exclude a portion of the boundary delimiting the entire area of the image, the portion being located in a region that is not included in a region that overlaps a table region extracted from the image.

4. The image identification apparatus according to claim 2 , wherein the processor is further programmed to:

exclude a portion of the boundary delimiting the entire area of the image, the portion being located in a region that is not included in a region that overlaps a table region extracted from the image.

5. The image identification apparatus according to claim 1 , wherein the processor is further programmed to:

exclude, as the boundary delimiting the entire area of the image, a line that is enclosed by a smallest substantially rectangular frame whose size is largest among smallest substantially rectangular frames by which the extracted lines are enclosed.

6. The image identification apparatus according to claim 5 , wherein the processor is further programmed to:

exclude, as the boundary delimiting the entire area of the image, a line that is enclosed by a smallest substantially rectangular frame whose size is largest among the smallest substantially rectangular frames by which the extracted lines are enclosed, the smallest substantially rectangular frame being larger than or equal to a predetermined first threshold.

7. The image identification apparatus according to claim 6 , wherein the predetermined first threshold is a threshold determined in accordance with a size of a recording medium if the image is a read image obtained by reading an image recorded on the recording medium.

8. The image identification apparatus according to claim 1 , wherein the processor is further programmed to:

exclude, as the boundary delimiting the entire area of the image, a line whose length in a predetermined first direction is longest among lengths of the extracted lines.

9. The image identification apparatus according to claim 8 , wherein the processor is further programmed to:

exclude, as the boundary delimiting the entire area of the image, a line whose length in the predetermined first direction is longest among the lengths of the extracted lines and is longer than a predetermined second threshold.

10. The image identification apparatus according to claim 9 , wherein the predetermined second threshold is a threshold determined in accordance with a size of a recording medium if the image is a read image obtained by reading an image recorded on the recording medium.

11. The image identification apparatus according to claim 1 , wherein the processor is further programmed to:

exclude, as the boundary delimiting the entire area of the image, a line that is extracted by the processor when the line is located in a predetermined peripheral region in the image.

12. The image identification apparatus according to claim 11 , wherein the predetermined peripheral region is a peripheral region determined in accordance with a size of a recording medium if the image is a read image obtained by reading an image recorded on the recording medium.

13. The image identification apparatus according to claim 1 , wherein the processor is further programmed to:

identify, as a table region, a region enclosed by a plurality of first lines substantially parallel to a predetermined first direction and a plurality of second lines substantially parallel to a second direction that intersects the first direction if the plurality of the first lines intersect the plurality of the second lines.

14. The image identification apparatus according to claim 1 , wherein the processor is further programmed to:

calculate first coordinates of the boundary delimiting the entire area of the image in the image;

regard a plurality of regions as being aligned continuously in a predetermined first direction if the plurality of regions are aligned continuously in a second direction that intersects the first direction, each of the plurality of regions including a group of pixels of the extracted lines aligned continuously in the first direction; and

exclude the boundary delimiting the entire area of the image using the first coordinates of the boundary delimiting the entire area of the image.

15. The image identification apparatus according to claim 14 , wherein the processor is further programmed to:

regard the plurality of regions as being aligned continuously in the first direction if a sum of lengths of the plurality of regions in the first direction is larger than or equal to a predetermined third threshold; and

calculate the first coordinates of the boundary delimiting the entire area of the image in the image.

16. The image identification apparatus according to claim 15 , wherein the predetermined third threshold is a threshold determined in accordance with a size of a recording medium if the image is a read image obtained by reading an image recorded on the recording medium.

17. The image identification apparatus according to claim 1 , wherein the processor is further programmed to:

calculate second coordinates of the boundary delimiting the entire area of the image in the image;

regard regions in the image as being aligned continuously in a predetermined first direction when the regions, each including a group of pixels of the extracted lines aligned continuously in the predetermined first direction, are aligned with one or more gaps interposed therebetween; and

exclude the boundary delimiting the entire area of the image using the second coordinates of the boundary delimiting the entire area of the image.

18. The image identification apparatus according to claim 17 , wherein the processor is further programmed to:

regard the regions as being aligned continuously in the predetermined first direction when a ratio of a sum of lengths of the regions in the predetermined first direction to a sum of lengths of the one or more gaps in the predetermined first direction satisfies a predetermined requirement; and

calculate the second coordinates of the boundary delimiting the entire area of the image in the image.

19. A non-transitory computer readable medium storing a program causing a computer to execute a process for image identification, the process comprising:

extracting lines from an image;

excluding from objects to be identified a boundary delimiting an entire area of the image among the extracted lines; and

identifying as an object a plurality of lines that are among the extracted lines and that are not excluded if the plurality of lines are connected to each other.

20. An image identification apparatus comprising:

extraction means for extracting lines from an image;

exclusion means for excluding from objects to be identified a boundary delimiting an entire area of the image among the extracted lines; and

identification means for identifying as an object a plurality of lines that are among the extracted lines and that are not excluded by the exclusion means if the plurality of lines are connected to each other.