Automated tattoo recognition techniques
In some implementations, a computer-implemented method is capable of automatically segmenting and detecting a tattoo within an image. An image may be initially obtained. A block coverage pattern that identifies multiple blocks within the obtained image may be determined. A set of processing operations may then be performed for each block. The processing operations may include calculating a plurality of statistical features. A confidence score reflecting a likelihood that at least a portion of the block includes a predetermined graphical attribute associated with tattoos may be calculated. A subset of the multiple blocks of the image that have a respective confidence score greater than a predetermined threshold value may be identified. A portion of the image that includes one or more blocks from among the subset of the multiple blocks may then be determined to correspond to a tattoo.
1. A method performed by one or more computers, the method comprising:
obtaining an image;
obtaining block pattern data identifying multiple blocks within the image, the block pattern data comprising (i) a two-dimensional array of overlapping blocks for the image, each block within the two-dimensional array corresponding a particular region within the obtained image, and (ii) a subset of the multiple blocks of the image that have a respective confidence score satisfying a predetermined threshold value;
identifying candidate tattoo regions of the image based on the subset of the multiple blocks of the image that have a respective confidence score satisfying the predetermined threshold;
processing one or more attributes of each of the candidate tattoo regions; and
determining, based on processing the one or more attributes of each of the candidate tattoo regions, that the image includes a tattoo.
2. The method of claim 1 , wherein the confidence scores for each block included in the subset of multiple blocks represents a likelihood that at least a portion of a corresponding block includes a predetermined graphical attribute associated with tattoos.
3. The method of claim 1 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises an average confidence score for blocks included in a particular candidate tattoo region;
processing the one or more attributes of each of the candidate tattoo regions comprises identifying regions from among the candidate tattoo regions that have an average confidence score that exceeds a threshold value.
4. The method of claim 1 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises a number of blocks included in a particular candidate tattoo region;
processing the one or more attributes of each of the candidate tattoo regions comprises identifying regions from among the candidate tattoo regions that have a number of blocks that exceeds a threshold number of blocks.
5. The method of claim 1 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises a number of pixels included in a particular candidate tattoo region;
determining that the image includes a tattoo comprises determining, based on the number of pixels included in the candidate regions, whether the image includes a partial tattoo or a full tattoo.
6. The method of claim 5 , wherein determining whether the image includes a partial tattoo or a full tattoo comprises:
determining that at least one of the candidate tattoo regions has a number of pixels that exceeds a threshold number of pixels; and
based on determining that at least one of the candidate tattoo regions has a number of pixels that exceeds a threshold number of pixels, determining that the image includes a partial tattoo.
7. The method of claim 1 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises a shape that is identified as being included in a particular candidate tattoo region;
processing the one or more attributes of each of the candidate tattoo regions comprises determining that two or more regions from among the candidate tattoo regions includes a portion of the same shape; and
determining that the image includes a tattoo comprises determining that the image includes a single tattoo based on determining that two or more regions from among the candidate tattoo regions includes a portion of the same shape.
8. The method of claim 1 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises a shape that is identified as being included in a particular candidate tattoo region;
processing the one or more attributes of each of the candidate tattoo regions comprises determining that none of the regions from among the candidate tattoo regions includes a portion of the same shape; and
determining that the image includes a tattoo comprises determining that the image includes multiple tattoos based on none of the regions from among the candidate tattoo regions includes a portion of the same shape.
9. A system comprising:
one or more computers; and
one or more storage devices storing instructions that, when executed by the one or more computers, cause the one or more computers to perform operations comprising:
obtaining an image;
obtaining block pattern data identifying multiple blocks within the image, the block pattern data comprising (i) a two-dimensional array of overlapping blocks for the image, each block within the two-dimensional array corresponding a particular region within the obtained image, and (ii) a subset of the multiple blocks of the image that have a respective confidence score satisfying a predetermined threshold value;
identifying candidate tattoo regions of the image based on the subset of the multiple blocks of the image that have a respective confidence score satisfying the predetermined threshold;
processing one or more attributes of each of the candidate tattoo regions; and
determining, based on processing the one or more attributes of each of the candidate tattoo regions, that the image includes a tattoo.
10. The system of claim 9 , wherein the confidence scores for each block included in the subset of multiple blocks represents a likelihood that at least a portion of a corresponding block includes a predetermined graphical attribute associated with tattoos.
11. The system of claim 9 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises an average confidence score for blocks included in a particular candidate tattoo region;
processing the one or more attributes of each of the candidate tattoo regions comprises identifying regions from among the candidate tattoo regions that have an average confidence score that exceeds a threshold value.
12. The system of claim 9 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises a number of blocks included in a particular candidate tattoo region;
processing the one or more attributes of each of the candidate tattoo regions comprises identifying regions from among the candidate tattoo regions that have a number of blocks that exceeds a threshold number of blocks.
13. The system of claim 9 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises a number of pixels included in a particular candidate tattoo region;
determining that the image includes a tattoo comprises determining, based on the number of pixels included in the candidate regions, whether the image includes a partial tattoo or a full tattoo.
14. The system of claim 13 , wherein determining whether the image includes a partial tattoo or a full tattoo comprises:
determining that at least one of the candidate tattoo regions has a number of pixels that exceeds a threshold number of pixels; and
based on determining that at least one of the candidate tattoo regions has a number of pixels that exceeds a threshold number of pixels, determining that the image includes a partial tattoo.
15. The system of claim 9 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises a shape that is identified as being included in a particular candidate tattoo region;
processing the one or more attributes of each of the candidate tattoo regions comprises determining that two or more regions from among the candidate tattoo regions includes a portion of the same shape; and
determining that the image includes a tattoo comprises determining that the image includes a single tattoo based on determining that two or more regions from among the candidate tattoo regions includes a portion of the same shape.
16. A non-transitory computer-readable storage device encoded with computer program instructions that, when executed by one or more computers, cause the one or more computers to perform operations comprising:
obtaining an image;
obtaining block pattern data identifying multiple blocks within the image, the block pattern data comprising (i) a two-dimensional array of overlapping blocks for the image, each block within the two-dimensional array corresponding a particular region within the obtained image, and (ii) a subset of the multiple blocks of the image that have a respective confidence score satisfying a predetermined threshold value;
identifying candidate tattoo regions of the image based on the subset of the multiple blocks of the image that have a respective confidence score satisfying the predetermined threshold;
processing one or more attributes of each of the candidate tattoo regions; and
determining, based on processing the one or more attributes of each of the candidate tattoo regions, that the image includes a tattoo.
17. The non-transitory computer-readable storage device of claim 16 , wherein the confidence scores for each block included in the subset of multiple blocks represents a likelihood that at least a portion of a corresponding block includes a predetermined graphical attribute associated with tattoos.
18. The non-transitory computer-readable storage device of claim 16 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises an average confidence score for blocks included in a particular candidate tattoo region;
processing the one or more attributes of each of the candidate tattoo regions comprises identifying regions from among the candidate tattoo regions that have an average confidence score that exceeds a threshold value.
19. The non-transitory computer-readable storage device of claim 16 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises a number of blocks included in a particular candidate tattoo region;
processing the one or more attributes of each of the candidate tattoo regions comprises identifying regions from among the candidate tattoo regions that have a number of blocks that exceeds a threshold number of blocks.
20. The non-transitory computer-readable storage device of claim 16 , wherein:
the one or more attributes of each of the candidate tattoo regions comprises a number of pixels included in a particular candidate tattoo region;
determining that the image includes a tattoo comprises determining, based on the number of pixels included in the candidate regions, whether the image includes a partial tattoo or a full tattoo.