Evaluating three-dimensional security features on document images
The disclosure includes a system and method for obtaining at least one image of a document under test, wherein valid instances of the document under test include at least one three-dimensional security feature; obtaining a first image snippet from the document under test, the first image snippet corresponding to at least a first portion of the three-dimensional security feature present in the valid instances of the document; deriving first dimensional data associated with the first image snippet; analyzing the first dimensional data to determine one or more of: whether the first dimensional data is consistent with a presence of the three-dimensional security feature, and whether the first dimensional data is consistent with second dimensional data; and modifying a likelihood that the document under test is accepted as valid, or rejected as invalid, based on the analysis of the first dimensional data.
1. A method comprising:
obtaining, using one or more processors, at least one image of a document, wherein valid instances of the document include at least a first three-dimensional security feature;
obtaining, using the one or more processors, a first image snippet from the document, the first image snippet corresponding to at least a first portion of the first three-dimensional security feature present in the valid instances of the document;
deriving, using the one or more processors, first dimensional data associated with the first image snippet;
analyzing, using the one or more processors, the first dimensional data to determine one or more of:
whether the first dimensional data is consistent with a presence of the first three-dimensional security feature, and
whether the first dimensional data is consistent with second dimensional data, wherein the second dimensional data corresponds to one of: a second portion of the first three-dimensional security feature in the valid instances of the document and at least a portion of a second three-dimensional security feature in the valid instances of the document; and
modifying, using one or more processors, a likelihood that the document under test is accepted as valid, or rejected as invalid, based on the analysis of the first dimensional data.
2. The method of claim 1 , the method further comprising:
obtaining a second image snippet from the image of the document; and
deriving the second dimensional data associated with the second image snippet,
wherein the first three-dimensional security feature includes a set of tactile text,
wherein the first dimensional data and first image snippet correspond to the first portion of the first three-dimensional security feature by representing a first character in the set of tactile text, and
wherein the second dimensional data and second image snippet correspond to the second portion of the first three-dimensional security feature by representing a second character in the set of tactile text.
3. The method of claim 1 , wherein deriving the first dimensional data includes:
determining, for each pixel in a set of pixels within the first image snippet, an intensity;
assigning a first subset of pixels that satisfy an intensity threshold to a first color;
assigning a second subset of pixels that fail to satisfy the intensity threshold to a second color; and
generating, as the first dimensional data, a first and second color version of the first image snippet.
4. The method of claim 1 , wherein determining whether the first dimensional data is consistent with the presence of the first three-dimensional security feature includes:
determining whether the first dimensional data is indicative of specular reflection; and
responsive to determining that the first dimensional data is not indicative of specular reflection, determining that first dimensional data is inconsistent with the presence of the first three-dimensional security feature.
5. The method of claim 1 , wherein deriving the first dimensional data includes:
identifying color variation within the first image snippet, wherein a color variation in portions of the first three-dimensional security feature known to have a common color is determined to be indicative of three-dimensionality.
6. The method of claim 1 , wherein deriving the first dimensional data includes:
applying nearest neighbor analysis to the first image snippet.
7. The method of claim 1 , wherein deriving the first dimensional data includes:
generating a depth map from stereoscopic images of the document, the depth map indicating whether portion of the document associated with the first image snippet is three dimensional.
8. The method of claim 1 , wherein determining whether the first dimensional data is consistent with the second dimensional data is based at least in part on determining a common light source.
9. The method of claim 8 , wherein determining the common light source is based on one or more of specular reflection and ray tracing.
10. The method of claim 1 , wherein determining whether the first dimensional data is consistent with second dimensional data is determined subsequent to determining that the first dimensional data is consistent with the presence of the first three-dimensional security feature.
11. A system comprising:
a processor; and
a memory, the memory storing instructions that, when executed by the processor, cause the system to:
obtain at least one image of a document, wherein valid instances of the document include at least a first three-dimensional security feature;
obtain a first image snippet from the document-under test, the first image snippet corresponding to at least a first portion of the first three-dimensional security feature present in the valid instances of the document;
derive first dimensional data associated with the first image snippet;
analyze the first dimensional data to determine one or more of:
whether the first dimensional data is consistent with a presence of the first three-dimensional security feature, and
whether the first dimensional data is consistent with second dimensional data, wherein the second dimensional data corresponds to one of: a second portion of the first three-dimensional security feature in the valid instances of the document and at least a portion of a second first three-dimensional security feature in the valid instances of the document; and
modify a likelihood that the document is accepted as valid, or rejected as invalid, based on the analysis of the first dimensional data.
12. The system of claim 11 , includes instructions that, when executed, cause the system to:
obtain a second image snippet from the image of the document; and
derive the second dimensional data associated with the second image snippet,
wherein the first three-dimensional security feature includes a set of tactile text,
wherein the first dimensional data and first image snippet correspond to the first portion of the first three-dimensional security feature by representing a first character in the set of tactile text, and
wherein the second dimensional data and second image snippet correspond to the second portion of the first three-dimensional security feature by representing a second character in the set of tactile text.
13. The system of claim 11 , wherein deriving the first dimensional data includes:
determining, for each pixel in a set of pixels within the first image snippet, an intensity;
assigning a first subset of pixels that satisfy an intensity threshold to a first color;
assigning a second subset of pixels that fail to satisfy the intensity threshold to a second color; and
generating, as the first dimensional data, a first and second color version of the first image snippet.
14. The system of claim 11 , wherein determining whether the first dimensional data is consistent with the presence of the first three-dimensional security feature includes:
determining whether the first dimensional data is indicative of specular reflection; and
responsive to determining that the first dimensional data is not indicative of specular reflection, determining that first dimensional data is inconsistent with the presence of the first three-dimensional security feature.
15. The system of claim 11 , wherein deriving the first dimensional data includes:
identifying color variation within the first image snippet, wherein a color variation in portions of the first three-dimensional security feature known to have a common color is determined to be indicative of three-dimensionality.
16. The system of claim 11 , wherein deriving the first dimensional data includes:
applying nearest neighbor analysis to the first image snippet.
17. The system of claim 11 , wherein deriving the first dimensional data includes:
generating a depth map from stereoscopic images of the document, the depth map indicating whether portion of the document associated with the first image snippet is three dimensional.
18. The system of claim 11 , wherein determining whether the first dimensional data is consistent with the second dimensional data is based at least in part on determining a common light source.
19. The system of claim 18 , wherein determining the common light source is based on one or more of specular reflection and ray tracing.
20. The system of claim 11 , wherein determining whether the first dimensional data is consistent with second dimensional data is determined subsequent to determining that the first dimensional data is consistent with the presence of the first three-dimensional security feature.