Signature authentications based on features

Abstract

An example system includes a feature extraction engine. The feature extraction engine is to determine a plurality of scale-dependent features for a portion of a target. The system also includes a signature-generation engine to select a subset of the plurality of scale-dependent features based on a strength of each feature. The signature-generation engine also is to store a numeric representation of the portion of the target and the subset of the plurality of scale-dependent features.

What is claimed is: 
     
         1 . A system comprising:
 a processor; and   a memory storing instructions executable by the processor to:
 convolve a kernel with an image of a portion of a target to determine a plurality of scale-dependent features for the portion of the target; 
 select a subset of the plurality of scale-dependent features based on a strength of each feature; and 
 store a numeric representation of the portion of the target and the subset of the plurality of scale-dependent features, 
   wherein the numeric representation is a thumbnail image of the portion of the target,   and wherein the numeric representation and the subset of the plurality of scale-dependent features form an authentication signature of the target.   
     
     
         2 . The system of  claim 1 , wherein the instructions are executable by the processor to determine the plurality of scale-dependent features in a small frequency range without determining features at frequencies other than the small frequency range. 
     
     
         3 . The system of  claim 2 , further comprising an image sensor to capture the image of the portion of the target, wherein the instructions are executable by the processor to determine the plurality of scale-dependent features based on the image. 
     
     
         4 . The system of  claim 3 , wherein the plurality of scale-dependent features in the small frequency range are reproducibly detectable across multiple images of the portion of the target and distinct between the target and duplicates of the target. 
     
     
         5 . The system of  claim 1 , wherein the target is selected from the group consisting of a two-dimensional print and a three-dimensional print. 
     
     
         6 . The system of  claim 1 , wherein the instructions are executable by the processor to further:
 authenticate a purported target using the numeric representation and the subset of the plurality of scale-dependent features.   
     
     
         7 . The system of  claim 6 , further comprising an image sensor to display the thumbnail image of the portion of the target and to capture an image of a corresponding portion of the purported target, the instructions executable by the processor to authenticate the purported target by:
 convolving the kernel with the image of the corresponding portion of the purported target to determine a plurality of scale-dependent features for the corresponding portion of the purported target;   select a subset of the plurality of scale-dependent features for the corresponding portion of the purported target based on a strength of each feature;   form an authentication signature of the purported target from a numeric representation of the portion of purported target and the subset of the plurality of scale-dependent features for the portion of the purported target; and   compare the authentication signature of the purported target against the authentication signature of the target.   
     
     
         8 . The system of  claim 7 , wherein the image sensor is to display the thumbnail image of the portion of the target as a ghost image that aligns with the corresponding portion of the purported target to a user,
 and wherein the image sensor is to capture the image of the corresponding portion of the purported target as has been user-selected.   
     
     
         9 . A method comprising:
 displaying a thumbnail image of a portion of a target;   capturing an image of a corresponding portion of a purported target;   convolving, using a processor, a kernel with the image to identify scale-dependent features in the image;   selecting, using the processor, a plurality of the scale-dependent features from the image without regard to distance between feature locations; and   generating, using the processor, an authentication signature of the purported target based on the selected scale-dependent features.   
     
     
         10 . The method of  claim 9 , further comprising comparing, using the processor, the authentication signature of the purported target against an authentication signature of the target to authenticate the purported target. 
     
     
         11 . The method of  claim 9 , further comprising authenticating, using the processor, the purported target by:
 convolving, using the processor, the kernel with an image of the portion of the target to identify scale-dependent features in the image of the portion of the target;   selecting, using the processor, a predetermined number of the scale-dependent features separated in location by at least a predetermined distance from the image of the portion of the target, the predetermined number less than a number of the plurality of the scale-dependent features selected from the image of the corresponding portion of the purported target; and   determining whether a subset of the plurality of scale-dependent features selected from the image of the corresponding portion of the purported target most closely the predetermined number of scale-dependent features selected from the image of the portion of the target satisfies a matching threshold.   
     
     
         12 . The method of  claim 11 , wherein selecting the predetermined number of the scale-dependent features separated in location by at least the predetermined distance comprises selecting scale-dependent features of a first polarity separated from each other by the predetermined distance and selecting scale-dependent features of a second polarity separated from each other by the predetermined distance regardless of distances between scale-dependent features of different polarities. 
     
     
         13 . The method of  claim 11 , wherein selecting the predetermined number of the scale-dependent features comprises selecting the predetermined number of the scale-dependent features at a small frequency range low enough to limit selecting the plurality of features corresponding to noise within the captured image and high enough to limit selecting the plurality of features present in duplicates of the purported target without selecting the plurality of features at high and low frequencies. 
     
     
         14 . The method of  claim 11 , wherein selecting the predetermined number of the scale-dependent features comprises selecting features resulting from unique variations in a printing process without detecting features resulting from print content. 
     
     
         15 . The method of  claim 9 , wherein displaying the thumbnail image of the portion of the target comprises displaying a ghost image that aligns with the corresponding portion of the purported target to a user,
 and wherein capturing the image of the corresponding portion of the purported target comprises capturing a user-selected image of the corresponding portion of the purported target.   
     
     
         16 . A non-transitory computer-readable medium comprising instructions that, when executed by a processor, cause the processor to:
 display a thumbnail image of a portion of a target;   convolve a kernel with an image of a corresponding portion of a purported target to determine a plurality of scale-dependent features for the portion of the purported target;   select a subset of the plurality of scale-dependent features with descriptors having a smallest mathematically determined distance from descriptors of a predetermined set of features of the target; and   determine an authenticity of the purported target based on whether the smallest mathematically determined distance satisfies a threshold.   
     
     
         17 . The computer-readable medium of  claim 16 , wherein the instructions cause the processor to select the subset based on locations of the subset of scale-dependent features approximately mapping to locations of the predetermined set of features under an affine transform. 
     
     
         18 . The computer-readable medium of  claim 17 , wherein the instructions cause the processor to further determine a location corresponding to an additional signature in the purported target, wherein the instructions cause the processor to determine the location based on the affine transform. 
     
     
         19 . The computer-readable medium of  claim 16 , wherein the instructions cause the processor to compute the smallest mathematically determined distance by calculating a Euclidean distances between the descriptor of each feature in the subset and the descriptor of the corresponding feature in the predetermined set. 
     
     
         20 . The computer-readable medium of  claim 16 , wherein the instructions cause the processor to display the thumbnail image of the portion of the target as a ghost image that aligns with the corresponding portion of the purported target to a user,
 and wherein the instructions cause the processor to further receive the image of the corresponding portion of the purported target as has been user-selected.