Image based target analysis

Abstract

In some examples, image based target analysis may include performing a search to identify an image, and determining whether a target is present in the image, and image metadata. Image based target analysis may further include performing, in response to a determination that the target is present in the image, based on a relationship of the target to an entity, and the image metadata, a further search to identify another image, and generating, based on the image and/or the other image, a result indicative of further information on the target.

What is claimed is: 
     
         1 . An apparatus comprising:
 a processor; and   a memory storing machine readable instructions that when executed by the processor cause the processor to:
 perform, based on an input image of a target and an input image metadata associated with the input image, a search of a database to identify a further image; 
 determine whether the target is present in the further image; 
 determine further image metadata associated with the further image; 
 perform, in response to a determination that the target is present in the further image, based on a relationship of the target to an entity, and the further image metadata, a further search of the database or another database to identify another image; and 
 generate, based on at least one of the further image and the other image, a result indicative of further information on the target. 
   
     
     
         2 . The apparatus according to  claim 1 , wherein the target includes a person, the machine readable instructions to determine whether the target is present in the further image and determine the further image metadata associated with the further image further comprise machine readable instructions to cause the processor to:
 determine whether another person is present in the further image and the further image metadata associated with the further image; and   perform, in response to a determination that the other person is present in the further image, based on the relationship of the target to the entity, and the further image metadata, a yet further search of the database or another database to identify yet another image.   
     
     
         3 . The apparatus according to  claim 1 , further comprising machine readable instructions to cause the processor to:
 ascertain the relationship of the target to the entity that includes at least one of a kinship relationship, a membership relationship, a friend relationship, a locale relationship, an ethnicity relationship, a date relationship, a weather relationship, and a work relationship of the target to the entity to constrain the search associated with the input image.   
     
     
         4 . The apparatus according to  claim 1 , wherein the relationship of the target to the entity includes a plurality of relationships combined by at least one Boolean logical operator that includes AND, OR, XOR, and NOT, wherein AND represents a conjunction, OR represents a disjunction, XOR represents an exclusive OR, and NOT represents a negation. 
     
     
         5 . The apparatus according to  claim 1 , wherein the machine readable instructions, when executed by the processor, further cause the processor to:
 ascertain a search threshold;   limit, based on the search threshold, the search of the database to identify the further image; and   limit, based on the search threshold, the search of the database or the other database to identify the other image.   
     
     
         6 . The apparatus according to  claim 1 , wherein the machine readable instructions to determine whether the target is present in the further image determine further image metadata associated with the further image, and generate, based on at least one of the further image and the other image, the result indicative of further information on the target further comprise machine readable instructions to cause the processor to:
 perform, in response to a determination that the target is not present in the further image, based on the relationship of the target to the entity, and the further image metadata, a yet further search of the database or the another database to identify yet another image;   determine, based on an analysis of the yet other image, whether the target is present in the yet other image, and yet other image metadata associated with the yet other image; and   generate, based on at least one of the further image and the yet other image, the result indicative of further information on the target.   
     
     
         7 . The apparatus according to  claim 1 , wherein the machine readable instructions to determine whether the target is present in the further image determine further image metadata associated with the further image further comprise machine readable instructions to cause the processor to:
 ascertain a confidence level for matching the target in the input image to an object in the further image;   determine, based on the analysis of the further image, whether a confidence of a match of the target in the input image to the object in the further image is at or greater than the confidence level; and   determine, in response to a determination that the confidence of the match of the target in the input image to the object in the further image is at or greater than the confidence level, that the target is present in the further image.   
     
     
         8 . The apparatus according to  claim 1 , further comprising machine readable instructions to cause the processor to:
 perform an age-regression on the input image to a specified age of the target;   ascertain the relationship of the target to the entity to constrain the search associated with the age-regressed input image;   perform, based on the age-regressed input image and the input image metadata, the search of the database to identify the further image;   perform the age-regression on the further image to the specified age of the target;   determine, based on the analysis of the age-regressed further image, whether the target is present in the age-regressed further image, and further image metadata associated with the age-regressed further image;   perform, in response to the determination that the target is present in the age-regressed further image, based on the relationship of the target to the entity, and the further image metadata, a further search of the database or another database to identify another image; and   generate, based on at least one of the age-regressed further image and the other image, a result indicative of further information on the target.   
     
     
         9 . A method comprising:
 performing an age-regression or age-progression on an input image of a target to a specified age of the target;   performing, by a processor, based on the age-regressed or age-progressed input image, a search of a database to identify a further image;   performing the age-regression or age-progression on the further image to the specified age of the target;   determining whether the target is present in the age-regressed or age-progressed further image;   determining further image metadata associated with the age-regressed or age-progressed further image;   performing, in response to a determination that the target is present in the age-regressed or age-progressed further image, based on a relationship of the target to an entity, and the further image metadata, a further search of the database or another database to identify another image; and   generating, based on at least one of the age-regressed or age-progressed further image and the age-regressed or age-progressed other image, a result indicative of further information on the target.   
     
     
         10 . The method according to  claim 9 , further comprising:
 ascertaining the relationship of the target to the entity that includes at least one of a kinship relationship, a membership relationship, a friend relationship, a locale relationship, an ethnicity relationship, a date relationship, a weather relationship, and a work relationship of the target to the entity to constrain the search associated with the input image.   
     
     
         11 . The method according to  claim 9 , wherein the relationship of the target to the entity includes a plurality of relationships combined by at least one Boolean logical operator that includes AND, OR, XOR, and NOT, wherein AND represents a conjunction, OR represents a disjunction, XOR represents an exclusive OR, and NOT represents a negation. 
     
     
         12 . A non-transitory computer readable medium having stored thereon machine readable instructions to provide image based target analysis, the machine readable instructions, when executed, cause a processor to:
 perform, based on a keyword associated with a target that is to be analyzed, a search of a database to identify an initial image, wherein the search of the database is limited based on a search threshold;   determine, based on a query associated with the initial image, whether the target is present in the initial image, and initial image metadata associated with the initial image;   perform, in response to a determination that the target is present in the initial image, based on a relationship of the target to an entity, and the initial image metadata, a further search of the database or another database to identify another image, wherein the search of the other database is limited based on the search threshold; and   generate, based on at least one of the initial image and the other image, a result indicative of further information on the target.   
     
     
         13 . The non-transitory computer readable medium according to  claim 12 , wherein the machine readable instructions are further to cause the processor to:
 perform, in response to a determination that the target is not present in the initial image, based on the relationship of the target to the entity, and the initial image metadata, a yet further search of the database or the another database to identify yet another image;   determine, based on a query associated with the yet other image, whether the target is present in the yet other image, and yet other image metadata associated with the yet other image; and   generate, based on at least one of the initial image and the yet other image, the result indicative of further information on the target.   
     
     
         14 . The non-transitory computer readable medium according to  claim 12 , wherein the machine readable instructions are further to cause the processor to:
 ascertain an input image of the target;   ascertain a confidence level for matching the target in the input image to an object in the initial image;   determine, based on an analysis of the initial image, whether a confidence of a match of the target in the input image to the object in the initial image is at or greater than the confidence level; and   determine, in response to a determination that the confidence of the match of the target in the input image to the object in the initial image is at or greater than the confidence level, that the target is present in the initial image.   
     
     
         15 . The non-transitory computer readable medium according to  claim 12 , wherein the machine readable instructions are further to cause the processor to:
 perform an age-regression on the initial image to a specified age of the target;   determine, based on the query associated with the age-regressed initial image, whether the target is present in the age-regressed initial image, and initial image metadata associated with the age-regressed initial image;   perform, in response to the determination that the target is present in the age-regressed initial image, based on the relationship of the target to the entity, and the initial image metadata, a further search of the database or another database to identify another image; and   generate, based on at least one of the age-regressed initial image and the other image, a result indicative of further information on the target.