System and method for dynamic thresholding for multiple result image cross correlation
AssigneeRevenue Management Solutions, LLC
InventorJay Wagner
Life of this patent
Estimating the term…
Abstract
The present disclosure relates to a computer-implemented system and method for finding matching occurrences of an item of interest (or image or sub-image) within a document (or larger image) via cross correlation and setting a dynamic threshold for each document (or larger image). The described system and method are capable of matching and locating the one or more items of interest within each specific document (or larger image).
Claims19
A method of locating one or more preselected items of interest in an image, the method being performed by one or more processors and comprising the steps of:
- (a) searching the image for the preselected item of interest;
- wherein the searching step includes scanning one or more locations of the image for the preselected item of interest and assigning a value for each scanned location, wherein the assigned value is representative of a similarity between each scanned location of the image and the preselected item of interest;
- (b) after step (a), locating one or more sets of one or more outliers of said image by performing a median absolute deviation analysis on each assigned value for each scanned location of said image;
- (c) after step (b), selecting a dynamic threshold value based on a maximum value derived from said median absolute deviation analysis performed in step (b) of the located one or more outliers;
- and
- (d) after step (c), for locations greater than the selected dynamic threshold value from step (c), identifying a location in the image for each of said one or more preselected items of interest.
A computer-implemented method performed by one or more processors comprising the steps of:
- (a) searching an image for a preselected sub-image;
- (b) after step (a), locating one or more sets of outliers of said image by performing a median absolute deviation analysis on an assigned value for each location of a plurality of locations of said image;
- (c) selecting a threshold value based on a maximum value derived from said median absolute deviation analysis;
- and
- (d) for locations greater than the selected threshold value defined in step (c), identifying one or more locations of said preselected sub-image within said image.
A non-transitory computer-readable storage media having stored thereon a plurality of computer-executable instructions for locating one or more preselected items of interest within an image and selecting a dynamic threshold value for each image which, when executed by a processor, cause the processor to:
- search the image for one or more occurrences of the preselected item of interest, wherein searching the image for one or more occurrences of the preselected item of interest includes scanning one or more locations of the image for the preselected item of interest and assigning a value for each scanned location, wherein the assigned value is representative of a similarity between each scanned location of the image and the preselected item of interest;
- locate one or more sets of one or more outliers of said image by performing a median absolute deviation analysis on each assigned value for each scanned location of said image;
- select a dynamic threshold value based on a maximum value derived from said median absolute deviation analysis to locate said one or more outliers;
- and
- for locations greater than the selected dynamic threshold value, identify a location in said image for each of said one or more preselected items of interest.
A system comprising:
- a processor;
- and
- a computer-readable storage media operably connected to said processor, said computer-readable storage media including instructions that when executed by said processor, cause performance said processor to perform a method of locating one or more preselected items of interest in an image and selecting a dynamic threshold value for each image, the method comprising:
- searching the image for the preselected item of interest, wherein the searching step includes scanning one or more locations of the image for the preselected item of interest and assigning a value for each scanned location, wherein the assigned value is representative of a similarity between each scanned location of the image and the preselected item of interest;
- locating one or more sets of one or more outliers of said image by performing a median absolute deviation analysis on each assigned value for each scanned location of said image;
- selecting a dynamic threshold value based on a maximum value derived from said median absolute deviation analysis to locate said one or more outliers;
- and
- for locations greater than the selected dynamic threshold value, identifying a location in the image for each of said one or more preselected items of interest.
Dependencies are inferred from the claim text.