METHOD AND APPARATUS FOR GENERATING SPECIAL-PURPOSE IMAGE ANALYSIS ALGORITHMS
Provides quantitative data about a two or more dimensional image. Classifies and counts number of entities an image contains. Each entity comprises a structure, or some other type of identifiable portion having definable characteristics. The entities located within an image may have different shape, color, texture, etc., but still belong to the same classification. Alternatively, entities comprising a similar color/texture may be classified as one type while entities comprising a different color/texture may be classified as another type. May quantify image data according to set of changing criteria and derive one or more classifications for entities in image. I.e., provides a way for a computer to determine what kind of entities (e.g., entities) are in image and counts total number of entities visually identified in image. Information utilized during a training process may be stored and applied across different images.
1 . In a computer system, a method for automating the expert quantification of image data using a product algorithm comprising:
obtaining a product algorithm for analysis of a first set of image data wherein said product algorithm is configured to recognize at least one entity within said first set of image data via a training mode that utilizes iterative input to an evolving algorithm obtained from at least one first user, wherein said training mode comprises:
presenting a first set of said at least one entity to said user for feedback as to the accuracy of said first set of identified entities;
obtaining said feedback from said user;
executing said evolving algorithm using said feedback;
presenting a second set of said at least one entity to said user for feedback as to the accuracy of said second set of identified entities;
obtaining approval from said user about said second set of entities; storing said evolving algorithm as a product algorithm;
providing said product algorithm to at least one second user so that said at least one second user can apply said product algorithm against a second set of image data having said at least one entity.
2 . The method of claim 1 wherein said evolving algorithm comprises a neural network.
3 . The method of claim 1 wherein said evolving algorithm comprises a classification engine.
4 . The method of claim 1 wherein said product algorithm comprises a pixel zoo.
6 . The method of claim 1 wherein said product algorithm comprises an entity zoo.
7 . A computer program product comprising:
a memory medium embodying computer readable program code for automating the expert quantification of image data, said computer readable program code configured to:
obtain image data having a plurality of chromatic data points;
identify which of said plurality of chromatic data points comprise an entity;
group said plurality of chromatic data points into a plurality of spatially connected subsets;
determine a plurality of characteristics about said spatially connected subsets;
pass said plurality of characteristics to a classification engine;
classify said plurality of spatially connected subsets into at least one classification;
obtaining affirmation of the veracity of said at least one classification from a user;
evaluate said spatially connected subset to derive a set of relative harmonic amplitudes;
pass said relative harmonics into a neural network, wherein said neural network is trained to classify said spatially connected subsets using shape information provided by said set of relative harmonic amplitudes;
present a result of said classification to said user;
obtain verification of said classification from said user;
using said verification to adjust said neural network.