Methods And Systems For Determining Optimal Features For Classifying Patterns Or Objects In Images

Abstract

Provided are methods for determining optimal features for classifying patterns or objects. Also provided are methods for image analysis. Further provided are methods for image searching.

What is claimed is: 
     
         1 . A method for image analysis comprising:
 receiving at least one image; and   determining features from the at least one image by classifying the at least one image using a trained classifier wherein the trained classifier utilizes one or more independent components.   
     
     
         2 . The method of  claim 1 , wherein the at least one image is of a non-natural scene. 
     
     
         3 . The method of  claim 2 , wherein the at least one image is one of, a color retinal image, a monochrome retinal image, a color stereo pair, a monochrome stereo pair, an x-ray image, a computer-aided tomographic (CAT) scan image, an angiogram image, a fMRI image, or a PET image. 
     
     
         4 . The method of  claim 1 , further comprising training the classifier, rein training the classifier comprises:
 presenting, to the classifier, a pre-classified image;   separating the pre-classified image into a plurality of channels;   determining a set of samples from the plurality of channels; and   for each channel, performing an Independent Component Analysis (ICA) on the set of samples resulting in a set of independent components.   
     
     
         5 . The method of  claim 1 , wherein the trained classifier is one of, a k-Nearest Neighbors classifier, a linear discriminant classifier, a quadratic discriminant classifier, or a support vector machine. 
     
     
         6 . The method of  claim 4 , wherein the ICA uses two dimensional samples from two-dimensional channels 
     
     
         7 . The method of  claim 4 , wherein the ICA uses three dimensional samples from three-dimensional channels. 
     
     
         8 . The method of  claim 4 , wherein the plurality of channels comprises one or more of, red, green, blue, a vectorial representation of multichannel data, color opponency channels as in a mammalian visual cortex, hue, saturation, and brightness. 
     
     
         9 . The method of  claim 1 , further comprising outputting a visual representation of the determined features. 
     
     
         10 . A system for image analysis comprising:
 a memory; and   a processor configured for performing steps comprising,
 receiving at least one image, and 
 determining features from the at least one image by classifying the at least one image using a trained classifier wherein the trained classifier utilizes one or more independent components. 
   
     
     
         11 . The system of  claim 10 , wherein the at least one image is of a non-natural scene. 
     
     
         12 . The system of  claim 11 , wherein the at least one image is one of, a color retinal image, a monochrome retinal image, a color stereo pair, a monochrome stereo pair, an x-ray image, a computer-aided tomographic (CAT) scan image, an angiogram image, a fMRI image, or a PET image. 
     
     
         13 . The system of  claim 10 , further comprising training the classifier, wherein training the classifier comprises:
 presenting, to the classifier, a pre-classified image;   separating the pre-classified image into a plurality of channels;   determining a set of samples from the plurality of channels; and   for each channel, performing an Independent Component Analysis (ICA) on the set of samples resulting in a set of independent components.   
     
     
         14 . The system of  claim 10 , wherein the trained classifier is one of, a k-Nearest Neighbors classifier, a linear discriminant classifier, a quadratic discriminant classifier, or a support vector machine. 
     
     
         15 . The system of  claim 13 , wherein the ICA uses two dimensional samples from two-dimensional channels 
     
     
         16 . The system of  claim 13 , wherein the ICA uses three dimensional samples from three-dimensional channels. 
     
     
         17 . The system of  claim 13 , wherein the plurality of channels comprises one or more of, red, green, blue, a vectorial representation of multichannel data, color opponency channels as in a mammalian visual cortex, hue, saturation, and brightness. 
     
     
         18 . The system of  claim 10 , further comprising outputting a visual representation of the determined features. 
     
     
         19 . A non-transitory computer readable medium with computer executable instructions embodied thereon for image analysis comprising:
 receiving at least one image; and   determining features from the at least one image by classifying the at least one image using a trained classifier wherein the trained classifier utilizes one or more independent components.   
     
     
         20 . The non-transitory computer readable medium of  claim 19 , wherein the at least one image is of a non-natural scene. 
     
     
         21 . The non-transitory computer readable medium of  claim 20 , wherein the at least one image is one of, a color retinal image, a monochrome retinal image, a color stereo pair, a monochrome stereo pair, an x-ray image, a computer-aided tomographic (CAT) scan image, an angiogram image, a fMRI image, or a PET image. 
     
     
         22 . The non-transitory computer readable medium of  claim 19 , further comprising training the classifier, wherein training the classifier comprises:
 presenting, to the classifier, a pre-classified image;   separating the pre-classified image into a plurality of channels;   determining a set of samples from the plurality of channels; and   for each channel, performing an Independent Component Analysis (ICA) on the set of samples resulting in a set of independent components.   
     
     
         23 . The non-transitory computer readable medium of  claim 19 , wherein the trained classifier is one of, a k-Nearest Neighbors classifier, a linear discriminant classifier, a quadratic discriminant classifier, or a support vector machine. 
     
     
         24 . The non-transitory computer readable medium of  claim 22 , wherein the ICA uses two dimensional samples from two-dimensional channels 
     
     
         25 . The non-transitory computer readable medium of  claim 22 , wherein the ICA uses three dimensional samples from three-dimensional channels. 
     
     
         26 . The non-transitory computer readable medium of  claim 22 , wherein the plurality of channels comprises one or more of, red, green, blue, a vectorial representation of multichannel data, color opponency channels as in a mammalian visual cortex, hue, saturation, and brightness. 
     
     
         27 . The non-transitory computer readable medium of  claim 19 , further comprising outputting a visual representation of the determined features.