Spectral image dimensionality reduction system and method

A method for reducing dimensionality of hyperspectral image data having a number of spatial pixels, each associated with a number of spectral dimensions, the method comprising:nreceiving sets of coefficients associated with each pixel of the hyperspectral image data, a set of basis vectors utilized to generate the sets of coefficients, and a maximum error value;calculating, using a processor, a first set of errors for each pixel associated with the set of basis vectors, and one or more additional sets of errors for each pixel associated with one or more subsets of the set of basis vectors;calculating, using the processor, a percent of the number of spatial pixels having an error greater than the maximum error value, for each of the first set of errors and the one or more additional sets of errors;calculating, using the processor, a plurality of reduction factors associated with each of the first set of errors and the one or more additional sets of errors, the plurality of reduction factors being calculated based on both the percent of the number of spatial pixels having the error greater than the maximum error value and the number of spectral dimensions associated with the hyperspectral image data; andselecting, using the processor, a maximum reduction factor from the plurality of reduction factors, and an optimum size of the set of basis vectors or the subset of basis vectors associated therewith.

A method for reducing dimensionality of hyperspectral image data having a number of spatial pixels, each associated with a number of spectral dimensions, the method comprising:nreceiving sets of coefficients associated with each pixel of the hyperspectral image data, a set of basis vectors utilized to generate the sets of coefficients, and a maximum data size value;calculating, using a processor, a maximum number of members in the set of basis vectors based on the maximum data size value and the number of spatial pixels, and establishing a maximum subset of the set of basis vectors associated with the maximum data size;calculating, using the processor, a first set of errors for each pixel associated with the maximum subset of basis vectors, and one or more additional sets of errors for each pixel associated with one or more subsets of the maximum subset of basis vectors;calculating, using the processor, a percent of the number of spatial pixels that can be set aside, based on the maximum data size, for each of the maximum subset of basis vectors and the one or more additional subsets of basis vectors;calculating, using the processor, a plurality of maximum error values associated with each of the percents of the numbers of spatial pixels that can be set aside, associated with each of the maximum subset of basis vectors and the one or more additional subsets of basis vectors; andselecting, using the processor, a minimum error value from the plurality of maximum error values, and an optimum size of the maximum subset of basis vectors or the one or more additional subsets of the maximum subset of basis vectors associated therewith.