Spectral reflectance compression
AssigneeHewlett-Packard Developmet Company, L.P.
InventorsPeter Morovic, Jan Morovic, Jordi Arnabat Benedicto
Life of this patent
Estimating the term…
Abstract
In some examples, a method for compressing a spectral reflectance dataset may be performed through compression circuitry. The method may include computing a principal component analysis basis for the spectral reflectance dataset; projecting the spectral reflectance dataset onto the principal component analysis basis to obtain a weight matrix; quantizing the weight matrix; performing a Huffman encoding process on the quantized weight matrix to generate a Huffman table and Huffman codes for the quantized weight matrix; and providing compressed spectral reflectance data as the principal component analysis basis, the Huffman table, and the Huffman codes.
Claims15
A method comprising:
- compressing a spectral reflectance dataset through compression circuitry by:
- computing a principal component analysis (PCA) basis for the spectral reflectance dataset;
- projecting the spectral reflectance dataset onto the PCA basis computed for the spectral reflectance dataset to obtain a weight matrix;
- quantizing the weight matrix to obtain a quantized weight matrix;
- performing a Huffman encoding process on the quantized weight matrix to generate a Huffman table and Huffman codes for the quantized weight matrix;
- and
- providing compressed spectral reflectance data as the PCA basis, the Huffman table, and the Huffman codes.
A system comprising:
- compression circuitry to compress a spectral reflectance dataset that satisfies a preconfigured accuracy level by:
- computing a principal component analysis (PCA) basis for the spectral reflectance dataset;
- projecting the spectral reflectance dataset onto the PCA basis computed for the spectral reflectance dataset to obtain a weight matrix;
- quantizing the weight matrix to obtain a quantized weight matrix;
- performing a Huffman encoding process on the quantized weight matrix to generate a Huffman table and Huffman codes for the quantized weight matrix;
- and
- providing compressed spectral reflectance data as one of the following data combinations with a lesser data size:
- the PCA basis, the Huffman table, and the Huffman codes;
- and
- the PCA basis and the weight matrix.
A product comprising:
- a non-transitory computer-readable medium storing executable instructions to compress a spectral reflectance dataset to meet an accuracy level by:
- computing a principal component analysis (PCA) basis for the spectral reflectance dataset to identify a dimensionality O for the PCA basis that meets the accuracy level;
- projecting the spectral reflectance dataset onto the PCA basis computed for the spectral reflectance dataset to obtain a weight matrix;
- when the PCA basis with dimensionality O meets a dimensionality criterion and the PCA basis with dimensionality O and the weight matrix meet a compression criterion:
- quantizing the weight matrix to obtain a quantized weight matrix;
- performing a Huffman encoding process on the quantized weight matrix to generate a Huffman table and Huffman codes for the quantized weight matrix;
- and
- when the Huffman table and Huffman codes for the quantized weight matrix meet the compression criterion:
- providing, as compressed spectral reflectance data, the Huffman table and Huffman codes for the quantized weight matrix and the PCA basis.
Dependencies are inferred from the claim text.