Parallel Image Convolution Processing with SVD Kernel Data
AssigneeSiemens Medical Solutions USA, Inc.
InventorFaguo Yang
Life of this patent
Estimating the term…
Abstract
A method of image processing includes receiving camera data representative of an object, implementing, with a processor, a convolution operation on the camera data to generate filtered data representative of the object, the convolution operation being configured with singular value decomposition (SVD) kernel data, and storing the filtered data in a memory. Implementing the convolution operation includes applying the SVD kernel data in parallel to a plurality of subsets of the camera data.
Claims21
A method of image processing, the method comprising:
- receiving camera data representative of an object;
- implementing, with a processor, a convolution operation on the camera data to generate filtered data representative of the object, the convolution operation being configured with singular value decomposition (SVD) kernel data;
- and
- storing the filtered data in a memory;
- wherein implementing the convolution operation comprises applying the SVD kernel data in parallel to a plurality of subsets of the camera data.
A non-transitory computer readable storage medium having stored therein data representing instructions executable by a programmed processor for image processing, the instructions comprising computer code to:
- receive camera data representative of an object;
- apply singular value decomposition (SVD) kernel data to the camera data, the application convolving the camera data with a non-separable kernel represented by the SVD kernel data and generating filtered data representative of the object;
- and
- store the filtered data in a memory;
- wherein the instructions to apply the SVD kernel data are configured to apply the SVD kernel data in parallel to multiple rows or multiple columns of the camera data.
A system for image processing, the system comprising:
- a first memory in which camera data representative of an object is stored;
- a processor configured to implement a convolution operation on the camera data to generate filtered data representative of the object, the convolution operation being configured with singular value decomposition (SVD) kernel data;
- and
- a second memory in communication with the processor to store the filtered data;
- wherein the processor is further configured to apply the SVD kernel data in parallel to a plurality of subsets of the camera data.
Dependencies are inferred from the claim text.