Image recognition device and method of recognizing image thereof
AssigneeElectronics and Telecommunications Research Institute
InventorsSanghun Yoon, Chun-Gi Lyuh, Ik Jae Chun, Jung Hee Suk, Tae Moon Roh
Life of this patent
Estimating the term…
Abstract
An image recognition device in accordance with the inventive concept may include an input vector extraction part extracting an input vector from an input image; a compression vector conversion part converting the input vector into a compression vector using a projection vector; a training parameter generation part receiving a training vector to generate a training parameter using a projection vector obtained through a folding operation of the training vector; and an image classification part classifying the compression vector using the training vector to output image recognition data.
Claims6
An image recognition device comprising:
- an input vector extraction part extracting an input vector from an input image;
- a compression vector conversion part converting the input vector into a compression vector using a first projection vector;
- a training parameter generation part receiving a training vector to generate a training parameter using a second projection vector obtained through a folding operation of the training vector, wherein the training parameter is a vector and the folding operation produces the second projection vector having an order less than an order of the training vector;
- and
- an image classification part classifying the compression vector using the training vector to output image recognition data.
An image recognition device comprising:
- an input vector extraction part extracting an input vector from an input image;
- a compression vector conversion part converting the input vector into a compression vector using a first projection vector;
- a training parameter generation part receiving a training vector to generate a training parameter using a second projection vector obtained through a folding operation of the training vector;
- and
- an image classification part classifying the compression vector using the training vector to output image recognition data,
- wherein the training parameter generation part comprises:
- a vector conversion part converting the training vector into a two-dimensional vector using a folding operation;
- a vector decomposition part decomposing a main element of the converted two-dimensional vector;
- a projection vector generation part selecting a plurality of elements having a large energy level using the decomposed vector to generate the second projection vector;
- a training compression vector conversion part converting the training vector into a training compression vector using the second projection vector;
- and
- a training parameter extraction part extracting a training parameter from a training vector gathering composed of the training compression vector.
Dependencies are inferred from the claim text.