Method and device for processing multi-channel feature map images
US10733767B2 · kind B2 · utility
8Cited by
3References
20Claims
0Family size
Assignee
Inventors
Key dates
| Filing date | Apr 30, 2018 |
| Grant date | Aug 4, 2020 |
| Priority date | — |
| Expiry date | Jan 24, 2039 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06V10/82
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A convolutional neural network-based image processing method is provided. The method includes: receiving, in a second layer, multi-channel feature map images generated by applying a convolution operation to an input image of a convolutional neural network having a plurality of layers with a plurality of filter kernels of a first layer; analyzing a dynamic range of the multi-channel feature map images; re-ordering the multi-channel feature map images, based on the dynamic range; and processing the re-ordered multi-channel feature map images in the second layer.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.