Robust blur estimation for shapes with known structural elements
US12148198B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Apr 13, 2022 |
| Grant date | Nov 19, 2024 |
| Priority date | — |
| Expiry date | Jun 24, 2043 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T2207/30168
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Various implementations disclosed herein assess the blurriness of portions of images depicting shapes such as codes or text that have known structural elements. This may involve determining whether a portion of an image of a code or text is sufficiently clear (not blurry) to be accurately interpreted. Blur may be assessed based on spatial frequency of statistical analysis. Blur may be assessed using a machine learning model that is trained using target blur metrics determined based on spatial frequency (e.g., analysis of high frequency portions of discrete cosine transforms of image portions) or statistical analysis (e.g., based on corner/edge detection in image portions).
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.