Methods for generating an undistorted image from a distorted original image of a graphical representation

Inventors

ROVDER, Simon • Herman, Martin • Herman, Robert • JURÁŠEK, Viktor • PALUŠ, Timotej

Assignees

Powerful Medical SRO

Interested in licensing this patent?

MTEC can help explore whether this patent might be available for licensing for your application.

Publication Number

US-12614244-B2

Patent

Publication Date

2026-04-28

Expiration Date


Abstract

Embodiments relate to a method for generating an image comprising a graphical representation. The method includes determining a transformation relationship for a unit cell of a graphical representation of an original image. The transformation relationship is based on position information of grid vertices of the unit cell of the original image graphical representation and position information of grid vertices of a unit cell of a reference grid arrangement of the image to be generated. The method further includes determining position information of a one or more pixels of the unit cell of the original image graphical representation based on the transformation relationship. The method further includes applying data information sampled at the one or more pixels of the unit cell of the original image graphical representation to one or more pixels of the unit cell of the reference grid arrangement.

Core Innovation

The invention relates to methods for producing an undistorted image from a distorted original image having a graphical representation on an original grid arrangement. The graphical representation is formed on the original grid arrangement with a plurality of unit cells that forms a background onto which graphical information is recorded, and the method produces an undistorted image based on a reference grid arrangement comprising a plurality of undistorted unit cells creatable on a blank image.

The method includes creating the reference grid arrangement by associating the plurality of undistorted unit cells on the blank page with corresponding unit cells of the original grid arrangement derivable from the distorted original image. For each current unit cell of the reference grid arrangement, a transformation relationship is determined based on position information of grid vertices of a target unit cell and position information of grid vertices of a corresponding source unit cell in the distorted original image.

Using the determined transformation relationship, the method determines position information in the source unit cell corresponding to one or more pixels of the current unit cell and supplies data information sampled in the source unit cell to the corresponding one or more pixels of the target unit cell to produce the undistorted image. The sampled data includes color value information representing waveform data of the original image graphical representation, and the disclosure further describes detecting distorted grid vertices, constructing unit cells from detected vertices and selected edges, tracing closed unit-cell paths, selecting the largest network of unit cells, and inferring missing vertices using linear regression.

Claims Coverage

Independent claim clm-00001 defines a transformation-based mapping approach for generating an undistorted image from a distorted original image by creating an undistorted reference grid, determining per-unit-cell transformations from grid-vertex position information, sampling source data, and writing sampled data into corresponding pixels of the target unit cells. The provided material also indicates dependent claim refinements that add constraints and specific mechanisms for grid processing and topology construction.

Undistorted image producing from distorted grid-based graphical representation

A method for producing an undistorted image from a distorted original image comprising a graphical representation on an original grid arrangement with a plurality of unit cells forming a background onto which graphical information is recorded, based on a reference grid arrangement comprising a plurality of undistorted unit cells creatable on a blank image.

Reference-to-source grid association on blank page

Creating the reference grid arrangement by associating the plurality of undistorted unit cells on the blank page with corresponding unit cells of the original grid arrangement derivable from the distorted original image.

Grid-vertex transformation relationship for unit cells

Determining a transformation relationship based on position information of grid vertices of a current unit cell, referred to as target unit cell, and position information of grid vertices of a corresponding unit cell, referred to as source unit cell, of the original grid arrangement in the distorted original image.

Pixel-position mapping using transformation relationship

Determining position information in a source unit cell corresponding to one or more pixels of the current unit cell, referred to as target unit cell, using the transformation relationship.

Sampling source data and supplying to target pixels

Supplying data information sampled in the source unit cell, using the position information determined in the source unit cell, to the corresponding one or more pixels of the target unit cell to produce the undistorted image.

Same size reference and original grid arrangements

The reference grid arrangement and the original grid arrangement are the same size.

Waveform data as sampled color values

The sampled data information consists of color value information representing waveform data of the original image graphical representation.

Scatter-plot-based unit cell construction from detected grid vertices

Determining position information for a plurality of grid vertices in an original image by generating a two-dimensional scatter plot of dots representing detected grid vertices, determining edge sets for each dot to form a unit cell, and constructing a corresponding unit cell for each dot based on those edge sets.

Maximum-likelihood edge tracing for closed unit cell paths

Constructing a unit cell by tracing paths starting from each reference dot along selected maximum-likelihood edges and then building a network of unit cells that each have a closed path.

Breadth-first search for largest unit-cell network and linear-regression inference

Determining the largest grid arrangement network of unit cells using breadth-first search, identifying missing unit cells, and calculating vertex position information for the missing unit cells using linear regression.

Overall, the claim set coverage centers on creating an undistorted reference grid on a blank page, determining unit-cell transformation relationships from grid-vertex positions between target and source unit cells, mapping pixel positions via the transformation, and supplying sampled source data to corresponding target pixels to form the undistorted image. Dependent refinements constrain grid size and define the sampled waveform data as color values, while further specifying grid-vertex detection and unit-cell topology construction using probabilistic or maximum-likelihood edge selection, closed unit-cell path tracing, breadth-first search selection of the largest network, and linear regression for missing vertex inference.

Stated Advantages

Not explicitly described in patent.

Documented Applications

Not explicitly described in patent.

JOIN OUR MAILING LIST

Stay Connected with MTEC

Keep up with active and upcoming solicitations, MTEC news and other valuable information.