Methods and arrangements for localizing machine-readable indicia

Patent metadata

Field	Value
Publication number	US-2025028919-A1
Application number	US-202418675848-A
Country	US
Kind code	A1
Filing date	May 28, 2024
Priority date	Feb 8, 2018
Publication date	Jan 23, 2025
Grant date	—

Abstract

The present technology relates to image signal processing. One aspect of the present technology involves analyzing reference imagery gathered by a camera system to determine which parts of an image frame offer high probabilities of—relative to other image parts—containing decodable signal data. Another aspect of the present technology whittles-down such determined image frame parts based on detected content (e.g., a cereal box) vs expected background within such determined image frame parts.

First claim

1 - 6 . (canceled)

7 . A method of processing image data, comprising:
obtaining image data captured by one or more cameras; analyzing a subset of the image data to determine whether it represents a content object or background imagery, said analyzing yielding a determination; and based on the determination, operating a signal localizer to assess whether the subset likely includes a two-dimensional encoded signal carried by a plurality of print elements.

8 . The method of claim 7 wherein the one or more cameras comprise retail scanner cameras.

9 . The method of claim 7 wherein the content object comprises a retail item.

10 . The method of claim 7 further comprising:
dividing the subset of the image data into a plurality of subareas, each subarea comprising n×m pixels, where n and m are both positive integers;
for each subarea:
determining an image characteristic representing the n×m pixels;
comparing the determined image characteristic to a baseline characteristic associated with the subarea; and
classifying the subarea as background or content based on said comparing.

11 . The method of claim 10 , wherein the image characteristic comprises a pixel mean value representing the n×m pixels.

12 . The method of claim 10 , further comprising:
maintaining an array or table of baseline values associated with the baseline characteristic; and maintaining a histogram of pixel values associated with each subarea.

13 . The method of claim 7 , wherein operating the signal localizer comprises:
downsampling the subset of the image data; determining an image mean value for the downsampled subset; for each pixel in the subset, subtracting the image mean from the pixel value to yield a residue value; comparing the residue value to a representation of image noise to yield a collection of image points; filtering the collection of image points to yield a reduced collection of image points; and counting the number of points within the reduced collection of points.

14 . The method of claim 13 further comprising determining whether the subset should be processed by a signal decoder based on the number of points within the reduced collection of points.

15 . The method of claim 13 , wherein the image mean value comprises a pixel greyscale mean value.

16 . The method of claim 13 , further comprising, prior to subtracting the image mean:
determining whether the image mean value is above a threshold; and stopping processing of the subset when the image mean value is below the threshold.

17 . An image processing system comprising:
one or more cameras positioned to capture imagery depicting an object moved past the one or more cameras; one or more light sources positioned to illuminate the object as it is moved past the one or more cameras; one or more processors configured to: determine whether a block of said imagery represents background or the object; and determine whether the block of said imagery likely depicts a two-dimensional dot pattern conveying encoded data, wherein determining whether the block likely depicts a two-dimensional dot pattern is performed only when the block is determined to represent the object.

18 . The system of claim 17 , wherein the one or more processors are further configured to generate a multi-color heatmap representing the two-dimensional dot pattern conveying encoded data.

19 . The system of claim 17 , wherein the object comprises a retail item and the one or more cameras comprise retail scanner cameras.

20 . The system of claim 17 , wherein to determine whether the block of said imagery represents background or the object, the one or more processors are configured to:
divide the block into a plurality of subareas, each subarea comprising n×m pixels, where n and m are both positive integers; for each subarea: determine an image characteristic representing the n×m pixels; compare the determined image characteristic to a baseline characteristic associated with the subarea; and classify the subarea as background or content based on said comparing.

21 . The system of claim 20 , wherein the image characteristic comprises a pixel mean value representing the n×m pixels.

22 . The system of claim 17 , wherein to determine whether the block of said imagery likely depicts a two-dimensional dot pattern, the one or more processors are configured to:
downsample the block of imagery; determine an image mean value for the downsampled block; for each pixel in the block, subtract the image mean from the pixel value to yield a residue value; compare the residue value to a representation of image noise to yield a collection of image points; filter the collection of image points to yield a reduced collection of image points; and count the number of points within the reduced collection of points.

23 . The system of claim 22 , wherein the one or more processors are further configured to determine whether the block should be processed by a signal decoder based on the number of points within the reduced collection of points.

24 . A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
obtaining image data captured by one or more cameras; analyzing a subset of the image data to determine whether it represents a content object or background imagery, said analyzing yielding a determination; and based on the determination, operating a signal localizer to assess whether the subset likely includes a two-dimensional encoded signal carried by a plurality of dots.

25 . The non-transitory computer-readable medium of claim 24 , wherein the operations further comprise:
dividing the subset of the image data into a plurality of subareas, each subarea comprising n×m pixels, where n and m are both positive integers; for each subarea: determining an image characteristic representing the n×m pixels; comparing the determined image characteristic to a baseline characteristic associated with the subarea; and classifying the subarea as background or content based on said comparing.

26 . The non-transitory computer-readable medium of claim 24 , wherein operating the signal localizer comprises:
downsampling the subset of the image data; determining an image mean value for the downsampled subset; for each pixel in the subset, subtracting the image mean from the pixel value to yield a residue value; comparing the residue value to a representation of image noise to yield a collection of image points; filtering the collection of image points to yield a reduced collection of image points; and counting the number of points within the reduced collection of points.

Assignees

• Digimarc Corp

Inventors

• Holub Vojtech
• Deshmukh Utkarsh
• Filler Tomas

Classifications

• G06V10/255

  Detecting or recognising potential candidate objects based on visual cues, e.g. shapes · CPC title

• G06V10/30

  Noise filtering · CPC title

• G06V10/32

  Normalisation of the pattern dimensions · CPC title

• H04N23/56

  provided with illuminating means · CPC title

• G06K7/1413

  1D bar codes · CPC title