Image transformation for indicia reading

• US 10,249,030 B2
• Filed: 10/30/2015
• Issued: 04/02/2019
• Est. Priority Date: 10/30/2015
• Status: Active Grant

• Current Assignee: Hand Held Products Incorporated (Honeywell International Inc.)
• Original Assignee: Hand Held Products Incorporated (Honeywell International Inc.)
• Inventors: McCloskey, Scott, Lloyd, Ryan Andrew
• Primary Examiner(s): Couso, Yon J
• Application Number:
  US14/928,032
• Publication Number:
  US 20170124687A1
• Time in Patent Office: 1,250 Days
• CPC Class Codes: G06K 7/10722 Photodetector array or CCD scanning
• CPC Class Codes: G06K 7/1447 extracting optical codes from image or text carrying said optical code
• CPC Class Codes: G06K 7/1456 determining the orientation of the optical code with respect to the reader and correcting therefore
• CPC Class Codes: G06T 2207/10028 Range image; Depth image; 3D point clouds
• CPC Class Codes: G06T 5/80 Geometric correction
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/50 Depth or shape recovery
• CPC Class Codes: G06T 7/62 of area, perimeter, diameter or volume
• CPC Class Codes: G06V 30/224 of printed characters having additional code marks or containing code marks

Abstract

Obtaining a good image is important for reading indicia. Often, however, objects are not well aligned with an indicia reader'"'"'s digital imager. This misalignment may lead to images in which an indicium may appear distorted due to perspective. This perspective distortion makes indicia reading difficult. In general, perspective distortion of an imaged object may be corrected using knowledge of the true size, shape, and/or orientation of the physical object. The present invention embraces the use of a dimensioning system to obtain the 3D information necessary to provide this knowledge. Then, using this knowledge, to transform a distorted image of an indicium into a corrected image of the indicium without distortion. This corrected image may then be processed to read the undistorted indicium.

Claims

1. A method for reading an indicium, the method comprising:

capturing an image of an object using a digital imager;

computing in real-time, using a dimensioning system, three-dimensional (3D) information corresponding to the object;

selecting an image-region-of-interest (image-ROI) in the image, the image-ROI containing at least a portion of the object;

creating an ideal-region-of-interest (ideal-ROI) using the 3D information corresponding to the portion of the object within the image-ROI, wherein dimensions of the ideal-ROI correspond to physical dimensions of the portion of the object contained in the image-ROI, and wherein the physical dimensions are determined using the 3D information;

transforming the portion of the image within the image-ROI into a corrected image within the ideal-ROI;

searching for an indicium in the corrected image; and

if an indicium can be found, then reading the indicium in the corrected image.

2. The method according to claim 1, further comprising:

if an indicium cannot be found in the corrected image, then repeating the steps of selecting, creating, transforming, and searching until an indicium is found or until all portions of the object that are visible to the digital imager have been searched.

3. The method according to claim 1, further comprising after the step of reading the indicium:

if additional indicia are found in the corrected image, then reading the additional indicia.

4. The method according to claim 1, wherein the selecting an image-ROI in the image comprises:

detecting an indicium in the image; and

selecting a region in the image that includes just the indicium.

5. The method according to claim 1, wherein the selecting an image-ROI in the image comprises:

detecting a surface on the object using the 3D information; and

selecting a region of the image that includes just the surface.

6. The method according to claim 1, wherein the 3D information corresponding to the portion of the object within the image-ROI comprises the physical curvature of the portion of the object contained in the image-ROI.

7. The method according to claim 1, wherein the 3D information corresponding to the portion of the object within the image-ROI comprises the surface-roughness and/or surface-features of the portion of the object contained in the image-ROI.

8. The method according to claim 1, wherein the 3D information corresponding to the portion of the object within the image-ROI comprises the physical position of the portion of the object contained in the image-ROI, the physical position being relative to the digital imager'"'"'s image sensor.

9. The method according to claim 1, wherein transforming the portion of the image within the image-ROI into a corrected image within the ideal-ROI comprises computing a homography between the image-ROI and the ideal-ROI.

10. A method for removing perspective distortion from an image using three-dimensional (3D) information, the method comprising:

capturing, using a digital imager, an image of a scene;

computing in real time, using a dimensioning system, 3D information corresponding to the scene;

selecting an image-region-of-interest (image-ROI) in the image;

creating, using the 3D information corresponding to the portion of the scene within the image-ROI, an ideal-region-of-interest (ideal-ROI), wherein dimensions of the ideal-ROI correspond to physical dimensions of the portion of the scene within the image-ROI, and wherein the physical dimensions are determined using the 3D information; and

transforming, the image in the image-ROI into a corrected image in the ideal-ROI to remove the perspective distortion.

11. The method according to claim 10, wherein the computing in real time 3D information comprises:

identifying objects in the image of the scene;

choosing a particular object;

classifying the particular object by its shape;

obtaining, based on the particular object'"'"'s shape, 3D information corresponding to the particular object.

12. The method according to claim 11, wherein creating an ideal-ROI comprises using the particular-object'"'"'s shape classification.

13. The method according to claim 11, wherein transforming the image in the image-ROI into a corrected image in the ideal-ROI to remove perspective distortion, comprises:

selecting a transformation from a plurality of transformations, the selection based on the particular-object'"'"'s shape classification.

14. An indicia-reading system comprising:

a dimensioning system to compute in real time 3D information about a scene, the dimensioning system comprising a pattern projector and a range camera;

a digital imager to capture an image of the scene; and

a processor communicatively coupled to the dimensioning system and the digital imager, the processor configured by software to;

select an image-region-of-interest (image-ROI) within the image of the scene, the image-ROI containing a distorted indicium,

transform the portion of the image within the image-ROI into a corrected image using 3D information corresponding to the portion of the scene within the image-ROI, the corrected image containing an undistorted indicium, and

read the undistorted indicium in the corrected image.

15. The indicia-reading system according to claim 14, wherein the dimensioning system comprises:

the pattern projector configured to project infrared (IR) light in the form of a transmitted pattern, and

the range camera configured to capture images of IR light reflected from the scene in the form of a reflected pattern.

16. The indicia-reading system according to claim 15, wherein the pattern projector and the range camera are positioned stereoscopically so that spatial differences between the transmitted pattern and the reflected pattern determine range.

17. The indicia-reading system according to claim 14, wherein the digital imager is an imaging indicia reader.

18. The indicia-reading system according to claim 14, wherein the dimensioning system, the digital imager, and the processor are contained in a handheld housing.

19. The indicia-reading system according to claim 14, wherein the digital imager is a camera for imaging visible light.