IMAGE BASED MEASUREMENT SYSTEM

• US 20160371855A1
• Filed: 06/19/2015
• Published: 12/22/2016
• Est. Priority Date: 06/19/2015
• Status: Active Grant

• Current Assignee: Smart Picture Technologies Inc.
• Original Assignee: Kari Myllykoski
• Inventors: Jovanovic, Dejan, Beardmore, Keith, Myllykoski, Kari, Freeman, Mark O.
• Granted Patent: US 10,083,522 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 2207/30208 Marker matrix
• CPC Class Codes: G06T 2207/30244 Camera pose
• CPC Class Codes: G06T 5/80 Geometric correction
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06T 7/60 Analysis of geometric attributes
• CPC Class Codes: G06T 7/62 of area, perimeter, diameter or volume
• CPC Class Codes: G06T 7/74 involving reference images or patches
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 10/245 by locating a pattern; Special marks for positioning
• CPC Class Codes: G06V 10/32 Normalisation of the pattern dimensions
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59

Abstract

A Quantified Image Measurement System that creates accurate physical measurement data from digital pictures is disclosed. The system can use any image format and enhances the image file with measurement data and data transformation information that enables the creation of any type of geometrical or dimensional measurement from the stored photograph. This file containing the original digital image along with the supplemental data is referred to as a Quantified Image File or QIF. The QIF can be shared with other systems via email, cloud syncing or other types of sharing technology. Once shared, existing systems such as CAD applications or web/cloud servers can use the QIF and the associated QIF processing software routines to extract physical measurement data and use the data for subsequent processing or building geometrically accurate models of the objects or scene in the image. Additionally smart phones and other portable devices can use the QIF to make measurements on the spot or share between portable devices. In addition, the quantified image measurement system of this invention eliminates the need for capturing the image from any particular viewpoint by using multiple reference points and software algorithms to correct for any off-angle distortions.

Claims

1. A system and apparatus for making accurate dimensional or geometrical measurements of the features in a predominantly 2D scene from an image of the scene taken from even a non-orthogonal perspective comprising

a. a digital scanner that creates digital image data;

b. a reference object containing at least four co-planar reference fiducials physically placed proximate to the predominantly 2D surface containing the features to be measured.

c. a data processing system that;

i. contains information regarding the geometry of the reference template or reference object;

ii. determines the locations of the coplanar reference fiducials within the digital image;

iii. From the locations of the fiducials, determines the non-orthogonal distortion(s) of the objects features caused by the non-orthogonal viewpoint of the camera relative to the predominantly 2D surface;

iv. determine corrections for the distortions;

v. apply the corrections to compute accurate dimensions or geometry of features in the scene

d. A user interface to allow user input and to output the accurate dimensions or geometry of the objects predominantly 2D surface.

e.

2. The measuring device of claim 1 where the dimensions or geometric features which are measured are automatically selected by the computational system.

3. The measuring device of claim 1 further comprising:

a. a user interface allowing the user to manually select features in the scene for which measurements are to be computed, and which displays the results of the measurement computations.

4. The measurement system of claim 1 where the reference object is a printed pattern.

5. The measurement system of claim 4 where the printed pattern is on sticker which can be affixed to the object to be measured.

6. The measurement system of claim 4 where the pattern is printed in color(s) which can more easily be extracted via a digital filter to expedite the data processing determination of the location of the fiducials.

7. The measurement system of claim 1 where the reference object or template includes a pattern of electromagnetic emitting elements.

8. The measurement system of claim 7 where the light emitting elements are of wavelengths which can more easily be extracted via a digital filter to expedite the data processing determination of the location of the fiducials.

9. The measurement system of claim 1 where the digital scanner is a camera.

10. The measurement system of claim 1 where the data processing system also creates a digital image file which includes digital image and the non-orthographic correction data.

11. The measurement system of claim 3 where the data processing system also creates a digital image file which includes digital image and the non-orthographic correction data.

12. The measurement system of claim 11 where the digital image file also contains automatically generated measurement data or measurement data of parameters which were manually identified.

13. The measurement system of claim 1 where fiducial(s) further comprise:

a. a circle with a center;

b. the space within the circle is divided into pie piece wedges intersecting the circle center;

c. some but not all of the pie wedge pieces contain a gradient.

14. The measurement system of claim 13 where gradient increases in density from the outside of the circle toward the center of the circle.

15. The measurement system of claim 14 were the fiducial locations are discovered by the data processing system via:

a. recognizing and locate the reference template as a whole within the camera image based on using the results of a Haar Training procedure;

b. using Hough Circle detection to find the circles of the fiducials;

c. using corner detection is within the circles to locate the corner at the center of the circle using the gradient;

to avoid false recognition of corners;

d. refine location of the center of the fiducial with sub-pixel precision using the gradient pattern.

16. The measurement system of claim 1 where the data processing occurs locally.

17. The measurement system of claim 1 where the data processing occurs remotely and data is transmitted via Cloud processing