Method for capturing, measuring and analyzing motion

• US 7,428,318 B1
• Filed: 12/13/2004
• Issued: 09/23/2008
• Est. Priority Date: 12/11/2003
• Status: Active Grant

• Current Assignee: Motion Reality, Inc.
• Original Assignee: Motion Reality, Inc.
• Inventors: Wilmink, Ronnie J. H., McLaughlin, Thomas M., Madsen, Nels H.
• Primary Examiner(s): Mehta; Bhavesh
• Assistant Examiner(s): Tabatabai; Abolfazl
• Application Number:
  US11/011,540
• Time in Patent Office: 1,380 Days
• Field of Search: 382/103, 382/107, 382/326, 348/14.1, 348/97, 348/154, 348/155, 348/208.4, 348/208.13, 348/208.16, 348/352, 348/407.1
• US Class Current: 382/107
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 2207/30204 Marker
• CPC Class Codes: G06T 7/136 involving thresholding
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments

Abstract

A system and method for capturing, measuring and analyzing motion in real-time is provided which operates with or without markers. A threshold image is based on a background image and is created to minimize noise during the motion capture. A captured image is compared to the threshold image to identify collections of hot pixels or globs. The globs are compared to expected characteristics of the markers or the subject and are tracked between frames. Glob information is used to determine the location of the markers or subject in each frame.

Claims

1. A method for capturing motion, comprising:

receiving a captured image having a plurality of pixels arranged in rows and columns;

for each pixel of the captured image, comparing an intensity of the pixel of the captured image to an intensity of a corresponding pixel in a threshold image;

if the intensity of the pixel of the captured image exceeds the intensity of the corresponding pixel in the threshold image, then designating the pixel of the captured image as a hot pixel;

comparing hot pixels in adjacent rows to determine related hot pixels;

designating the related hot pixels as a glob;

comparing characteristics of the glob with predetermined characteristics of an item to track the motion associated with the item;

if the characteristics of the glob satisfy the predetermined characteristics of the item, then determining three dimensional coordinates for candidate points corresponding to the glob;

determining track attributes for the candidate points, wherein the track attributes describe the candidate point'"'"'s movement from frame to frame;

comparing the track attributes for the candidate points to expected attributes of the motion of the item; and

if the track attributes satisfy the expected attributes, then determining that the candidate points correspond to the item.

2. The method of claim 1 wherein the item is a reflective marker.

3. The method of claim 1, wherein the item is an object.

4. A method for capturing a background image, comprising:

for each of a plurality of cameras,

capturing a plurality of images over a period of time using a predetermined shutter speed, wherein each image has a plurality of pixels;

for each pixel of a selected captured image,

comparing an intensity of the pixel with an intensity of a corresponding pixel of other ones of the captured images;

determining a maximum intensity of the pixels; and

using the maximum intensity of the pixels as the intensity for the corresponding pixel of the background image;

capturing a second plurality of images over a second period of time using a second predetermined shutter speed;

for each pixel of the second captured images comparing an intensity of the pixel with a predetermined intensity threshold;

if the pixel satisfies the predetermined intensity threshold, then setting the corresponding pixel in the background image to the predetermined intensity threshold.

5. A method for capturing motion associated with a first item, wherein the first item is associated with a marker and a second item, wherein the second item is not associated with a marker, comprising:

receiving a captured image having a plurality of pixels arranged in rows and columns;

comparing the captured image to a threshold image on a pixel-by-pixel basis;

based on the comparison, identifying globs within the captured image;

comparing the globs to a first set of criteria, wherein the first set of criteria correspond to characteristics of the marker;

comparing the globs to a second set of criteria, wherein the second set of criteria correspond to characteristics of the second item;

based on the comparison to the first set of criteria and the second set of criteria, identifying any globs that correspond to the marker and identifying any globs that correspond to the second item to track the motion of the marker and the second item;

wherein a region of interest defines an expected range of locations for the second item, and wherein comparing the captured image to a threshold image on a pixel-by-pixel basis, comprises;

for each pixel of the captured image, comparing an intensity of the pixel of the captured image to an intensity of a corresponding pixel in the threshold image; and

for each pixel of the captured image that corresponds to the region of interest, comparing the intensity of the pixel of the captured image to a predetermined intensity associated with the second item.

6. A method for creating a threshold image, comprising:

determining a center of an image for curvature falloff;

lowering a threshold intensity until detected noise exceeds a predetermined threshold;

raising the threshold intensity until the detected noise satisfies a second predetermined threshold;

lowering a threshold curvature until detected peripheral noise exceeds a third predetermined threshold; and

raising the threshold intensity and the threshold curvature until the detected noise satisfies a third predetermined threshold corresponding to noise in the center of the image and the detected peripheral noise satisfies a fourth predetermined threshold that corresponds to peripheral noise in the image,

wherein determining the detected noise and the peripheral noise includes determining a number of globs in the image and comparing the globs to certain thresholding parameters.

7. The method of claim 6, wherein raising the threshold intensity and the threshold curvature comprises:

raising the threshold intensity to limit noise in the center of the image; and

raising the curvature to address noise in the periphery of the image.