LOCATION MEASURING DEVICE AND METHOD

• US 20090175498A1
• Filed: 07/02/2008
• Published: 07/09/2009
• Est. Priority Date: 07/06/2007
• Status: Active Grant

• Current Assignee: Topcon Corporation
• Original Assignee: Topcon Corporation
• Inventors: ANAI, Tetsuji, OTANI, Hitoshi, KOCHI, Nobuo
• Granted Patent: US 7,844,077 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06T 2200/08 involving all processing steps from image acquisition to 3D model generation
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 7/579 from motion
• CPC Class Codes: G06T 7/74 involving reference images or patches

Abstract

To realize high speed and high precision with device and method of three-dimensional measurement by applying estimating process to points corresponding to feature points in a plurality of motion frame images. With the device and method of calculating location information through processes of choosing a stereo pair, relative orientation, and bundle adjustment and using corresponding points of feature points extracted from respective motion frame images, each process is made up of two stages. To the first process section (stages:5A and 6A), robust estimation is applied to determine estimated values of calculation parameters. In the second process section (stages:5B and 6B), corresponding points and stereo pair of large residual errors are excluded, and estimating calculation is carried out again to make measurements in three dimensions.

Claims

1. A location measuring device comprising:

a feature point extracting section for extracting a plurality of feature points on a photographing object from each of successively photographed images in a plurality of frames obtained with an image obtaining means, a location of the image obtaining means relatively varying to a location of the photographing object with time;

a feature point tracking section for searching each of the successively photographed images for points corresponding to the feature points and tracking the feature points;

a first relative orientation process section for choosing one photographed image frame and another photographed image frame from tracked results of the feature point tracking section, carrying out relative orientation based on a plurality of points randomly chosen out of the respective corresponding points between the two photographed image frames, determining vertical parallaxes, applying robust estimation to the determined vertical parallaxes, and determining estimated values of parameters for the relative orientation while reducing influence of errors;

a second relative orientation process section for carrying out relative orientation between the respective corresponding points of the two image frames based on the parameters for the relative orientation estimated in the first relative orientation process section, calculating vertical parallaxes of the feature points, applying threshold process to the vertical parallaxes, and excluding incorrect corresponding points or determining appropriateness of the photographed image frames; and

a location measuring section for measuring a coordinate of the photographing object, or the location or a photographing posture of the image obtaining means using the plurality of photographed image frames.

2. The location measuring device as recited in claim 1, wherein the device comprises a GPS location data acquiring section to obtain location data of the image obtaining means from the GPS data, and choose a stereo pair having a sufficient baseline length from acquired location data.

3. The location measuring device as recited in claim 1, wherein the robust estimation is one of LMedS method, median estimation (M estimation) method, and RANSAC method.

4. A location measuring device comprising:

a feature point extracting section for extracting a plurality of feature points on a photographing object from each of successively photographed images in a plurality of image frames obtained with an image obtaining means, a location of the image obtaining means relatively varying to a location of the photographing object with time;

a feature point tracking section for searching each of the successively photographed images for points corresponding to the feature points and tracking the feature points;

a bundle adjustment process section for choosing a plurality of image frames from results of the tracking by the feature point tracking section, applying bundle adjustment process to coordinate locations of the feature points between the plurality of image frames or to photographing locations of the respective frames, and applying robust estimation to coordinate location data of the feature points determined by the bundle adjustment process or to residuals of the photographing locations of the respective frames to determine estimated data of the coordinate location or the residuals;

a choice-removal process section for, based on the estimated data determined in the bundle adjustment process section, determining threshold, and removing incorrect corresponding points or choosing image frames according to the threshold; and

a location measuring section for measuring coordinates of the photographing object, or the location or a photographing posture of the image obtaining means by using the bundle adjustment results obtained using the feature points and the image frames determined to be appropriate.

5. The location measuring device as recited in claim 4, wherein a GPS location data taking section for acquiring the location data of the image obtaining means from GPS data is provided to choose the photographed image frames from the location data.

6. The location measuring device as recited in claim 4, wherein the robust estimation is one of LMedS method, median estimation (M estimation) method, and RANSAC method.

7. A location measuring method comprising:

a feature point extracting step of extracting a plurality of feature points on a photographing object from each of successively photographed images in a plurality of image frames obtained with an image obtaining means, a location of the image obtaining means relatively varying to a location of the photographing object with time;

a feature point tracking step of searching each of the successively photographed images for points corresponding to the feature points and tracking the feature points;

a first relative orientation process step of choosing a stereo pair of one photographed image frame and another photographed image frame from results of the tracking by the feature point tracking section, carrying out relative orientation based on a plurality of points randomly chosen out of the respective corresponding points between the two photographed image frames, determining vertical parallaxes, applying robust estimation to the determined parallaxes, and determining estimated values of parameters for the relative orientation while reducing influence of errors;

a second relative orientation process step of carrying out relative orientation between the respective corresponding points of the two image frames based on the parameters for the relative orientation estimated in the first relative orientation process step, calculating vertical parallaxes of the feature points, applying threshold process to the vertical parallaxes, and excluding incorrect corresponding points or determining appropriateness of the photographed image frames; and

a location measuring step of measuring a coordinate of the photographing object, or the location or a photographing posture of the image obtaining means using the plurality of photographed image frames.

8. A location measuring method comprising:

a feature point extracting step of extracting a plurality of feature points on a photographing object from each of successively photographed images in a plurality of frames obtained with an image obtaining means, a location of the image obtaining means relatively varying to a location of the photographing object with time;

a feature point tracking step of searching each of the successively photographed images for points corresponding to the feature points and tracking the feature points;

a first bundle adjustment process step of choosing a plurality of image frames from results of the tracking by the feature point tracking section, applying bundle adjustment process to coordinate locations of the feature points between the plurality of image frames or to photographing locations of the respective frames, and applying robust estimation to the coordinate location data of determined feature points or to the residuals of the photographing locations of the respective frames to determine estimated data of the coordinate location or the residuals;

a choice-removal process step of, based on the estimated data determined in the bundle adjustment process step, determining threshold, and removing incorrect corresponding points or choosing image frames according to the threshold; and

a location measuring step of measuring coordinates of the photographing object, or the location or a photographing posture of the image obtaining means by using the bundle adjustment results obtained using the feature points and the image frames determined to be appropriate.