Method and apparatus for classifying objects and clutter removal of some three-dimensional images of the objects in a presentation

Patent metadata

Field	Value
Publication number	US-9842282-B2
Application number	US-201514719523-A
Country	US
Kind code	B2
Filing date	May 22, 2015
Priority date	May 22, 2015
Publication date	Dec 12, 2017
Grant date	Dec 12, 2017

Abstract

An approach is provided for classifying objects that are present at a geo-location and providing an uncluttered presentation of images of some of the objects in an application such as a map application. The approach includes determining one or more regions of interest associated with at least one geo-location, wherein the one or more regions of interest are at least one textured three-dimensional representation of one or more objects that may be present at the at least one geo-location. The approach also includes processing and/or facilitating a processing of the at least one textured three-dimensional representation to determine at least one two-dimensional footprint and three-dimensional geometry information for the one or more objects. The approach further includes causing, at least in part, a generation of at least one two-dimensional image representation of the one or more regions of interest by causing, at least in part, a projection of three-dimensional texture information of the at least one textured three-dimensional representation onto the at least one two-dimensional footprint. The approach also includes causing, at least in part, a classification of the one or more objects based, at least in part, on the at least one two-dimensional image representation and the three-dimensional geometry information.

First claim

What is claimed is:

1. A method comprising:
determining, by an apparatus, regions of interest associated with a geo-location, wherein each of the regions of interest is a textured polygon-based three-dimensional aerial photography representation of objects present at the geo-location;
processing, by the apparatus, the textured polygon-based three-dimensional representation to determine a two-dimensional footprint and three-dimensional geometry information for the objects per region;
in response to a selection of one or more aerial images, initiating, by the apparatus, an orthographic projection of three-dimensional texture information of the textured polygon-based three-dimensional aerial photography representation onto the two-dimensional footprint and a separation of the two-dimensional footprint into two-dimensional color image patches per region using top view information of the orthographic projection that includes respective colors and textures of tops of the objects while discarding front and side view information of the orthographic projection that include colors and textures of fronts and sides of the objects; and
initiating, by the apparatus, a classification of one or more of the two-dimensional color image patches each as corresponding to a plurality of building objects and other one or more of the two-dimensional color image patches each corresponding to one or more non-building objects based on information of respective aerial photography heights of the objects,
wherein a map application presentation is provided on a user interface based on the one or more of the two-dimensional color image patches each corresponding to the respective plurality of building objects.

2. A method of claim 1 , further comprising:
causing, at least in part, an extraction of one or more features from the two-dimensional color image patches,
wherein the classification of the one or more of the two-dimensional color image patches is based, at least in part, on one or more features.

3. A method of claim 2 , wherein the one or more features include, at least in part, a geometry feature, a structural feature, an edge intensity feature, a color feature, a color histogram feature, an edge orientation feature, or a combination thereof.

4. A method of claim 2 , further comprising:
causing, at least in part, a training of one or more classification models based, at least in part, on the one or more features; and
causing, at least in part, a classification of one or more other regions of interest based, at least in part, on the one or more classification models.

5. A method of claim 4 , further comprising:
determining classification accuracy information for the one or more classification models based, at least in part, on the classification of the one or more of the two-dimensional color image patches, the one or more other regions of interest, or a combination thereof that are associated with one or more previously labeled objects; and
causing, at least in part, an updating of (a) the one or more classification models; (b) the classification of the one or more of the two-dimensional color image patches, the classification of the one or more other regions of interest, or a combination thereof; or (c) a combination thereof based, at least in part, on the classification accuracy information.

6. A method of claim 1 , further comprising:
causing, at least in part, an initiation of at least one clutter removal process with respect to the regions of interest based, at least in part, on the classification.

7. A method of claim 1 , wherein the geo-location is a city, and the one or more non-building objects are further classified into a tree, a road, a car, a human, or a combination thereof, and the method further comprising:
determining probability information that the one or more non-building objects are a tree, a road, a car, a human, or a combination thereof per region.

8. A method of claim 1 , further comprising:
initiating, by the apparatus, a switch on the user interface between a presentation of only the other one or more two-dimensional color image patches each corresponding to the one or more non-building objects and the presentation of only the one or more of the two-dimensional color image patches each corresponding to the respective plurality of building objects; and
calculating one or more statistic features for the one or more two-dimensional color image patches each corresponding to the respective plurality of building objects per region, wherein the one or more statistic features includes geometry, color, structure, color histogram, edge density, edge orientation, or a combination thereof,
wherein the classification of the objects uses the one or more statistic features.

9. A method of claim 1 , wherein the textured polygon-based three-dimensional aerial photography representation is a triangular mesh three-dimensional representation on which the three-dimensional texture information is mapped.

10. A method of claim 1 , wherein the respective aerial photography heights of the objects are calculated based, at least in part, on base and roof elevations associated with the objects in the region.

11. An apparatus comprising:
at least one processor; and
at least one memory including computer program code for one or more programs,
the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to perform at least the following, determine regions of interest associated with a geo-location, wherein each of the regions of interest is a textured polygon-based three-dimensional aerial photography representation of objects present at the geo-location;
process the textured polygon-based three-dimensional representation to determine a two-dimensional footprint and three-dimensional geometry information for the objects per region;
in response to a selection of one or more aerial images, initiate an orthographic projection of three-dimensional texture information of the textured polygon-based three-dimensional aerial photography representation onto the two-dimensional footprint and a separation of the two-dimensional footprint into two-dimensional color image patches per region using top view information of the orthographic projection that includes respective colors and textures of tops of the objects while discarding front and side view information of the orthographic projection that include colors and textures of fronts and sides of the objects; and
initiate a classification of one or more of the two-dimensional color image patches each as corresponding to a plurality of building objects and other one or more of the two-dimensional color image patches each corresponding to one or more non-building objects based on information of respective aerial photography heights of the objects,
wherein a map application presentation is provided on a user interface based on the one or more of the two-dimensional color image patches each corresponding to the respective plurality of building objects.

12. An apparatus of claim 11 , wherein the apparatus is further caused to:
cause, at least in part, an extraction of one or more features from the two-dimensional color image patches,
wherein the classification of the one or more of the two-dimensional color image patches is based, at least in part, on one or more features.

13. An apparatus of claim 12 , wherein the one or more features include, at least in part, a geometry feature, a structural feature, an edge intensity feature, a color feature, a color histogram feature, an edge orientation feature, or a combination thereof.

14. An apparatus of

Assignees

• Here Global Bv

Inventors

• Liu Xiaoqing
• Adachi Jeffrey
• Haro Antonio
• Macfarlane Jane

Classifications

• G06T15/04 Primary

  Texture mapping · CPC title

• G06F18/24

  Classification techniques · CPC title

• G06T17/05

  Geographic models · CPC title

• G06K2009/4666

  Physics · mapped topic

• G06K9/46

  Physics · mapped topic