METHODS AND SYSTEMS FOR OBJECT BASED GEOMETRIC FITTING

• US 20160343152A1
• Filed: 05/20/2016
• Published: 11/24/2016
• Est. Priority Date: 05/22/2015
• Status: Active Grant

• Current Assignee: Myheat Inc.
• Original Assignee: Myheat Inc.
• Inventors: Hay, Geoffrey J., Couloigner, Isabelle
• Granted Patent: US 10,013,785 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 11/60 Editing figures and text; Combining figures or text
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/10048 Infrared image
• CPC Class Codes: G06T 2207/30184 Infrastructure
• CPC Class Codes: G06T 2215/16 Using real world measurements to influence rendering
• CPC Class Codes: G06T 7/337 involving reference images or patches

Abstract

Described herein are various implementations of a process and a system for automatically performing object based geometric fitting. The process may include spatially superimposing a mask dataset comprising mask-objects on a raster image comprising image-objects, and locally superimposing individual mask-objects on individual image-objects, until a dataset is obtained wherein mask-objects are geometrically fitted to image-objects.

Claims

1. An automated process for geometric data fitting, the process comprising:

(a) obtaining a raster image having a spatial resolution and comprising a plurality of image-objects representing real-world objects;

(b) obtaining a mask dataset comprising a plurality of mask-objects, each mask-object having a center, and each mask-object spatially corresponding to one of the image-objects in the raster image at the same spatial resolution;

(c) defining (i) a distance threshold and (ii) a pseudo-rotation angle applicable to a shift function;

(d) spatially superimposing the mask dataset and the raster image in such a manner that the locations of the plurality of mask-objects approximately corresponds to the locations of the plurality of corresponding image-objects;

(e) locally superimposing a plurality of mask-objects by incrementally moving the plurality of mask-objects over the spatially corresponding image-objects using a shift function including the defined (i) distance threshold and (ii) pseudo-rotation angle until an optimal homogeneity is attained for a plurality of mask-objects; and

(f) generating an optimally fitted dataset comprising a plurality of mask-objects which are individually geometrically fitted.

3. The process according to claim 1 wherein the real-world objects represented by the raster image are geographical objects.

2. An automated process for geometric data fitting, the process comprising:

(a) obtaining a raster image having a spatial resolution and comprising a plurality of image-objects representing real-world objects;

(b) obtaining a mask dataset comprising a plurality of mask-objects, each mask-object having a center, and each mask-object spatially corresponding to one of the image-objects in the raster image at the same spatial resolution;

(c) defining (I) a distance threshold and (ii) a pseudo-rotation angle applicable to a shift function;

(d) spatially superimposing the mask dataset and the raster image in such a manner that the locations of the plurality of mask-objects approximately corresponds to the locations of the plurality of corresponding image-objects;

(e) locally superimposing a plurality of mask-objects by incrementally moving the plurality of mask-objects over the spatially corresponding image-objects using a shift function by, for each mask-object

(i) determining the homogeneity based on digital numbers in the corresponding raster image that fall within the area of the mask-object;

(ii) incrementally shifting the mask-object around its center based on a pseudo-rotation angle projected in the x and y direction;

(iii) at each shift increment, calculating and saving the homogeneity of the digital numbers in the raster image that fall within the overlapping area of the mask-objects;

(iv) iterating (e)(i) to (e)(iii) until the distance threshold is reached or until an optimal homogeneity is attained; and

(v) defining the location and boundary of the mask-object at the optimal homogeneity; and

(f) generating a fitted dataset comprising a plurality of mask-objects which are individually geometrically fitted.

4. A system for providing automated process for geometric data fitting, wherein the system comprises:

an input interface for receiving a raster image having a spatial resolution and comprising a plurality of image-objects representing real-world objects;

an output interface for providing an indication of geometrically fitted mask-objects corresponding to the image-objects; and

a processing unit coupled to the input and the output interfaces, the processing unit being configured to generate a mask dataset comprising a plurality of mask-objects that each have a center and spatially correspond to one of the image-objects in the raster image at the same spatial resolution;

to spatially superimpose the mask dataset and the raster image in such a manner that the locations of the plurality of mask-objects approximately corresponds to the locations of the plurality of corresponding image-objects;

to locally superimpose the plurality of mask-objects by incrementally moving the plurality of mask-objects over the spatially corresponding image-objects using a shift function operating under a defined distance threshold and using a pseudo-rotation angle until an optimal homogeneity is attained for the plurality of mask-objects; and

to define a geometrically fitted dataset comprising the plurality of geometrically shifted mask-objects which are individually optimally and geometrically fitted to the corresponding image-objects.

5. The system of claim 4, wherein the processor is configured to provide a display geometrically fitted mask-objects via the output interface.

6. The system of claim 4, wherein the processor is configured to store the geometrically fitted dataset in a memory element.

7. The system of claim 4, wherein the processor is configured to send the geometrically fitted dataset to another electronic device.

8. The system of claim 4, wherein the processor is configured to define the distance threshold and the pseudo-rotation angle applicable to a shift function prior to shifting the plurality of mask-objects.