Constructing map data using laser scanned images
First Claim
Patent Images
1. A computer, comprising a processor and memory that stores processor-executable instructions, wherein the processor is programmed to:
- receive a plurality of map-perspective patch images of a physical region near a vehicle and imaged by at least one laser scanner on the vehicle, wherein each of the plurality of map-perspective patch images represent a measure of ground reflectivity;
generate a plurality of transform patch images by applying a mathematical transform operator to at least some of the plurality of map-perspective patch images;
determine, from among the plurality of transform patch images, at least two of the plurality of transform patch images having the largest gradients;
combine the at least two of the plurality of transform patch images into a single patch image; and
construct a map patch image using the single patch image and at least one of the plurality of map-perspective patch images.
1 Assignment
0 Petitions
Accused Products
Abstract
A computer that includes a processor and memory that stores processor-executable instructions. The processor can be programmed to: generate a plurality of transform patch images (PTPIs) by applying a mathematical transform operator to a plurality of map-perspective patch images (PMPPIs) received from at least one vehicle laser-scanner; combine at least some of the PTPIs into a single patch image; and construct a map patch image using the single patch image.
-
Citations
20 Claims
-
1. A computer, comprising a processor and memory that stores processor-executable instructions, wherein the processor is programmed to:
-
receive a plurality of map-perspective patch images of a physical region near a vehicle and imaged by at least one laser scanner on the vehicle, wherein each of the plurality of map-perspective patch images represent a measure of ground reflectivity; generate a plurality of transform patch images by applying a mathematical transform operator to at least some of the plurality of map-perspective patch images; determine, from among the plurality of transform patch images, at least two of the plurality of transform patch images having the largest gradients; combine the at least two of the plurality of transform patch images into a single patch image; and construct a map patch image using the single patch image and at least one of the plurality of map-perspective patch images. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14)
-
-
15. A method, comprising:
at a computer in a vehicle; receiving a plurality of uncalibrated map-perspective patch images of a physical region near the vehicle and imaged by at least one scanner on the vehicle, wherein each of the plurality of uncalibrated map-perspective patch images represent a measure of ground reflectivity; generating a plurality of transform patch images by applying a mathematical transform operator to at least some of the plurality of map-perspective patch images; determining, from among the plurality of transform patch images, at least two of the plurality of transform patch images having the largest gradients; combining the at least two of the plurality of transform patch images into a single fused patch image; and constructing a map patch image using the fused patch image and at least one of the plurality of map-perspective patch images. - View Dependent Claims (16, 17, 18)
-
19. A computer, comprising a processor and memory that stores processor-executable instructions, wherein the processor is programmed to:
-
generate a plurality of transform patch images (PTPIs) by applying a mathematical transform operator to a plurality of map-perspective patch images (PMPPIs) received from at least one vehicle laser-scanner; determine, from among the PTPIs, at least two of the PTPIs having the largest gradients; combine the at least two of the PTPIs into a single patch image; and construct a map patch image using the single patch image. - View Dependent Claims (20)
-
Specification