System and method for correcting high-definition map images

A processor-implemented method for correcting high-resolution map images, the method comprising:

• receiving, by a cloud-based processor from a plurality of vehicles, ground view image data of a map anomaly hotspot captured by the vehicles, wherein the map anomaly hotspot is an area described by a high-definition bird view map at which physical characteristics of the map anomaly hotspot as observed by the vehicles are inconsistent with a description for the area provided by the high-definition bird view map;
• converting the ground view image data to bird view image data using homography;
• removing outliers from the bird view image data;
• applying an error reduction algorithm for error reduction to the bird view image data with the removed outliers;
• stitching together a sequence of consecutive images from the bird view image data after error reduction to produce a high-definition image, wherein each of the consecutive images has been generated through converting the ground view image data to the bird view image data using homography, removing outliers from the bird view image data, and applying an error reduction algorithm for error reduction;
• comparing the stitched sequence of consecutive images to a pre-existing high-definition image of the map anomaly hotspot to determine revisions to the pre-existing high-definition image;
• and
• providing revised high-definition image data to an autonomous vehicle for use in navigating in an area near the map anomaly hotspot.

A non-transitory computer-readable medium having stored thereon instructions that when executed by a cloud-based processor cause the processor to perform a method for correcting high-resolution map images, the method comprising:

• receiving, from a plurality of vehicles, ground view image data of a map anomaly hotspot captured by the vehicles, wherein the map anomaly hotspot is an area described by a high-definition bird view map at which physical characteristics of the map anomaly hotspot as observed by the vehicles are inconsistent with a description for the area provided by the high-definition bird view map;
• converting the ground view image data to bird view image data using homography;
• removing outliers from the bird view image data;
• applying an error reduction algorithm for error reduction to the bird view image data with the removed outliers;
• stitching together a sequence of consecutive images from the bird view image data after error reduction to produce a high-definition image, wherein each of the consecutive images has been generated through converting the ground view image data to the bird view image data using homography, removing outliers from the bird view image data, and applying an error reduction algorithm for error reduction;
• comparing the stitched sequence of consecutive images to a pre-existing high-definition image of the map anomaly hotspot to determine revisions to the pre-existing high-definition image;
• and
• providing revised high-definition image data to an autonomous vehicle for use in navigating in an area near the map anomaly hotspot.