TomTom Global Content B.V.

A first digital map and a second digital map re processed to identify matching objects by a map data processing system. The first digital map is used to generate a first ordered array of pixel values. The second digital map is used to generate a second ordered array of pixel values. A machine learning (ML) model is used to determine at least one score indicative of a likelihood that the first digital map object corresponds to the second digital map object.

Certain examples provide a computer-implemented method for managing map data, the method comprising: receiving one or more map data change requests indicative of a request for a change to be applied to a state of map data to generate another state of the map data; sending the one or more changes of the one or more map data change requests to a second apparatus, wherein the second apparatus stores one or more states of map data; receiving a request for at least a portion of a particular state of map data; determining whether the particular state of map data is stored at the second apparatus; and obtaining, from the second apparatus, the at least portion of the particular state of map data.

A method and system for obtaining, from at least one camera associated with a vehicle traveling through a road network, a sequence of images of a road or a road environment, each image being associated with a location where that image is captured, generating a local map representation of an area of the road network using at least some images from the sequence of images and the locations associated therewith, the generating including: processing the at least some of the images to detect an object in the road or the road environment, determining at least one transformation for tracking the object between the at least some of the images and, based on the at least one transformation and the locations associated with the at least some of the images, generating a two- or three-dimensional representation of the object relative to the area of the road network, comparing the local map representation with a reference map covering the area of the road network, and determining, based on the comparison, a geographical location and an orientation of the vehicle within the road network.

Various embodiments of methods and systems for automated update of digital map data based on media messages are described herein. Communication messages are collected from one or more media services. Information relevant to an accident and comprising at least one road reference is extracted from the collected messages. Further, the extracted information is matched with a digital map to identify at least one road and/or road segment affected by the incident. At least one property associated with the at least one road and/or road segment is adjusted in the digital map based on the extracted information as matched. Based on the adjusting, information about an intensity of the incident associated with the at least one road and/or road segment is presented to clients.

A method and a system include associating a value with each of one or more maneuvers available at a central node of the set of nodes, selecting a path between a pair of nodes of the set of nodes based on a predefined criterion, and changing the value associated with a respective maneuver from the one or more maneuvers at the central node when the respective maneuver is part of the selected path.

A device is provided to extract scan lines data from images of a route, said scan line data comprising only a portion of the an image that extends along a scan line, and to store this data in memory together with respective positional data pertaining to the position of the device traveling along the route in a navigable network. This scan line and positional data can be obtained from a plurality of devices in a network and transmitted to computing apparatus, for example, for use in a production of a realistic view of a digital map.

For performance of a map data-related function, a processing system is operative to access electronic map data stored in a plurality of files. Each of the plurality of files is associated with a tile of a tiling. Different tiles relate to different geographic regions. At least some of the files may include references, which are used to obtain object data for at least one map object. A tile directory may provide version information useful for retrieval of the object data.

A processing system processes probe data to determine a change in a number of vehicles parked in an area, the probe data comprising information on probe traces for a plurality of probes. The processing system identifies occurrence of an event based at least on the determined change in the number of vehicles parked in the area. The processing system causes the identified occurrence of the event to be used for performing one or several operations.

The invention relates to a method of training a model for use in semantic image segmentation. The model includes a first classifier neural network and a second classifier neural network. The method includes training the first network by inputting a first training image containing a target object to the first network, using the first network to identify and erase pixels of the first training image that are discriminative for the target object, inputting the first training image to the second network, using the second network to determine a likelihood of the first modified training image containing the target object and updating weights of the first network using a first loss function that is a monotonically-increasing function of the determined likelihood.

Disclosed herein is a technique for generating and providing an indication to an autonomous vehicle regarding the confidence level for the accuracy or quality of the map data in which the indication is determined from observation data received from other vehicles.