HERE Global

An approach is provided for path-based map matching. The approach involves, for instance, executing a first pass map-matching process to determine on-road candidate matches for observation points of the trajectory. The approach also involves determining emission probabilities of the candidate matches and transition probabilities between on-road candidates of subsequent observations, and determining a most probable path based on the emission and transition probabilities. The approach further involves determining a set of a predetermined number of on-road candidate matches from the most probable path. The approach further involves marking at least one match in the set as not matched based on determining that an alternate transition with a higher transition probability than the at least one transition exists. The approach further involves redetermining the most probable path using a second pass map-matching process based on the marked match.

Example embodiments describe a computer-implemented method, an apparatus, and a computer program product for generating personalized content for users traversing a route. The method may obtain a route and its content attention profile, which assigns an estimated content attention level to each road link based on user attention levels corresponding to various road link features. A control prompt may be generated based on this profile and a first content data file containing text, audio, or video data. The control prompt may instruct adjusting the detail level of the content. A generative artificial intelligence component may process the first content data file and control prompt to generate a second content data file. The second content data file may be presented on a user device. This method may provide an adaptive entertainment experience for users along their route, potentially enhancing engagement and enjoyment during travel.

Methods, apparatuses and computer program products for updating map data are disclosed. The method for updating a map comprises obtaining first map data to be updated, obtaining second map data to be used for updating the first map data, determining a matched second map road segment of a road of the second map data matched to a corresponding matched first map road segment of a road of the first map data, registering the matched second map road segment to the matched first map road segment; and updating the first map data by updating the first map road segment.

A road sign detection method comprises obtaining an unlabeled image of a road segment and executing a neural network trained to identify road signs within the unlabeled image. The neural network comprises a first model configured to localize an object in the unlabeled image. The neural network may further comprise a second model configured to generate classification data associated with the cropped image, based on content of the cropped image, where the classification data corresponds to one or more appearance-based attributes of the cropped image and also to determine a category of the cropped image, based on the classification data of the cropped image. The neural network may comprise a third model configured to generate road sign label data based on the determined category of the cropped image, where the road sign label data represents the identified road sign. The method further comprises outputting the road sign label data.

An approach is provided for matching to directed graph representation of lanes of a road topology network of a digital map. The approach, for instance, involves determining a plurality of lane group centerlines respectively representing a plurality of lane groups of a road network in a geographic database. The approach also involves for each lane group centerline of the plurality of lane group centerlines, determining one or more edges having a pair of node identifiers. The approach further involves inserting the one or more edges and the pair of node identifiers into a lane group routing graph, wherein the lane group routing graph is a directed graph. The approach further involves matching an input directed graph to the lane group routing graph, wherein the input directed graph represents a set of trajectories, and providing a map matching result based on the matching.

An approach is provided for lane-level direction-of-travel detection from overhead imagery. The approach, for instance, involves processing overhead imagery data of a road network to determine a plurality of features indicative of a direction of travel of at least one road lane of the road network. Each feature of the plurality of features indicates a candidate direction of travel. The approach also involves determining the direction of travel of the at least one road lane based on a weighted voting mechanism applied based on the candidate direction of travel of each feature. The weighted voting mechanism weighs the candidate direction of travel of each feature based on a feature type of each feature. The approach further involves associating the determined direction of travel as an attribute of the at least one road lane in a geographic database.

An approach is provided for map matching a vehicle path network to a road topology network of a digital map. The approach, for instance, involves aggregating vehicle trajectory data into a vehicle path network that is a directed graph comprising a plurality of nodes. The approach also involves segmenting the vehicle path network into a plurality of linear features. Each linear feature of the plurality of linear features connects two nodes of the plurality of nodes in the vehicle path network. The approach further involves using a path-based map matcher to separately match each linear feature by incorporating context of the vehicle path network to a map topology of a digital map to generate map matching results. The approach further involves aggregating the map matching results of each linear feature into an overall map matching result for the directed graph, and providing the overall map matching result as an output.

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …