Probe Trace Correlation via Pixel Map Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for processing probe data from navigation devices, such as GPS-enabled PNDs, face challenges in determining the reliability of probe traces due to noise and spatial contamination, leading to inaccurate digital map updates and inefficient data analysis.

Innovation Solution

A method and system that utilize a pixel map to assign correlation scores to probe traces by mapping them to a parameter space, allowing for efficient and accurate determination of trace correlation, which reduces data generalization and spatial contamination issues by weighting and normalizing the data, thereby improving the reliability of digital map updates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If spatial clustering techniques are used to process probe traces, then probe data can be organized and analyzed, but data generalization occurs and important information is lost or distorted

Engineering Contradiction:
Improvedata processing efficiencyVSAvoidprobe trace information
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent divides the parameter space into discrete pixels, creating a grid-based representation of probe trace data. Each pixel represents a specific region in the parameter space, allowing data to be organized without forcing unrelated traces into the same cluster. This segmentation approach maintains information integrity while enabling efficient processing through the pixelated map structure.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If probe traces are clustered spatially, then nearby segments can be identified, but traces may be incorrectly matched to parallel road segments due to spatial contamination

Engineering Contradiction:
Improvetrace position accuracyVSAvoidspatial contamination
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent applies local quality by allowing different pixels to have different correlation values based on the density and distribution of probe traces in their respective regions. Each pixel independently evaluates the correlation of traces passing through it, enabling local accuracy assessment without being influenced by contamination from distant or unrelated traces in other spatial regions.

Inventive Principle:
Principle #3Local quality

3Reliability

If all probe trace points are connected by lines and placed into a geographic index, then spatial relationships can be analyzed, but computational burden increases significantly

Engineering Contradiction:
Improvetrace correlation accuracyVSAvoidcomputational complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent transforms the continuous geographic index into a discrete pixel-based representation, adding a dimensional abstraction layer. Instead of directly processing continuous spatial coordinates and their relationships, the system maps all probe trace data onto a pixel grid, converting complex spatial relationships into simpler pixel-coordinate relationships. This dimensional transformation reduces computational complexity while preserving essential spatial correlation information.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP2556338B1Probe data processing
Publication Date: 2018.02.21 TOMTOM NORTH AMERICA INC
  • EP2556338B1 patent drawingFigure 1~2
  • EP2556338B1 patent drawingFigure 3~4
  • EP2556338B1 patent drawingFigure 5

AI summary

A method of processing probe trace data to determine a measure of correlation of a probe trace with other probe traces comprises obtaining a map that associates a parameter space with a plurality of pixels, wherein at least one correlation value is assigned to each of the plurality of pixels, and determining a correlation score for a probe trace by mapping the probe trace to at least one pixel of the map and determining the correlation score for the probe trace from at least one correlation value of the at least one pixel to which the probe trace is mapped.