Radar-Image Object Matching with Likelihood-Based Candidate Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object determination apparatuses face challenges in accurately determining whether an object detected by an electromagnetic wave sensor and an image sensor are the same object, especially when multiple objects are in proximity, leading to potential false determinations due to overlapping search regions.
Innovation Solution
The apparatus employs a same-object determiner, object-type identifier, likelihood calculator, and candidate-object selector to calculate likelihoods for candidate objects based on detection information from the electromagnetic wave sensor, preferentially selecting candidates with higher likelihoods for the identified object type, thereby reducing false determinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If search regions are set based on positions of objects detected by electromagnetic wave sensor and image sensor, then object matching can be performed, but false determinations occur when multiple objects are in proximity
Solution Approach 1:
The patent changes the parameter of search region size dynamically. When multiple candidate objects are detected, the search region size is adjusted based on the number of candidates and their spatial distribution. This allows the system to differentiate between closely spaced objects by adapting the matching criteria to the specific scene complexity, thereby reducing false determinations while maintaining detection accuracy.
Solution Approach 2:
The patent segments the object matching process into multiple stages: initial detection by electromagnetic wave sensor, candidate identification by image sensor, likelihood calculation for each candidate, and selective detailed matching. This segmentation allows the system to handle multiple objects in proximity by processing them hierarchically, reducing computational complexity and improving accuracy by focusing detailed analysis only on promising candidates.
2Measurement precision
If likelihood calculation is performed for all candidate objects, then accurate object type identification is achieved, but computational complexity increases
Solution Approach 1:
The patent applies partial action by performing comprehensive likelihood calculation only for a limited number of candidate objects that meet certain criteria (e.g., within a threshold distance, matching basic object type characteristics). Instead of calculating likelihoods for all detected objects, the system selectively processes only those that are plausible candidates, thereby maintaining high identification accuracy while significantly reducing computational complexity.
Solution Approach 2:
The patent performs preliminary filtering of candidate objects based on basic criteria (position overlap, object type consistency) before conducting detailed likelihood calculation. This preliminary action eliminates obviously incorrect candidates early in the process, reducing the number of objects requiring computationally intensive likelihood analysis and thus lowering overall system complexity while preserving accuracy for relevant candidates.
Data Source
AI summary
In an object determination apparatus, a same-object determiner is configured to make a same-object determination as to whether a first object ahead of a subject vehicle that is a vehicle carrying the object determination apparatus, detected by an electromagnetic wave sensor, and a second object ahead of the subject vehicle, detected by an image sensor, are the same object. A candidate-object identifier is configured to identify a candidate for the first object, between which and the second object the same-object determination is to be made, as a candidate object. A candidate-object selector is configured to, in response to there being a plurality of the candidate objects, preferentially select, from the plurality of candidate objects, a candidate object whose likelihood for the identified object type of the second object is higher than a predetermined likelihood threshold, as a candidate object to be subjected to the same-object determination.