Vehicle Location Estimation Using Weighted Identification Cues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle location estimation systems face challenges in accurately estimating the location of driving vehicles, particularly when using low-resolution cameras, which can lead to misrecognition and reduced accuracy due to the difficulty in extracting property information such as license plates from images.
Innovation Solution
A vehicle location estimating apparatus and method that applies a weighted value based on identification information obtained from surrounding images to improve the accuracy of location estimation, considering probability information for properties other than license plates to reduce misrecognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If low-resolution cameras are used to capture surrounding images, then device complexity and cost are reduced, but measurement precision and reliability of vehicle identification deteriorate
Solution Approach 1:
The patent combines multiple identification features (license plate, vehicle type, color, size) into a composite identification system. By merging these different property information sources and applying weighted values based on their extraction probabilities, the system achieves reliable vehicle identification even when individual features are difficult to extract from low-resolution images
Solution Approach 2:
The system dynamically adjusts the weighting of different identification properties based on their extraction probabilities. When certain properties (like license plate) have low extraction probability due to image quality, the system automatically reduces their weight and relies more on other properties (like vehicle type and color) that can be reliably extracted, thus adapting to varying image conditions
2Adaptability or versatility
If property information extraction is relied upon for vehicle identification, then identification capability is enhanced, but misrecognition errors increase due to low image quality
Solution Approach 1:
The system implements feedback through probability calculation - it continuously evaluates the extraction probability of each property information and adjusts the weighted value accordingly. This feedback mechanism allows the system to recognize when extraction is unreliable and compensates by relying on other properties with higher extraction probabilities, thereby reducing misrecognition
Solution Approach 2:
Instead of requiring all identification properties to be extracted with high confidence, the system uses partial information effectively. It applies weighted values based on extraction probabilities, allowing identification to proceed with incomplete or partially certain property information, thus maintaining identification capability while managing misrecognition risk
3Measurement precision
If weighted values based on extraction probability are applied to location information, then location estimation accuracy is improved, but computational complexity increases
Solution Approach 1:
The system applies different weighted values to different property information based on their local extraction probabilities. Each property (license plate, vehicle type, color) receives a weight appropriate to its reliability in the current context, allowing accurate location estimation without requiring uniform high-quality extraction of all properties
Data Source
AI summary
In accordance with an aspect of the present disclosure, there is provided an apparatus for estimating a location of a vehicle including, a communication unit configured to receive, from an information providing vehicle, identification information and location information on a driving vehicle in a vicinity of the information providing vehicle, a weighted value obtaining unit configured to obtain a weighted value representing accuracy of the location information based on the received identification information and a location estimating unit configured to estimate a location of the driving vehicle by applying the weighted value to the location information.