Image Region Selection for Occlusion-Robust Position Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Occlusion caused by movable objects, such as people or vehicles, can lead to decreased accuracy in position estimation for autonomously movable apparatuses, as these objects hide stationary objects used for estimation, resulting in noise and reduced accuracy.
Innovation Solution
A position estimating apparatus that calculates an evaluation value for image regions and compares them with a reference image to determine the current position of the movable apparatus, excluding regions with low evaluation values to mitigate the impact of occlusion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If position estimation is performed using all regions of the captured image, then the position estimation can be conducted using maximum available information, but the accuracy decreases due to occlusion by movable objects introducing noise
Solution Approach 1:
The image is divided into multiple regions, and an evaluation value is calculated for each region to determine its suitability for position estimation. This segmentation allows the system to identify and exclude regions affected by occlusion while utilizing regions with high evaluation values, thereby maintaining position estimation accuracy despite the presence of movable objects.
Solution Approach 2:
Different regions of the image are assigned different evaluation values based on their local characteristics and suitability for position estimation. Regions with high evaluation values are selected for estimation, while regions with low evaluation values (likely occluded) are excluded. This local quality assessment ensures that only reliable regions contribute to the final position estimation.
2Measurement precision
If regions with occlusion are excluded from position estimation, then the accuracy is maintained, but the amount of usable information for estimation is reduced
Solution Approach 1:
The system calculates an evaluation value for each image region based on multiple parameters including distance information, image intensity, and variance. By changing and analyzing these parameters, the system can objectively determine which regions are suitable for position estimation and which should be excluded, optimizing the balance between accuracy and information utilization.
3Reliability
If the evaluation value calculation includes distance information and image characteristics, then the selection of suitable regions is improved, but the computational complexity increases
Solution Approach 1:
The system calculates evaluation values for all image regions using multiple parameters (distance, intensity, variance), which may seem excessive. However, this comprehensive approach ensures high reliability in region selection. The computational load is managed by efficiently processing these parameters and using the evaluation values to filter out unsuitable regions before performing the actual position estimation.
Data Source
AI summary
A position estimating apparatus includes a memory that stores a reference image, an interface circuit configured to communicate with a movable apparatus, and a processor. The processor is configured to, upon receipt of at least one image captured by the movable apparatus via the interface circuit, calculate an evaluation value for each of a plurality of regions of the image, and determine a current position of the movable apparatus by comparing the regions of the captured image where the calculated evaluation value exceeds a first threshold with the reference image.