Stereo Parallax Object Ranging Without Longer Camera Baselines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distance measurement systems face challenges in achieving high accuracy while maintaining a small apparatus size, particularly when using multiple imaging devices with different viewpoints, as the baseline length between viewpoints affects accuracy and distinguishing objects from backgrounds.
Innovation Solution
An information processing apparatus and method that utilizes multiple viewpoint images to detect object regions, set distance measurement points, and generate distance information using parallax calculations, employing techniques like ellipse fitting and stereo matching to improve accuracy without increasing the baseline length.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the baseline length between multiple imaging apparatuses is increased to improve distance measurement accuracy, then the measurement precision is improved, but the apparatus size increases
Solution Approach 1:
The patent segments the distance measurement process into multiple discrete steps: object region detection using difference images, identification of candidate pixels, parallax calculation for candidate pixels, and statistical filtering. This segmentation allows accurate distance measurement to be achieved through sophisticated processing rather than through increasing baseline length, thereby maintaining compact apparatus size while improving measurement precision.
Solution Approach 2:
The patent performs preliminary object region detection and candidate pixel identification before parallax calculation. By pre-identifying relevant regions and pixels that likely belong to the object, the system focuses computational resources on critical areas, improving measurement accuracy without requiring increased baseline length or larger apparatus size.
2Measurement precision
If the baseline length is increased to distinguish objects from background, then the measurement precision is improved, but the apparatus size increases
Solution Approach 1:
The patent applies local quality by performing object detection and parallax calculation specifically on candidate pixels within identified object regions, rather than processing all pixels uniformly. This localized approach enhances object-background distinction accuracy by focusing computational effort where it is most needed, without requiring increased baseline length or larger apparatus size.
Solution Approach 2:
The patent changes parameters by using statistical methods (median filtering, histogram analysis) to process parallax values and distinguish objects from background. This parameter-based approach enables accurate object-background distinction through data processing rather than through increased baseline length, maintaining compact apparatus size.
3Measurement precision
If multiple viewpoint images are used for distance measurement, then the measurement precision is improved, but the device complexity increases
Solution Approach 1:
The patent extracts only the essential elements needed for distance measurement: object region detection results, candidate pixel identification, and parallax calculations for those candidates. By extracting and processing only relevant data rather than all image data, the system achieves high measurement precision while reducing computational complexity and processing overhead.
Solution Approach 2:
The patent performs partial action by calculating parallax only for candidate pixels within detected object regions, rather than for all pixels in the image. This selective approach maintains high measurement precision while significantly reducing computational complexity compared to full-image processing.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate distance measurement and object tracking in a compact form factor by utilizing parallax calculations and filtering methods to distinguish objects from backgrounds, reducing the need for larger apparatus sizes.
Implementation Method 1
a distance measuring section configured to set a plurality of distance measurement points in the object region detected by the object detection processing section, and generate distance measurement information regarding an object indicated by an image of the object region on the basis of a parallax calculated for each of the plurality of distance measurement points using the plurality of viewpoint images
Data Source
AI summary
An object detection processing section 21 detects an object region, using at least one of multiple viewpoint images (e.g., left-viewpoint image) with different viewpoint positions on the basis of information learned through prior learning. A distance measuring section 23 performs ellipse fitting on the object region detected by the object detection processing section 21 to set a distance measurement point arrangement region, and sets multiple distance measurement points in the distance measurement point arrangement region. The distance measuring section 22 performs a stereo matching process using the multiple viewpoint images to calculate a parallax with sub-pixel accuracy for each of the distance measurement points corresponding to the object indicated by an image of the object region and, on the basis of the calculated parallaxes, generates distance measurement information regarding the object. Thus, the distance measurement information regarding the object can be generated without elongating a baseline distance between imaging sections that acquire the multiple viewpoint images with different viewpoint positions to such an extent that the object can be distinguished from the background.