Multi-Camera Object Recognition for Occluded Distance Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for autonomous vehicles struggle to accurately acquire distance information when markers are partially or entirely hidden by shielding objects, leading to incorrect or incomplete distance measurements.
Innovation Solution
An information processing apparatus utilizing multiple cameras to capture images, perform object recognition, calculate reliability, determine identical objects across images, and acquire distance information from the most reliable image, employing projective transformation and table information to ensure accurate distance measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single camera is used to capture images for distance measurement, then the device complexity is low, but the measurement precision deteriorates when markers are partially or entirely hidden by shielding objects
Solution Approach 1:
The patent divides the imaging task into multiple segments by using multiple cameras positioned at different locations. Each camera captures images from its own perspective, and the system segments the distance measurement problem into multiple viewpoint analyses. This allows the system to overcome occlusions by selecting images where markers are visible, thereby maintaining measurement precision without requiring a single complex camera system.
2Reliability
If multiple cameras are used to capture images from different perspectives, then the reliability of distance measurement improves when objects are obscured, but the device complexity increases
Solution Approach 1:
The patent merges the data from multiple cameras by identifying corresponding markers across different images and combining the distance measurement results. The system integrates information from multiple perspectives to determine the most reliable distance measurement, thereby improving overall reliability while managing the complexity of the multi-camera system through coordinated data processing.
Solution Approach 2:
The system implements feedback by evaluating the reliability of distance measurements from each camera and selecting or combining results based on this evaluation. When markers are obscured in one camera's view, the system receives feedback about the poor measurement quality and automatically relies more on measurements from cameras with better visibility, thereby maintaining high reliability despite the complexity of managing multiple cameras.
3Measurement precision
If distance information is acquired from a single captured image, then the processing time is short, but the measurement precision deteriorates when the marker is hidden
Solution Approach 1:
The patent applies preliminary action by pre-processing and evaluating the quality of captured images before performing detailed distance measurements. The system identifies images with visible markers in advance and prioritizes these for measurement, avoiding time-consuming processing of images where markers are hidden. This approach maintains measurement precision while minimizing processing time by preparing and selecting suitable images beforehand.
Data Source
AI summary
An information processing apparatus comprising: a unit configured to acquire a first captured image by a first camera and a second captured image by a second camera; a unit configured to recognize an object from each the first captured image and the second captured image; a reliability calculation unit configured to calculate reliability indicating a probability that the object is the object for each the first captured image and the second captured image; a determination unit configured to determine whether an identical object is present in the first captured image and the second captured image; and a distance acquisition unit configured to acquire distance information to the object, based on an object recognition result for a captured image having higher reliability calculated by the reliability calculation unit, in a case where the determination unit determines that the identical object is present.