Stereo Parallax Image Selection for Accurate Distance Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image pickup apparatuses struggle to determine the most suitable parallax images for calculating distance due to the lack of a clear selection of exposure conditions between high-sensitivity and low-sensitivity pixels, and HDR images, leading to inaccuracies in parallax amount calculation.
Innovation Solution
An information processing apparatus that acquires and evaluates parallax images with different exposure conditions, estimating reliability through contrast, saturation, and motion blur, and selects the most suitable images for distance calculation based on these factors, using a processor to execute instructions for image processing and distance measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple parallax images with different exposure conditions are acquired, then the availability of images for distance measurement is improved, but the difficulty of selecting the most suitable images for accurate parallax calculation increases
Solution Approach 1:
The system calculates reliability values for each parallax image based on contrast reliability, saturation reliability, and motion blur reliability. This feedback mechanism automatically evaluates and ranks the suitability of each image, resolving the selection difficulty by providing quantitative metrics for comparison.
Solution Approach 2:
The system changes the evaluation parameters by calculating multiple types of reliability (contrast, saturation, motion blur) for each image. This allows systematic comparison across different exposure conditions by transforming the selection problem into a parameter-based ranking system.
2Adaptability or versatility
If parallax images with different exposure conditions are used, then the measurement coverage is improved, but the measurement precision decreases due to uncertainty in selecting the optimal images
Solution Approach 1:
By calculating reliability values as feedback metrics, the system ensures that only high-quality images (those with high contrast, low saturation, and minimal motion blur) are selected for parallax calculation, thereby maintaining measurement precision while utilizing multiple exposure conditions.
Solution Approach 2:
The system introduces reliability parameters (contrast reliability, saturation reliability, motion blur reliability) to quantify image quality. This transforms the selection process into a parameter-driven decision system that objectively determines the most suitable images for accurate parallax measurement.
3Illumination intensity
If images with high sensitivity pixels are used, then the low light performance is improved, but the measurement precision deteriorates due to luminance saturation
Solution Approach 1:
The saturation reliability calculation provides feedback on whether an image suffers from luminance saturation. This allows the system to identify and exclude images where high-sensitivity pixels have saturated, preventing degradation of measurement precision while still utilizing high-sensitivity imaging when appropriate.
Solution Approach 2:
By introducing saturation reliability as an evaluation parameter, the system quantifies the impact of luminance saturation on image quality. This enables objective comparison between high-sensitivity and low-sensitivity images based on actual saturation levels rather than assuming one is always superior.
4Reliability
If images with long exposure time are used, then the signal-to-noise ratio is improved, but the measurement precision deteriorates due to motion blur
Solution Approach 1:
The motion blur reliability calculation provides feedback on the presence of motion blur in each image. This allows the system to identify and exclude images where long exposure has introduced significant motion blur, thereby maintaining measurement precision while still benefiting from longer exposures when the scene is static.
Solution Approach 2:
By introducing motion blur reliability as an evaluation parameter, the system quantifies the impact of motion blur on image quality. This enables objective comparison between images with different exposure times based on actual motion blur levels rather than assuming longer exposure is always superior.
Data Source
AI summary
An information processing apparatus acquires a first parallax image and a second parallax image, which are parallax images that are a group of images having parallax with each other and captured by an imaging unit with different exposure conditions, acquires reliability, which is an accuracy of a parallax amount of the parallax images, for the first parallax image and the second parallax image, and selects the parallax images for generating information about a distance from a plurality of the parallax images including the first parallax image and the second parallax image, based on the reliability.