Pitch Correction for Distance Estimation in Automated Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for estimating the distance to a target object, such as in automated driving, face accuracy decreases due to vehicle oscillation around the pitch axis, and current methods for correcting this are either lacking in specificity or burdensome in processing load.
Innovation Solution
An information processing apparatus that acquires time-series images and processes them to generate shifted images by shifting a reference image in the up-down direction by different amounts, allowing for the accurate specification of a correction amount for pitch deviations based on differences between the shifted images and the target image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a technique for calculating and integrating movement amount in a vertical direction of a predetermined region in an image is used to estimate pitch angle, then pitch angle can be estimated, but processing load increases due to the need to specify a region having high correlation between two images
Solution Approach 1:
The patent extracts only the necessary information (image height and pixel position of the imaging unit) from the complex image processing task, avoiding the need to analyze entire images or specify regions of high correlation. This selective extraction maintains pitch estimation accuracy while significantly reducing processing load.
Solution Approach 2:
The patent replaces complex mechanical image processing (comparing entire images or regions to find correlation) with a simplified computational approach using sensor data (accelerometer or GPS). This substitution maintains measurement precision while dramatically reducing processing complexity.
2Reliability
If the entire image processing approach is used to correct pitch deviation, then comprehensive correction is achieved, but processing time and computational resources increase
Solution Approach 1:
The patent extracts only the critical correction parameter (vertical shift amount) needed for pitch compensation, avoiding comprehensive processing of entire images. This selective approach maintains distance estimation reliability while minimizing processing time.
Solution Approach 2:
The patent performs preliminary pitch correction by calculating the vertical shift amount before detailed distance estimation. This preliminary action prevents unnecessary processing time during the main distance calculation phase while ensuring accurate correction is applied.
Data Source
AI summary
An information processing apparatus is provided. The apparatus comprises an image acquisition unit that acquires time-series images by an imaging unit; and a processing unit that performs image processing, generates a plurality of shifted images obtained by shifting a reference image acquired before a target image among the time-series images in an up-down direction by a plurality of types of different shift amounts, and specifies a correction amount by a pitch of the imaging unit on the basis of a difference between each of the plurality of shifted images and the target image.