Stereo Camera Parallax Correction for Road Step Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing devices for detecting steps on a road surface face challenges in accurate detection due to erroneous parallax measurement, particularly when line segments in stereo images have similar feature amounts, leading to incorrect identification of step locations.
Innovation Solution
The solution involves a feature image generation unit, step candidate extraction unit, line segment candidate extraction unit, analysis unit, and three-dimensional object detection unit that collaborate to analyze and validate step candidates based on their overlap with line segment candidates and inclination, correcting for erroneous parallax measurements to accurately detect road steps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If parallax is calculated by searching for corresponding points while shifting a small region horizontally, then the processing speed is improved, but erroneous parallax measurement occurs when line segments have similar feature amounts
Solution Approach 1:
The system performs collation between step candidates and line segment candidates, using the results to feedback and correct erroneous parallax measurements. The analysis unit compares whether step candidates overlap with line segment candidates and adjusts parallax values accordingly to eliminate measurement errors.
Solution Approach 2:
The system introduces an intermediary collation process that compares step candidates against line segment candidates. This intermediary step acts as a mediator to identify and filter out erroneous parallax measurements by checking for overlaps between the two candidate types.
2Productivity
If step detection is performed without considering line segment candidates, then the detection speed is improved, but false detection of steps increases
Solution Approach 1:
The system extracts line segment candidates in advance before performing final step detection. This preliminary extraction allows the analysis unit to later compare and filter step candidates against these pre-extracted line segments, improving accuracy without significantly impacting overall detection speed.
Solution Approach 2:
The detection process is segmented into separate stages: extracting step candidates, extracting line segment candidates, performing collation between them, and finally determining valid steps. This segmentation allows each component to be optimized independently while maintaining overall efficiency.
Data Source
AI summary
Erroneous detection due to erroneous parallax measurement is suppressed to accurately detect a step present on a road. An in-vehicle environment recognition device 1 includes a processing device that processes a pair of images acquired by a stereo camera unit 100 mounted on a vehicle. The processing device includes a stereo matching unit 200 that measures a parallax of the pair of images and generates a parallax image, a step candidate extraction unit 300 that extracts a step candidate of a road on which the vehicle travels from the parallax image generated by the stereo matching unit 200, a line segment candidate extraction unit 400 that extracts a line segment candidate from the images acquired by the stereo camera unit 100, an analysis unit 500 that performs collation between the step candidate extracted by the step candidate extraction unit 300 and the line segment candidate extracted by the line segment candidate extraction unit 400 and analyzes validity of the step candidate based on the collation result and an inclination of the line segment candidate, and a three-dimensional object detection unit 600 that detects a step present on the road based on the analysis result of the analysis unit 500.