Autonomous Parking Stop Position Detection Using Inverted Edge Line Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing devices face challenges in accurately determining a proper stop position for autonomous vehicles due to obstacles like fallen leaves obstructing the detection of three-dimensional indication bodies, leading to errors in measuring distances.
Innovation Solution
The image processing device employs an extraction unit to identify horizontal edge lines, a detection unit to detect three-dimensional indication bodies, and a determining unit to accurately determine the stop position by assuming a certain horizontal edge line corresponds to a top end line rather than a bottom end line, even if the bottom end line is obstructed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the detection unit detects the bottom end line of the three-dimensional indication body, then the stop position can be determined, but fallen leaves or obstacles may obstruct the bottom end line causing detection errors
Solution Approach 1:
Instead of detecting the bottom end line of the three-dimensional indication body, the patent inverts the approach by detecting the top end line and calculating the bottom end line position through coordinate transformation and geometric relationships. This inversion avoids the obstruction problem caused by fallen leaves at the bottom end line while still achieving accurate stop position determination.
Solution Approach 2:
The patent uses the top end line as an intermediary element to indirectly determine the bottom end line position. By detecting the top end line (which is not obstructed) and using the known geometry of the three-dimensional indication body, the system calculates the bottom end line position without directly observing it, thus avoiding the harmful effect of obstructions.
2Device complexity
If the determining unit uses the highest horizontal edge line as the top end line, then the detection is simplified, but this may lead to incorrect stop position determination when multiple edge lines are present
Solution Approach 1:
The patent applies local quality by differentiating the treatment of different horizontal edge lines based on their positions and characteristics. Instead of uniformly using the highest edge line, the system identifies which edge line corresponds to the top end line of the three-dimensional indication body by analyzing local geometric relationships and coordinate transformations, thereby improving accuracy without excessive complexity.
Solution Approach 2:
The patent changes the parameter selection criterion from simply using the highest horizontal edge line to selecting the edge line that corresponds to the top end line based on coordinate transformation and geometric analysis. This parameter change allows accurate identification of the correct edge line even when multiple edge lines are present, resolving the contradiction between simplicity and accuracy.
Data Source
AI summary
An image processing device includes: an extraction unit which extracts a horizontal edge line in which edge points are arranged consecutively in a horizontal direction from an image taken that contains an image of a parking frame; a detection unit which detects a three-dimensional indication body to serve as a stop position indication body in the parking frame based on the horizontal edge line extracted by the extraction unit; and a determining unit which, in a case where a plurality of the horizontal edge lines are arranged in a vertical direction in an image of the three-dimensional indication body detected by the detection unit, determines the stop position regarding one of the plurality of the horizontal edge lines other than a highest horizontal edge line in the vertical direction as corresponding to a top end line of the three-dimensional indication body.