Roadside Object Detection Using Height Extrapolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing roadside object detection systems face difficulties in detecting curbstones at distant locations due to challenges in accurately identifying the shorter side surface and parallel parts of curbstones in captured images, leading to reduced detection precision.
Innovation Solution
A roadside object detection apparatus that utilizes a height information detection unit, a height change detection unit, a characteristic point extrapolation unit, and a roadside object detection unit to analyze captured images, detect height changes, extrapolate characteristic points based on a road model, and set a search range for detecting curbstones in far-off regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image recognition methods are used to detect curbstones, then detection is feasible for near curbstones, but detection precision deteriorates for far-off curbstones due to shorter apparent length of side surface and parallel parts
Solution Approach 1:
The invention transitions from 2D image plane detection to 3D spatial detection by introducing height information. Instead of relying solely on the apparent length of curbstone features in the image, the system detects the height of the curbstone from the road surface, which remains relatively constant regardless of distance. This dimensional shift allows accurate detection of far-off curbstones by using vertical dimension information rather than horizontal dimension information.
Solution Approach 2:
The invention introduces height information as an intermediary parameter to bridge the gap between near and far curbstone detection. By detecting the height of the curbstone from the road surface and using this as a key feature, the system can identify curbstones at various distances without relying on the apparent length of side surface or parallel parts, which diminish with distance.
2Area of stationary object
If detection range is extended to far-off regions, then more curbstones can be detected, but detection accuracy decreases due to insufficient characteristic information
Solution Approach 1:
The invention extends detection coverage to far-off regions by utilizing height information, which provides sufficient characteristic information even at large distances. The height of the curbstone from the road surface remains detectable and meaningful regardless of how far the curbstone is from the vehicle, enabling accurate detection across extended areas.
Solution Approach 2:
The system uses the curbstone's own height property as the detection feature, which is an intrinsic characteristic that does not diminish with distance. By relying on the curbstone's self-contained height information rather than external reference features that become smaller with distance, the system maintains detection accuracy across extended coverage areas.
3Measurement precision
If only neighboring region height information is used, then reliable height data can be obtained, but far-off curbstones cannot be directly detected
Solution Approach 1:
The invention performs preliminary detection of height information in the neighboring region where reliable data can be obtained. This preliminary height information is then used as a reference to detect and verify curbstones in far-off regions. By establishing reliable height benchmarks nearby first, the system can then extend detection to distant curbstones using the same height-based criteria.
Solution Approach 2:
The system uses height information from neighboring regions as feedback to guide and validate detection in far-off regions. The reliable height data from nearby curbstones establishes a reference pattern that can be applied to identify and verify curbstones at greater distances, creating a feedback loop that extends detection capability while maintaining accuracy.
Data Source
AI summary
A roadside object detection apparatus to detect a roadside object by analyzing images, detecting height information of an object in the images in a neighboring region having a distance less than or equal to a threshold from a vehicle, detecting a change of a height of a road surface due to the roadside object from the height information in the neighboring region, determining positions having the change of the height detected as characteristic points of the roadside object, extrapolating the characteristic points into a far-off region having a distance greater than or equal to the threshold from the vehicle, based on a road model, setting a search range for the roadside object in the far-off region based on a virtual line that is obtained by the characteristic points being extrapolated by the characteristic point extrapolation unit, and detecting the roadside object in the far-off region based on the search range.