Traffic Cone Recognition via Boundary Pixel Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting traffic cones using laser radar and millimeter wave radar are inefficient due to low resolution and ineffective echo signals, and visual sensors struggle with fuzzy and discontinuous boundary images, leading to low accuracy in traffic cone identification.
Innovation Solution
A method and device that acquire images of potential traffic cones, perform differential and ternary processing to extract positive and negative boundary pixels, and use Hough transformation to determine straight line segments, matching their positions and angles with known traffic cone information to accurately identify traffic cones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of stationary object
If laser radar is used to detect traffic cones, then detection range is extended, but detection accuracy deteriorates due to low resolution
Solution Approach 1:
The patent segments the boundary detection process into multiple stages: initial boundary pixel extraction, straight line segment fitting, and iterative optimization. This segmentation allows the system to progressively refine detection accuracy from coarse to fine levels, resolving the contradiction between detection range and accuracy.
Solution Approach 2:
The patent implements dynamic iterative optimization where the boundary detection process continuously refines straight line segments through multiple iterations. The system dynamically adjusts detection parameters and re-processes boundary pixels to improve accuracy while maintaining extended detection range capability.
2Reliability
If millimeter wave radar is used to detect traffic cones, then detection capability is improved, but detection effectiveness deteriorates due to ineffective echo signals from plastic materials
Solution Approach 1:
The patent replaces radar-based detection (electromagnetic wave reflection) with vision-based detection (optical imaging). This substitution leverages the visual characteristics of traffic cones (colors, shapes, boundary patterns) rather than relying on echo signals from plastic materials, effectively resolving the information loss problem.
3Ease of operation
If visual sensor is used to capture traffic cone images, then detection is enabled, but identification accuracy deteriorates due to fuzzy and discontinuous boundaries
Solution Approach 1:
The patent performs preliminary boundary pixel extraction and straight line segment fitting before final identification. By pre-processing the image to extract boundary characteristics and fit straight line segments, the system prepares clean, structured data that improves subsequent identification accuracy while maintaining the ease of visual detection.
Solution Approach 2:
The patent creates an idealized geometric model (straight line segments) that copies the essential boundary characteristics of traffic cones. This simplified geometric representation captures the key features needed for identification while eliminating the fuzziness and discontinuities present in raw visual sensor data.
Data Source
AI summary
An image-based road cone recognition method, apparatus, storage medium, and vehicle. Said method comprises: acquiring, during vehicle driving, an image of an object to be recognized; performing differential processing of the image, so as to acquire an image on which the differential processing has been performed, and performing, according to a preset threshold, ternary processing of the image on which the differential processing has been performed, so as to acquire a ternary image comprising forward boundary pixels and negative boundary pixels; acquiring, according to the forward boundary pixels and the negative boundary pixels, a forward straight line segment and a negative straight line segment which represent the trend of the boundaries of the object to be recognized; when position information of the forward and negative straight line segments matches boundary position information of a known road cone, determining that the object to be recognized is a road cone.