Vehicle Collision Time Estimation via Edge Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional collision time estimation and alarm systems for vehicles face challenges in accurately calculating collision time due to high computational complexity and errors in edge position detection, particularly when objects move laterally relative to the vehicle, such as during sudden maneuvers like cutting in or merging.
Innovation Solution
The system standardizes edge width, increments count values for detected positions, calculates moving direction and speed based on count value inclinations, and estimates collision time without block matching, allowing for robust output against position detection errors and accurate calculation of collision time for objects with a possibility of collision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If block matching algorithm is used to detect optical flow, then optical flow can be detected, but computational complexity increases and collision time cannot be calculated easily
Solution Approach 1:
The patent extracts only the essential information needed for collision time calculation by detecting edge positions and their movement between frames, rather than performing full block matching on entire image regions. This extraction approach maintains optical flow detection capability while significantly reducing computational complexity.
Solution Approach 2:
The patent segments the image processing task by focusing specifically on edge detection and tracking rather than analyzing entire image blocks. By dividing the problem into edge extraction, position detection, and movement calculation steps, the system achieves optical flow detection with reduced computational burden.
2Measurement precision
If corresponding position is detected before edge speed calculation, then optical flow can be obtained, but positioning accuracy directly influences collision time estimation accuracy
Solution Approach 1:
The patent employs feedback by detecting edge positions in successive frames and using the detected positions to calculate both movement amount and collision time. The system continuously refines collision time estimation by incorporating actual detected edge positions from multiple frames, reducing the direct influence of single-frame positioning errors.
Solution Approach 2:
The patent performs preliminary edge detection and position identification before calculating edge speed and collision time. By预先 identifying edge positions in successive frames, the system prepares accurate reference points that enable more reliable collision time estimation even when objects move laterally.
3Adaptability or versatility
If conventional collision alarm apparatus is configured for objects running in parallel, then collision time can be calculated for overtaking or front-approaching objects, but collision time cannot be calculated for objects moving laterally such as cutting in or merging
Solution Approach 1:
The patent creates a universal collision detection system that handles multiple object movement patterns (parallel, lateral, cutting in, merging) using the same edge detection and tracking methodology. The system calculates collision time based on edge position changes regardless of movement direction, making it adaptable to various collision scenarios without requiring separate detection algorithms.
Solution Approach 2:
The patent uses dynamic edge tracking that adapts to changing object positions and movement patterns. By continuously detecting edge positions in successive frames and calculating movement based on actual position changes, the system can dynamically respond to lateral movements, cutting in, and merging scenarios rather than being constrained to predefined parallel movement patterns.
Data Source
AI summary
Imaging device picks up an image of the area around a vehicle, edge extraction function extracts an edge image from the image picked up by the imaging device, edge width standardization function standardizes an edge width of the edge image extracted by the edge extraction function, counting functions increments a count value corresponding to a position where the edge image standardized by the edge width standardization function is detected, and also initializes a count value corresponding to a position where the standardized edge image is not detected, moving speed detection function calculates a moving direction and moving speed of the edge image extracted by the edge extraction function based on the inclination of the count values, and collision time calculation function calculates the time of collision with an object by utilizing the position and the moving speed of the edge image calculated by the moving speed detection function.