Monocular Camera Distance Estimation via World Coordinate Grid
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Monocular video cameras provide excellent angular resolution but lack distance resolution, making it challenging to determine the distance of an object from a vehicle effectively for applications like collision avoidance and parking assistance.
Innovation Solution
A method involving pre-initialization of a grid of points on an electronic control unit, generating and processing video images to determine the location of a target point in world coordinates using bi-linear interpolation, and calculating the distance and direction from the video camera to the object, incorporating extrinsic camera calibration parameters and a touch screen interface for precise target selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a monocular video camera is used for vehicle applications, then cost and processing requirements are reduced, but distance resolution is lost
Solution Approach 1:
The patent introduces an intermediary coordinate transformation system that maps screen coordinates to world coordinates through a grid of reference points. This intermediary mapping layer enables distance estimation by comparing the position of target objects against pre-established world coordinate references, allowing a monocular camera to provide distance information without requiring stereoscopic vision.
Solution Approach 2:
The patent replaces the mechanical/optical depth sensing mechanism (stereoscopic vision) with a computational approach using coordinate transformation and interpolation algorithms. Instead of using multiple cameras or complex optical systems to perceive depth, the system uses image processing and mathematical mapping to estimate distance from a single camera's 2D image data.
2Measurement precision
If angular resolution is improved in a monocular video camera, then target identification accuracy increases, but distance measurement capability remains insufficient
Solution Approach 1:
The patent transforms 2D image coordinates into a 3D world coordinate system by introducing a mapping grid that relates screen positions to real-world distances. This dimensional transformation allows the system to recover distance information (the third dimension) from 2D image data, effectively adding depth perception capability to the monocular camera system through computational geometry.
Data Source
AI summary
A method and system for determining a distance and direction between a video camera secured on a vehicle and a target point relies on an electronic control unit. The system maps and stores grid points representing a world coordinate grid onto a screen coordinate grid and displays the video image on a display using the screen coordinate grid. The system obtains a target point of an object in the video image and determines a locus of four closest grid points of the screen coordinate grid that encircle the target point. The system determines screen distances from the target point to each of the four grid points and maps the four grid points onto the world coordinate grid. The electronic control unit interpolates the location of the target point in the world coordinate grid as weighted by the screen distances. Using the video camera location in world coordinates and the target point location in world coordinates, the system determines a distance between the video camera and the target point.