3D Measurement Error Correction for Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face challenges in precisely measuring the position and posture of a photographing device and the coordinates of an object from moving images, especially when feature points disappear and reappear, and in distinguishing between appropriate and inappropriate feature points for 3D measurement, leading to reduced measurement precision and reliability.
Innovation Solution
An image processing device and method that includes feature extraction, tracking, stereo image selection, orientation, 3D measurement, error range calculation, and erroneous point deletion sections to identify and remove inappropriate feature points, ensuring high-precision 3D coordinate measurement by selecting stereo images with small error ranges and using a predetermined threshold for appropriateness determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional 2D movement analysis is used to track feature points in moving images, then the tracking process is simple, but measurement precision of 3D coordinates deteriorates due to inability to properly remove erroneous corresponding points
Solution Approach 1:
The patent transitions from 2D image plane analysis to 3D space analysis by calculating actual movement distances of feature points in three-dimensional space. This allows the system to distinguish between pseudo feature points (which show inconsistent 3D movement patterns) and true feature points, thereby improving measurement precision while maintaining reasonable process complexity
Solution Approach 2:
The patent replaces conventional 2D image processing methods with a 3D coordinate-based analysis system. By calculating spatial positions and movement distances in three-dimensional space rather than analyzing two-dimensional image coordinates, the system achieves more accurate identification and removal of erroneous corresponding points
2Productivity
If all extracted feature points are used for 3D measurement, then the measurement process is efficient, but reliability reduces due to inclusion of erroneous corresponding points
Solution Approach 1:
The patent extracts and identifies erroneous corresponding points from the set of all feature points by analyzing their movement consistency in 3D space. These erroneous points (including pseudo feature points and points affected by photographing device sway) are then removed from the measurement dataset, improving reliability while maintaining efficiency by processing only the validated subset of feature points
Solution Approach 2:
The patent implements a feedback mechanism where the movement consistency of feature points is evaluated based on calculated 3D movement distances. Points that deviate from expected movement patterns provide feedback for identification as erroneous, allowing the system to iteratively refine the set of valid feature points used for measurement
3Quantity of substance
If feature points affected by photographing device sway are included in measurement, then the number of usable feature points increases, but manufacturing precision of coordinate measurement deteriorates
Solution Approach 1:
The patent replaces 2D image coordinate analysis with 3D spatial movement analysis to identify and exclude feature points affected by photographing device sway. By calculating actual three-dimensional movement distances and comparing them against expected values, the system can distinguish between valid feature points and those corrupted by device motion, thereby maintaining measurement precision while utilizing the maximum number of reliable feature points
Data Source
AI summary
An image processing device is provided that can precisely measure the photographing position or posture of a photographing device or the coordinates of an object based on sequentially changing photographed images. A series of sequentially photographed images are acquired, from which feature points are extracted. The feature points are tracked and correlated to each other. Stereo images are selected from the series of photographed images correlated, and subjected to an orientation and a 3D measurement. The error range of corresponding points obtained by the 3D measurement is calculated. Based on the calculated error range, it is determined whether or not the corresponding points are appropriate for 3D measurement. Those corresponding points determined as inappropriate are deleted. Another orientation and another 3D measurement are performed using those corresponding points excluding the deleted ones, thus improving the measurement precision.