Aerial Image Mosaic Generation Using Selective Feature Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional aerial image systems face challenges in generating high-quality mosaics efficiently due to computational intensity, limited image matching capabilities, and the need for complex motion control systems, especially in unpredictable motion patterns like those encountered with unmanned aerial vehicles (UAVs).
Innovation Solution
A method combining computer vision and photogrammetry to generate aerial image mosaics by establishing optical flow through selective matching of consecutive image pairs using tie points and cross-matching non-consecutive pairs with feature point descriptors, calculating three-dimensional motion and structure parameters, and optimizing these parameters for stable solutions without requiring additional motion control systems like GPS or gyroscopes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image matching methods are used to generate aerial mosaics, then image quality can be maintained, but the computational time and processing complexity increase significantly
Solution Approach 1:
The patent segments the image matching process into distinct stages: feature detection, feature matching, and mosaic generation. By processing images in sequential stages rather than attempting simultaneous full-image matching, the computational complexity is reduced while maintaining matching accuracy through progressive refinement.
Solution Approach 2:
The patent performs preliminary feature detection and extraction before the actual matching process. Key features such as corners, edges, and distinctive patterns are identified and stored in advance, allowing the matching algorithm to work with pre-processed data rather than raw images, significantly reducing computational time during the matching phase.
2Measurement precision
If high-precision cameras and motion control systems (GPS, gyroscopes) are used to improve mosaic accuracy, then measurement precision increases, but device complexity and cost increase
Solution Approach 1:
The patent enables the aerial imaging system to self-calibrate and self-correct for motion disturbances by analyzing the relationship between consecutive images. The system automatically detects and compensates for platform motion, camera orientation changes, and positioning errors through image correlation algorithms, eliminating the need for external motion control systems.
Solution Approach 2:
The patent replaces mechanical motion control systems (GPS receivers, gyroscopes, accelerometers) with a computational approach based on image correlation and feature matching. Instead of using hardware to physically stabilize and position the camera, the system uses software algorithms to correct for motion effects during image processing.
3Measurement precision
If manual feature point selection is used to ensure accurate matching, then measurement precision improves, but ease of operation deteriorates due to time-consuming manual intervention
Solution Approach 1:
The patent implements automatic feature detection and selection algorithms that identify and select optimal feature points without user intervention. The system autonomously detects corners, edges, and distinctive patterns in the images and selects features that maximize matching accuracy, replacing the manual clicking process with automated computer vision techniques.
Solution Approach 2:
The patent replaces the manual mechanical process of clicking feature points with an automated computational process. Image processing algorithms automatically detect and select feature points based on mathematical criteria such as corner detection, edge detection, and pattern recognition, eliminating the need for manual user interaction.
4Measurement precision
If cross-strip matching is performed to improve overall mosaic quality, then measurement precision increases, but computational intensity and processing time increase
Solution Approach 1:
The patent segments the cross-strip matching process by first performing matching within individual strips, then performing a second stage of matching between adjacent strips. This two-stage approach divides the computationally intensive cross-strip problem into smaller, more manageable sub-problems that can be processed sequentially with reduced memory and computational requirements.
Solution Approach 2:
The patent performs preliminary intra-strip matching to establish initial alignment and identify overlapping regions between strips before performing cross-strip matching. This preliminary action provides a foundation that reduces the search space and computational complexity of the subsequent cross-strip matching operation.
Data Source
AI summary
The method according to the invention gene rates an aerial image mosaic viewing a larger area than a single image from a camera can provide using a combination of computer vision and photogrammetry. The aerial image mosaic is based on a set of images acquired from a camera. Selective matching and cross matching of consecutive and non-consecutive images, respectively, are performed and three dimensional motion and structure parameters are calculated and implemented on the model to check if the model is stable. Thereafter the parameters are globally optimised and based on these optimised parameters the serial image mosaic is generated. The set of images may be limited by removing old image data as new images are acquired. The method makes it is possible to establish images in near real time using a system of low complexity and small size, and using only image information.