Dual-Swath Imaging System for Seamless Aerial Panorama
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current airborne digital frame camera systems face limitations due to the size of CCD and CMOS area arrays, requiring multiple arrays or cameras to achieve large ground coverage and high resolution, which results in complex and costly processes to correct fan-shaped images into geo-referenced formats, compromising image quality and resource efficiency.
Innovation Solution
A dual-swath photogrammetric imaging system using two 6 cm by 4.5 cm format frame transfer CCD sensors with a nadir shooting twin symmetrical lens configuration, capturing simultaneously overlapping images from each camera to form a seamless panoramic image with uniform resolution, integrated with GPS/IMU for direct geo-referencing and simplified processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple camera systems or multiple arrays are used to achieve large ground coverage and high resolution, then image coverage area and resolution are improved, but device complexity and processing difficulty increase
Solution Approach 1:
The imaging system is divided into two separate camera units, each capturing one half of the scene. This segmentation allows each camera to have simpler optics and processing requirements while collectively providing comprehensive coverage. The left camera captures the left half and the right camera captures the right half, avoiding the need for a single complex multi-sensor system.
Solution Approach 2:
The patent merges the outputs of two separate camera systems into a single panoramic image. By combining the left and right camera images through stitching algorithms, the system achieves large ground coverage without requiring a single overly complex camera array. The merging process integrates information from both cameras to create a unified view.
2Area of stationary object
If oblique cameras are used to capture wide swath images, then ground coverage is improved, but image quality deteriorates due to fan-shaped imaging areas with variable resolution
Solution Approach 1:
The patent employs asymmetric lens configurations where the left and right cameras have different focal lengths and optical characteristics optimized for their respective fields of view. This asymmetry allows each camera to capture its designated half with optimal resolution, avoiding the uniform but compromised resolution of symmetric oblique configurations.
Solution Approach 2:
Each camera is optimized for its specific local field of view, with the left camera tuned for capturing the left half and the right camera for the right half. This local optimization ensures that each region receives dedicated optical tuning, improving overall image quality compared to a single uniform oblique configuration that must compromise for the entire swath.
3Measurement precision
If tilted photographs are corrected to rectangular images for GIS applications, then geo-referencing capability is improved, but image quality is sacrificed through data interpolation and artifact introduction
Solution Approach 1:
The cameras are pre-configured with nadir-pointing optics and precise orientation data, capturing images in the correct geometric orientation from the start. This preliminary correct positioning eliminates the need for post-capture geometric correction, avoiding the quality degradation associated with interpolating and warping tilted images.
Solution Approach 2:
The system creates accurate digital copies of the scene directly in the correct geometric format through nadir-pointing cameras. By capturing the image directly in the desired orientation rather than capturing a tilted image and then creating a corrected version, the system avoids the quality loss inherent in the correction process.
4Area of stationary object
If four lenses and fifteen 35mm format area sensor arrays are used to achieve large format imaging, then image coverage and resolution are improved, but device complexity and failure rate increase
Solution Approach 1:
The patent extracts and uses only the essential components needed for the function - two cameras with appropriate sensors and optics. By removing the excessive number of sensors and lenses from the system, reliability is improved while maintaining adequate imaging capability. The solution uses minimal necessary hardware rather than maximum possible hardware.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system provides large-format, high-resolution images with centimeter-level accuracy, reducing complexity and cost by using only two cameras, enabling quick and precise geo-referencing of images, and simplifying data processing for efficient aerial mapping and surveillance applications.
Implementation Method 1
two 6 cm by 4.5 cm format frame transfer CCD sensors
Implementation Method 2
nadir pointing first optical lens, a first lens shift mount attached between a body of the first camera and the first optical lens
Data Source
AI summary
A portable, aerial, dual-swath photogrammetric imaging system comprising twin nadir pointing CCD cameras for simultaneously acquiring twin adjacent digital images for merging into a large panorama. A pair of lens shifters symmetrically shift twin images to the left and right sides of the focal points of two parallel lenses to extend imaging swathwidth. The twin adjacent images of the imaging system have a strip of narrow overlap at the center of the whole scene that are reserved for photogrammetric processing and stitching twin images into a seamless panorama. Each camera is connected to an embedded computer which controls imaging data acquisition, attaches GPS/IMU measurements, generates KML metadata files for its snapshots, and stores acquired images and metadata into removable SSDs. Direct geo-referenced panoramic digital stills are immediately registered on to Google™ Earth precisely. Its images can be further processed for advanced mapping, change detection and GIS applications.