Multispectral Imaging System Using Stacked Photodiodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multispectral surveillance imaging systems face challenges in spatial and temporal registration, leading to loss of processing gain and reduced ability to detect low contrast objects in ocean imagery due to mis-registration and atmospheric and surface clutter components.
Innovation Solution
A multispectral littoral surveillance system using a camera with a vertically stacked array of three photodiodes for RGB output separation, combined with a polarizing filter to reduce clutter, and a weighted difference process between spectral components to enhance object detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a color mosaic is affixed to the sensor in a single camera, then the camera can capture multiple color channels, but spatial mis-registration occurs between subpixels
Solution Approach 1:
The patent transitions from a 2D color mosaic arrangement to a 3D stacked photodiode architecture. Multiple photodiodes are stacked vertically at the same spatial location, allowing each to capture different color channels while maintaining perfect spatial alignment. This dimensional transition from planar to volumetric arrangement eliminates the spatial mis-registration problem inherent in 2D mosaics.
2Adaptability or versatility
If multiple cameras are used for multispectral imaging, then more color channels can be captured, but mechanical alignment and image registration become difficult to maintain
Solution Approach 1:
The patent merges multiple color channel capture functions into a single camera by stacking multiple photodiodes within the same sensor housing. This integration eliminates the need for multiple separate cameras and their associated mechanical alignment systems, reducing device complexity while maintaining multispectral imaging capability.
3Measurement precision
If dichroic filters and prisms are used to separate light components, then spectral separation is achieved, but polarization effects cause loss of processing gain
Solution Approach 1:
The patent extracts the spectral separation function from optical elements (dichroic filters and prisms) and implements it directly at the photodiode level through wavelength-selective absorption. By removing the intermediate optical components that cause polarization, the system achieves spectral separation without the associated energy losses.
4Measurement precision
If subpixels are not sampled simultaneously, then temporal mis-registration occurs, but simultaneous sampling requires precise timing control
Solution Approach 1:
The patent combines multiple photodiodes into a single integrated sensor unit where all photodiodes are exposed to light simultaneously and read out in parallel. This unified sampling architecture eliminates temporal mis-registration without requiring complex timing control mechanisms, as all color channels are captured at the exact same moment.
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 effectively eliminates spatial and temporal mis-registration, reduces atmospheric and surface clutter, and amplifies object contrast by up to 30 times, improving detection capabilities and processing gain.
Implementation Method 1
The sensor is a new type of sensor that uses an array of three vertically stacked photodiodes per pixel that produce a Red, Green and Blue (RGB) output separation due to the spectral absorption of silicon as a function of depth
Implementation Method 2
combined with a polarizing filter to reduce clutter
Data Source
AI summary
The multispectral imaging system provides detection of low contrast objects on or below the surface of a body of water through the elimination of most of the surface reflected light clutter components from multispectral images. The images are processed by taking the weighted difference of two or more spectral components and applying a demeaning or whitening filter to the result. The images can then be frame averaged where the appropriate corrections for motion, magnification, rotation, and translation have been applied. The resultant imagery can then be displayed to an operator or transmitted to a remote location.