Image Processing Device Weighted Product-Sum Aperture Synthesis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image pickup devices require multiple image takings with different encoded aperture arrays to capture images effectively, limiting their use to stationary objects and increasing complexity and cost due to the need for aperture array switching.
Innovation Solution
An image-processing device performs weighted product-sum calculations between the pupil image of the main lens and encoded aperture patterns, allowing for the selection and interchange of encoded aperture patterns in a single image taking, enabling the use of advanced encoded aperture methods for both stationary and moving objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple image takings are performed with different encoded aperture arrays, then images with multiple encoded apertures can be acquired, but the device complexity increases due to aperture array switching mechanisms
Solution Approach 1:
The patent creates virtual encoded aperture patterns through computational processing of pupil images rather than using physical aperture arrays. By calculating weighted product-sums of pupil images to synthesize different encoded aperture effects, the system achieves multiple aperture patterns without physical switching mechanisms, thereby reducing device complexity while maintaining adaptability
Solution Approach 2:
The patent replaces the mechanical aperture array switching system with a computational image processing system. Instead of physically moving or switching aperture arrays, the system uses digital calculations on pupil images to achieve the same effect, eliminating mechanical complexity while preserving the ability to select different encoded aperture patterns
2Adaptability or versatility
If multiple image takings are performed with different encoded aperture arrays, then images with multiple encoded apertures can be acquired, but the loss of time increases due to multiple capture sequences
Solution Approach 1:
The patent performs preliminary capture of a single pupil image that contains information from which multiple encoded aperture patterns can be derived. By pre-capturing the pupil image and then computationally generating different encoded aperture effects from it, the system eliminates the need for multiple sequential captures, thereby reducing time loss while maintaining pattern versatility
Solution Approach 2:
The patent segments the encoded aperture patterns into computationally derived components from a single pupil image. By dividing the pupil image into regions that can be independently processed and recombined to form different encoded aperture patterns, the system achieves multiple patterns from one capture, reducing time loss while maintaining adaptability
3Adaptability or versatility
If aperture array switching is implemented, then different encoded aperture patterns can be selected, but the ease of operation decreases due to complex switching control
Solution Approach 1:
The patent replaces mechanical aperture switching with computational processing. Users can select different encoded aperture patterns through software controls that process the captured pupil image, eliminating the need for complex mechanical switching mechanisms and their associated control systems, thereby improving ease of operation while maintaining adaptability
Solution Approach 2:
The patent creates virtual copies of encoded aperture patterns through computational processing rather than physical switching. Different aperture patterns are generated as digital processing variants of the single captured pupil image, allowing easy selection and interchange through software without mechanical intervention, thus improving ease of operation
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Light rays having passed through different regions in a pupil plane PP of a main lens 12 and having entered a micro-lens 14(0) of a micro-lens array 14 enter different light receiving cells 16(0, 1) to 16(0, 5), respectively, and are taken in an encoded aperture processing unit 80, as pixel data S(0, 1) to S(0, 5). The pixel data are generated as the pixel data of a weighted product-sum value S0, by a weighted product-sum calculating device U(0). The same process is performed for each micro-lens, and thereby, it is possible to obtain an image equivalent to an image when an encoded aperture pattern corresponding to weight coefficients is disposed on the pupil plane PP.