Composite Image Selection From Multi-Spectral Original Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing imaging technologies primarily acquire a predetermined spectral image, limiting the ability to obtain needed or desired images.
Innovation Solution
An imaging device that acquires multiple original images, selects one output image from candidate images using machine-learning prediction processing, and optionally composites these images to generate a composite image, with the capability to embed identification information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If only a predetermined spectral image is acquired, then the device complexity is reduced, but the adaptability to obtain needed images deteriorates
Solution Approach 1:
The imaging device segments the image acquisition process into multiple independent spectral image acquisitions at different wavelengths, allowing each spectral component to be processed and selected independently to form the final needed image
Solution Approach 2:
The device implements multi-functionality by enabling the same imaging system to acquire various types of images (spectral images, composite images, selected wavelength images) through software control rather than requiring separate dedicated hardware for each image type
2Manufacturing precision
If multiple original images and composite images are acquired and processed, then the quality and appropriateness of the output image is improved, but the processing time and complexity increase
Solution Approach 1:
The device performs preliminary actions by pre-acquiring multiple spectral images at different wavelengths and pre-processing them into original images and composite images, so that when a needed image is required, the selection can be made quickly from already-prepared candidates
Solution Approach 2:
The selection unit implements self-service by automatically selecting the most appropriate image from candidate images based on predetermined conditions or machine learning algorithms, eliminating the need for manual selection and reducing processing time
3Ease of operation
If automatic selection using machine learning is implemented, then the ease of operation is improved, but the device complexity and computational requirements increase
Solution Approach 1:
The selection unit acts as an intermediary between the multiple acquired images and the final output, using machine learning algorithms as a mediator to automatically determine the most appropriate image based on learned patterns and predetermined conditions
Data Source
AI summary
Obtaining a needed image has been difficult with conventional technology. A needed image can be obtained by an imaging device including an optical signal acquisition unit that performs imaging and acquires an optical signal, an original image acquisition unit that acquire two or more different original images using the optical signal, a selection unit that acquires one output image from candidate images including the two or more original images acquired by the original image acquisition unit, and an image output unit that outputs the output image acquired by the selection unit.