Hyperspectral Image Reconstruction Error Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current hyperspectral imaging technologies face challenges in accurately estimating reconstruction errors, which affect the quality of hyperspectral images generated by compressed sensing techniques, particularly due to variations in wavelength resolution and randomness of mask data in both space and wavelength directions.
Innovation Solution
A signal processing method that acquires designation information for N wavelength bands, estimates reconstruction errors based on this information and mask data, and outputs a signal indicative of these errors, allowing for improved estimation and display of reconstruction errors to users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If compressed sensing technique is used to acquire hyperspectral images, then the imaging speed and data acquisition efficiency are improved, but the reconstruction error increases due to loss of spectral information
Solution Approach 1:
The patent applies feedback by calculating reconstruction errors for multiple candidate hyperspectral images and using this error information to select the most accurate reconstruction. The system computes reconstruction errors based on the relationship between compressed images and candidate images, then uses this feedback to identify the optimal hyperspectral image that minimizes reconstruction error while maintaining imaging efficiency.
Solution Approach 2:
The patent changes parameters by generating multiple candidate hyperspectral images with different spectral characteristics and evaluating them based on reconstruction error. By varying the spectral parameters of candidate images and selecting those with lowest reconstruction error, the system optimizes both imaging speed and spectral accuracy.
2Measurement precision
If multiple wavelength bands are reconstructed to improve spectral detail, then the measurement precision is improved, but the computational complexity and processing time increase
Solution Approach 1:
The patent segments the problem by dividing the reconstruction process into independent candidate image generations, where each candidate represents a different spectral reconstruction hypothesis. By segmenting the spectral information into multiple candidate versions and evaluating them separately using reconstruction error metrics, the system achieves high spectral accuracy without overwhelming computational complexity.
3Measurement precision
If reconstruction error estimation is performed for all wavelength bands to ensure quality, then the measurement precision is improved, but the loss of time increases due to extensive calculations
Solution Approach 1:
The patent applies partial action by calculating reconstruction errors for a limited set of candidate hyperspectral images rather than exhaustively analyzing all possible wavelength band combinations. This selective approach provides sufficient quality assurance through error estimation while avoiding the excessive computational time that would result from complete analysis of all spectral possibilities.
Data Source
AI summary
A signal processing method executed by a computer includes acquiring designation information for designating N wavelength bands corresponding to N spectral images generated on a basis of a compressed image in which spectral information is compressed, N being an integer greater than or equal to 4; estimating a reconstruction error of each of the N spectral images on the basis of the designation information; and outputting a signal indicative of the reconstruction error.