Spectral Signal Database Construction for Food Nutrient Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional non-destructive food analysis methods rely on pre-constructed databases, limiting their reliability when analyzing new food samples, as they cannot accurately provide precise caloric nutrient information without extensive database construction and are time-consuming, especially in cases like microorganism cultivation.
Innovation Solution
A managing apparatus and method using machine learning to construct a spectral signal database by acquiring and synthesizing caloric nutrient information from standard and reference foods, allowing for high-precision analysis and reliable food information provision through a learning unit that generates spectral signals and updates the database dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If non-destructive ingredient analysis is used with pre-constructed database, then food analysis can be performed quickly and non-destructively, but the reliability is lowered when new inquiry is inputted before database is incompletely constructed
Solution Approach 1:
The system performs preliminary actions by pre-acquiring spectral signals from standard model foods with known caloric nutrient compositions and storing them in a database before actual analysis is needed. This preliminary database construction enables rapid analysis while maintaining reliability, as the system has advance reference data to compare against new food samples.
Solution Approach 2:
The patent introduces standard model foods as intermediaries between the spectral analysis system and actual food samples. These standard model foods with predetermined caloric nutrient compositions serve as reference mediators, allowing the system to reliably analyze unknown foods by comparing their spectral signals against the known spectral patterns of standard models, thus solving the reliability issue with new food inquiries.
2Measurement precision
If chemical food analysis is used to extract sample and measure chemical reaction, then highest accuracy is achieved, but it takes very long time especially in cultivation of microorganisms
Solution Approach 1:
The patent replaces the mechanical/chemical extraction and measurement system with an optical spectral analysis system. Instead of physically extracting food samples and measuring chemical reactions in the lab, the system uses light spectral signals to non-destructively analyze food composition, achieving high accuracy comparable to chemical analysis but without the time-consuming sample preparation and measurement processes.
Solution Approach 2:
The system changes the measurement parameter from chemical reaction measurement to spectral signal analysis. By measuring the interaction between light and food components across different wavelengths, the system obtains compositional information rapidly without requiring the slow chemical extraction and reaction measurement processes, thus reducing analysis time while maintaining precision.
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
This approach enables high-precision, reliable food information analysis and database construction with reduced effort and cost, providing accurate caloric nutrient information for both known and unknown food samples by leveraging machine learning to match spectral signals with caloric nutrient combinations.
Implementation Method 1
a spectrum of reflected light after irradiating food with light of a normal camera (visible area), a near-IR (infrared area)
Data Source
AI summary
The managing apparatus for food information according to one embodiment of the present invention provides database of spectral signals according to caloric nutrient concentrations based on spectral signals of standard model food randomly prepared according to caloric nutrient concentrations.