NIR Feedstuff Identification Using Ranked Spectral Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying feedstuff and feedstuff raw materials using near infrared spectroscopy are prone to errors due to incorrect classification, requiring costly lab equipment, expertise, and are susceptible to false positive determinations, especially when dealing with heterogeneous samples.
Innovation Solution
A computer-implemented method involving near infrared spectroscopy and similarity analysis, where a query vector is compared to a database of known spectra, with similarity values ranked and weighted by position to accurately predict the feedstuff or feedstuff raw material, using measures like Cosine coefficient and Euclidean distance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If qualitative analysis methods are used to precisely identify feedstuff identity and origin, then measurement precision is improved, but device complexity and cost increase due to requiring costly lab equipment
Solution Approach 1:
The patent uses near-infrared spectroscopy to create spectral copies or fingerprints of feedstuff materials. Instead of using complex lab equipment for direct analysis, the system captures spectral information that serves as a copy of the material's identity characteristics, which can then be compared against reference databases for identification.
Solution Approach 2:
The patent replaces complex mechanical/lab-based qualitative analysis equipment with an optical system (near-infrared spectrometer). This substitution uses light interaction with the feedstuff to obtain identification information, eliminating the need for expensive laboratory instruments while maintaining identification capability.
2Measurement precision
If traditional qualitative analysis methods are used, then identification accuracy is improved, but loss of time increases due to high standards for time required and expertise
Solution Approach 1:
The system enables self-service identification where the feedstuff's own spectral characteristics are used to identify it. The near-infrared spectrometer automatically captures the spectral fingerprint, and the system autonomously compares it against the reference database to provide identification, eliminating the need for expert operators and reducing analysis time.
Solution Approach 2:
The patent pre-establishes a database of reference spectral information for various feedstuff materials before actual identification is needed. This preliminary action of building the reference library enables rapid comparison and identification during operation, eliminating the need for time-consuming expert analysis during the actual identification process.
3Device complexity
If near infrared spectroscopy is used for routine identification, then device complexity is reduced, but reliability decreases due to human mistakes in selection and classification
Solution Approach 1:
The system incorporates feedback through automated spectral comparison. The near-infrared spectrum obtained from the feedstuff is automatically compared against the reference database, and the system provides feedback in the form of identification results. This closed-loop approach eliminates human selection errors and ensures consistent, reliable classification.
Solution Approach 2:
The patent introduces an intermediary computational system that mediates between the simple near-infrared measurement and the final identification result. This intermediary layer automatically processes the spectral data, compares it with reference information, and determines the feedstuff identity, removing human operators from the classification process and eliminating their errors.
4Measurement precision
If similarity analysis with reference spectra is performed, then measurement precision is improved, but device complexity increases due to requiring database systems and computational processing
Solution Approach 1:
The patent creates a universal reference database that serves multiple identification purposes for different feedstuff materials. This single database system handles the spectral comparison for various materials, providing multi-functional capability that justifies the computational complexity by enabling accurate identification across different feedstuff types through a unified approach.
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 method provides precise and reliable identification of feedstuff and feedstuff raw materials, reducing human error and improving prediction accuracy by considering multiple similarity values and their rankings.
Implementation Method 1
providing a near infrared (NIR) spectrum of a sample of an unknown feedstuff raw material and/or feedstuff
Implementation Method 2
transforming absorption intensities of wavelengths or wavenumbers in the spectrum
Data Source
AI summary
A computer-implemented method for predicting a feedstuff and/or feedstuff raw material is described. The method comprises providing a near infrared (NIR) spectrum of a sample of an unknown feedstuff raw material and/or feedstuff. The absorption intensities of wavelengths or wavenumbers in the spectrum are transformed to give a query vector. A set of database vectors of a population of spectra of known feedstuff raw materials and/or feedstuffs is also provided, and these comprise at least 50 spectra of samples of each feedstuff and/or feedstuff raw material from each of its global growing areas. The similarity between the query vector and each of database vectors is analyzed to produce a score, and the feedstuff raw material and/or feedstuff of the database vector with the highest score is assigned to the sample.