Laser Elemental
Classifier

Onteko develops and applies laser-based elemental analysis systems for seed-level discrimination, classification, and quantitative assessment. We convert elemental signatures into actionable biological and industrial decisions.

Seeds for elemental analysis

Elemental Signatures → Biological Decisions

We use micro-laser elemental analysis combined with supervised multivariate modeling to convert seed-level spectral data into actionable classification outputs.

  • Discriminate genotypes within seed lots
  • Classify seeds based on elemental signatures
  • Quantify surface elemental distribution
  • Support research, breeding, and QC decisions
  • Evaluate biological relevance at seed scale
Laser Spectroscopy
Statistical Modeling
Classification Algorithms
Biological Validation

Current Applications

Genotype Discrimination Seed-Lot Differentiation Process Variability Analysis Experimental Classification Ploidy Detection Research Off-Type Seed Detection

Why It Matters

Bulk chemical tests provide averages. Optical inspection evaluates appearance. Neither captures classification at the resolution of the individual seed.

Elemental signatures reveal statistically significant discrimination between classes, within-lot variability, and hidden structure not visible optically — enabling evidence-based go/no-go decisions.

Individual Seed Analysis for Industrial Decision Support

The Laser Elemental Classifier™ performs rapid elemental analysis of individual seeds using a single micro-laser pulse per seed. Each seed generates a characteristic elemental signature, processed by a supervised classification model to assign the seed to a defined class.

No imaging No bulk averaging No indirect surrogates Direct seed-level classification
Seed classification analysis

How It Works in Practice

  • One controlled laser pulse per seed
  • Multi-element spectral acquisition
  • Automated spectral preprocessing
  • Supervised model-based classification
  • Class assignment with confidence score

From Seed-Level Results to QC Decisions

  • Proportion of seeds in each class
  • Detection of off-type or anomalous seeds
  • Lot-to-lot comparison
  • Statistical confidence of classification model
  • Defined go/no-go thresholds

Why Seed-Level Classification Matters

Bulk chemical measurements provide averages. Optical inspection evaluates appearance. Neither method captures classification at the resolution of the individual seed. By operating at seed scale, variability becomes measurable and decision criteria become quantifiable.

1
Pulse / Seed
ms
Acquisition Time
4–6w
Pilot Delivery
300+
Seeds per Pilot

How It Works

From seed intake to classification output — a streamlined, traceable analytical pipeline.

Step 01
🌱
Seed Handling & Traceability
Seeds logged and tracked individually. No bulk pooling or averaging.
Step 02
Single-Pulse Elemental Acquisition
Micro-laser pulse generates transient plasma with multi-element spectral data.
Step 03
📊
ProLIBSpector™ Processing
Signal normalization, feature extraction, and supervised model application.
Step 04
Classification & QC Output
Class label + confidence score. Lot-level go/no-go threshold.
Step 01 — Seed Handling

Individual Traceability

Seeds are logged and tracked under controlled conditions. Each seed is evaluated individually — every classification result corresponds to a specific seed.

Step 02 — Laser Acquisition

Elemental Signature Capture

A precisely controlled micro-laser pulse interacts with the seed surface, generating a transient plasma emission with multi-element spectral information.

  • Single pulse per seed
  • No chemical reagents
  • Minimal preparation
  • Rapid acquisition
Step 03 — Processing

Spectral Processing & Classification

Raw spectral data are processed using ProLIBSpector™, performing signal normalization, feature extraction, supervised model application, and class assignment with confidence scores.

Step 04 — Aggregation

Statistical Aggregation & QC Metrics

  • Class proportions
  • Variability indicators
  • Lot-to-lot comparison metrics
  • Defined decision thresholds

Transforms analytical data into structured decision support.

LIBS Classifier Workflow

Pilot Studies

  • Off-site analytical evaluation
  • Model development and validation
  • Performance report

Long-Term Integration

  • Dedicated laboratory system
  • Workflow integration
  • Model updating
  • Training and support

MicroLIBS-Based Elemental Surface Analysis

The Laser Elemental Classifier™ is based on laser-induced breakdown spectroscopy (LIBS), adapted for seed-scale surface verification. The platform builds upon decades of laser spectroscopy development and classification research.

Multi-element detection
Spatial resolution
Rapid acquisition
Direct elemental signature
Surface-focused analysis
Non-destructive testing

Advanced Laser Analysis Facility

MicroLIBS System
MicroLIBS™ Analysis System
ProLIBSpector HMI
ProLIBSpector™ HMI Software
High-Throughput Automated Laser Analysis System
High-Throughput Automated Laser System
Sample preparation and analysis
Sample Preparation & Analysis

Peer-Reviewed Research Behind Our Technology

Our classification methodology is grounded in over a decade of published laser spectroscopy research across biological, agricultural, and industrial applications.

🎯
Origin Identification
Coffee, Tobacco, Amber
🦠
Disease Detection
HLB in Plants & Insects
🧬
Biological Classification
Varieties & Genotypes
🔬
Tissue Discrimination
Cancer Cells
🦟

Insect Disease Detection

LIBS-based bacterial infection identification in insect vectors of Huanglongbing (HLB) disease.

Scientific Reports 2019

Laser-Induced Breakdown Spectroscopy (LIBS) as a novel technique for detecting bacterial infection in insects

Killiny N., Etxeberria E., Ponce Flores A., Gonzalez Blanco P., Flores Reyes T., Ponce Cabrera L.

🍊

Plant Disease Detection (HLB)

Rapid identification of HLB-infected citrus plants through phloem analysis using LIBS.

Applied Optics 2018

Rapid identification of Huanglongbing-infected citrus plants using laser-induced breakdown spectroscopy of phloem samples

Ponce L., Etxeberria E., Gonzalez P., Ponce A., Flores T.

Coffee Variety Classification

Spectral differentiation between Arabica and Robusta coffee varieties.

Optics and Photonics Journal 2017

Laser-Induced Breakdown Spectroscopy (LIBS) Applied in the Differentiation of Arabica and Robusta Coffee

Diaz Guerrero A.M., Ponce Cabrera L.V., Flores Reyes T., Ortega Izaguirre R.

🚬

Tobacco Origin Classification

Quality control and geographic origin identification of handmade cigars.

Applied Optics 2015

Laser-Induced Breakdown Spectroscopy (LIBS) Quality Control and Origin Identification of Handmade Manufactured Cigars

Alvira F., Bilmes G.M., Flores T., Ponce L.

🔬

Cancer Cell Detection

Fast detection of prostate malignant tissue using multipulsed LIBS.

Revista Cubana de Fisica 2022

Fast Detection of Prostate Malignant Tissue by Multipulsed Laser-Induced Breakdown Spectroscopy (LIBS)

Ponce A., Flores T., Ponce L.

💎

Amber Classification

Analysis and classification of amber samples using LIBS and chemometric methods.

Revista Cubana de Fisica 2023

Analysis of Amber Samples by LIBS and Chemometrics Methods

Flores T., Alvira F.C., Ponce A., Ponce L.

All studies utilized LIBS combined with supervised classification models — the same core methodology powering the Laser Elemental Classifier™ for seed applications.

Initiate a Structured Pilot Evaluation

The Laser Elemental Classifier™ is introduced through a structured pilot engagement designed to evaluate seed-level classification performance using your defined samples. This is a formal technical evaluation — not a demonstration or free trial.

📋 Standard Pilot Scope

  • One defined seed type or species
  • Two to three predefined classification groups
  • Controlled sample size (typically 100–300 seeds)
  • Single-pulse seed-level elemental acquisition
  • Supervised classification model development
  • Cross-validated statistical performance assessment
  • Structured technical review session

📊 Deliverables

  • Classification accuracy metrics (accuracy, precision, recall)
  • Confusion matrix and validation summary
  • Class proportion analysis
  • Defined decision-threshold evaluation
  • Technical interpretation of results
Technical review session — analyzing classification results

Ready to Evaluate?

Discuss scope, feasibility, and objectives with our team. Decision-ready results delivered under controlled conditions.

Request Pilot Discussion
info@onteko.net
Olive Branch, MS, USA
863-521-8998
4–6
Weeks — Sample to Report