# US 2026/0287513 A1 · Systems and methods for analyzing and sorting treated seeds

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0287513 A1
- Document ID: US20260287513A1
- Status: Published application
- Kind: A1
- Application number: 19/549,201
- Filed: 2026-02-25
- Published: 2026-09-24
- Earliest claimed: 2025-02-27
- Assignees: Crc R&D, LLC
- Applicants: Crc R&D, LLC
- Inventors: Ronald H. Mains, JR.; Tom Lombard
- Web page: https://patentgrep.com/patent/US20260287513A1

## CPC

- G01N 21/85: Investigating moving fluids or granular solids
- B07C 5/3425: {of granular material, e.g. ore particles, grain (investigation moving fluids or granular solids using optical means G01N21/85)}
- G06T 7/0012: {Biomedical image inspection}
- G06T 7/60: Analysis of geometric attributes
- G06T 7/90: Determination of colour characteristics
- G01N 2021/8592: {Grain or other flowing solid samples}
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)
- B07C: POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING (specially adapted for a specific purpose covered by another class, see the relevant place, e.g. A43D33/06, B23Q7/12)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A process for using image information to identify germination viable and germination non-viable seeds. A classifier algorithm must is initially trained to identify non-viable seeds and viable seeds. Once the classifier algorithm is trained, the classifier algorithm may be implemented to sort viable and non-viable seeds in a semi-autonomous or fully autonomous system.

## Claims

### Claim 1

A computer implemented system for classifying a seed, comprising:
a database for storing seed information and germination information for a plurality of seeds, wherein the seed information and germination information for the plurality of seeds collectively define a plurality of seed training sets;
an imager for imaging individual seeds and recording individual seed information to the database;
a germination input for acquiring germination information for the individual seeds after planting and storing the individual germination information to the database;
a neural networ that receives the plurality of seed training sets as input and output values to generate a trained seed classifier that is configured to determine from an image of a seed if the seed is one of:
germination viable; and
germination non-viable.

### Claim 2 (depends on 1)

The system of claim 1, where the imager further comprises an image analyzer that acquires one or more measurements from individual seed images.

### Claim 3 (depends on 1)

The system of claim 1, where the measurements are at least one of width, height, area, color, hue, intensity, and texture.

### Claim 4 (depends on 1)

The system of claim 1, wherein the imager is a color digital scanner.

### Claim 5 (depends on 1)

The system of claim 1, wherein the germination input is an imager configured to input the germination information based on image analysis.

### Claim 6 (depends on 1)

The system of claim 1, further comprising:
a sorter system having an imager to image seeds, wherein the imager provides seed images to the trained seed classifier, wherein the sorter sorts the seeds based on an output of the trained seed classifier.

### Claim 7 (depends on 1)

The system of claim 1, wherein the neural network is a deep neural network.

### Claim 8

A method for training a seed classifier, comprising:
imaging a plurality of individual seeds;
acquiring image derived seed information for each individual seed and storing the image derived seed information into a neural network training database;
acquiring germination information for each individual seed after the individual seeds are imaged and planted, and storing individual seed germination information to the neural network training database;
using the image derived seed information and the seed germination information for a plurality of individual seeds as input and output values of a neural network to generate a classifier for classifying seeds as at least one of germination viable and germination non-viable.

### Claim 9 (depends on 8)

The method of claim 8, wherein acquiring image derived seed information comprises acquiring one or more of the width, height, area, color, hue, intensity, and texture from individual seed images.

### Claim 10 (depends on 8)

The method of claim 8, wherein acquiring germination information comprises acquiring the germination information from one or more images.

### Claim 11 (depends on 8)

The method of claim 8, wherein the germination information is acquired after the individual seeds are planted for a predetermined time.

### Claim 12

A seed sorting system configured to sort viable seeds from non-viable seeds, comprising:
a seed hopper for holing a plurality of individual seeds;
a conveyor for moving individual seeds from the seed hopper to an inspection area;
an imager for imaging individual seeds as they move through the inspection area and generating an image output for each individual seed;
an image analyzer configured to receive the image output for each individual seed and apply a trained germination classifier to the image and generate a germination output classifying the seed as germination viable and germination non-viable;
an ejector device in communication with the image analyzer, wherein the ejector device is operative to remove an individual seed at is moves through the inspection area upon receiving a germination non-viable classification output from the image analyzer.

### Claim 13 (depends on 12)

The system of claim 12, wherein the seed hopper receives seeds having a first coating having a first color and a second coating having a second color.

### Claim 14 (depends on 13)

The system of claim 13, wherein the image analyzer if further configured to:
identify the individual seed as having the first color or the second color.

### Claim 15 (depends on 14)

The system of claim 14, wherein the image analyzer applies a color specific trained germination classifier the image based on the individual seed being identified as the first color or the second color.

### Claim 16 (depends on 14)

The system of claim 14, wherein the image analyzer is further configured to generate a color output for the seed.

### Claim 17 (depends on 16)

The system of claim 16, wherein the ejector device is operative sort the seed based on both the color output and the germination output.
