Automated Seed Sorter Using Multi-Angle Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual seed analysis is inefficient and prone to human error, requiring high throughput and reliability in sorting seeds based on phenotypic traits in the agricultural industry.
Innovation Solution
An automated system that loads seeds into a tray, directs light from multiple angles and spectral wavelengths, collects image data from various seed portions, and analyzes it to sort seeds with desired phenotypes using imaging stations and a controller system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual seed analysis is used, then human judgment can identify phenotypic traits, but the process is tedious, slow, and prone to human error
Solution Approach 1:
The patent replaces the manual mechanical examination system with an automated optical imaging and analysis system. Multiple imaging stations with cameras capture images of seeds from different angles and wavelengths, and computer algorithms automatically analyze the images to identify phenotypic traits such as seed coat color, endosperm color, and seed size, eliminating human manual inspection while maintaining or improving accuracy and dramatically increasing throughput
Solution Approach 2:
The system enables seeds to be automatically analyzed without human intervention. The imaging system captures images, the computer system processes and analyzes the images to identify phenotypic traits, and the system automatically sorts seeds based on the analysis results, creating a self-service automated workflow that replaces manual labor throughout the entire process
2Ease of operation
If manual seed sorting is used, then seeds can be separated based on observed traits, but the process is cumbersome and time-consuming
Solution Approach 1:
The patent replaces manual visual inspection and physical sorting with an automated system that uses multiple imaging stations to capture seed images, computer algorithms to analyze phenotypic traits, and automated mechanisms to sort seeds. This substitution eliminates the cumbersome manual process while dramatically reducing the time required for sorting large numbers of seeds
Solution Approach 2:
The system performs preliminary imaging and analysis of all seeds before sorting. Multiple images are captured from different angles and wavelengths, and phenotypic traits are identified in advance, allowing seeds to be sorted based on pre-analyzed data rather than requiring real-time manual inspection during the sorting process
3Measurement precision
If automated imaging from multiple angles and wavelengths is used, then comprehensive seed analysis is achieved, but system complexity increases
Solution Approach 1:
The patent divides the imaging and analysis system into multiple independent imaging stations, each equipped with cameras and light sources for specific wavelengths. Each station captures images from specific angles, and the computer system processes images from multiple stations to compile comprehensive phenotypic information. This segmentation allows comprehensive analysis while keeping individual system components manageable and modular
Solution Approach 2:
The imaging system uses multi-functional imaging stations that can capture images at multiple wavelengths and from multiple angles using the same basic camera and lighting infrastructure. The computer system is designed to process various types of images and identify different phenotypic traits using unified algorithms, reducing overall system complexity through multi-functionality
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
Enables high-throughput, efficient sorting of seeds with improved accuracy by automating the analysis and separation of seeds with specific traits, reducing human intervention and error.
Implementation Method 1
directing light onto the seeds from at least two directional angles and at a plurality of sequentially changing spectral wavelengths, collecting image data from at least two portions of each seed
Data Source
AI summary
A seed sorter system is operable to sort seeds based on one or more characteristics of the seeds. The system includes a seed loading station operable to isolate individual seeds from a plurality of seeds and load the isolated seeds into a seed tray, an imaging and analysis subsystem operable to collect image data of at least a top portion and a bottom portion of each of the seeds in the seed tray and determine one or more characteristics of each of the seeds, a seed off-load and sort station operable to remove the seeds from the seed tray and sort the seeds to desired receptacles based on the determined one or more characteristics of the seeds, and a seed transport operable to move the seed tray between the seed loading station, the imaging and analysis subsystem, and the seed off-load and sort station.