Product Overview

Redistribute Labor & Enhance Potato Grading Accuracy with Artificial Intelligence

Benefits

  • Labor savings: Reduce and redistribute labor where needed at your facility.
  • Reduced chargebacks: Ensure your customers receive potatoes that meet their quality and sizing criteria.
  • Unmatched accuracy: Removes subjectivity from sorting and grading procedures.
  • Improved throughput: Allows processors to operate at higher capacities.
  • Superior process control: Real-time analysis and record keeping aid in business decisions and vetting of suppliers

Features

  • Discard unacceptable potatoes at rates up to 80-100 picks per minute (with two-robot configurations).
  • Programmed with robust AI models to automatically classify products for user-defined traits.
  • Robust, intuitive software.

Unmatched Defect Detection

Most vision systems can’t handle products with mass variations.​ SiftAI's advanced technology accounts for natural variations in your products with unrivalled precision, effectively discerning between defected and viable products. Defects in potatoes, for example, that we can find include:

  • Green
  • Growth Crack
  • Misshapen (banana, heart, knob, pear, and other)
  • Mud​​ ​
  • Sprouts​​​​​​​​​​​​
  • New Bruise
  • Old Bruise
  • Pressure Bruise
  • Rhizoct
  • ​Rot​
  • ​Frozen
  • And more...
No items found.
No items found.
No items found.
No items found.

How It Works

AI-Powered Vision Inspection & Automation

The SiftAI Robotic Sorter is a first-of-its-kind potato grader that combines an AI-based vision inspection system with a delta robot. Each system is programmed with AI models for overall potato size and shape or presence of defects like bruises, cracks, %green, and other cosmetic features. Should any potato fall outside of the AI model’s acceptance criteria, the system triggers the robotic arm to pick up and remove the potato from the process stream.


Proven AI Models to Ensure System Accuracy

Over time, the SiftAI Robotic Sorter “learns” to focus on products on the conveyor and detect visible defects across the entire product, while also applying AI models to determine whether the product meets the specified size. Each user can adjust defect and size tolerances instantly from the SiftAI Robotic Sorter Software.

1) The SiftAI® camera takes a picture of product on a rolling table.

2) The AI system separates the product from the rolling table.

3) The AI then scans the product for presence of defects based on the AI Model criteria. From there, it automatically calculates the defect percentage and determines next steps (let the product proceed or reject it from the product flow). 

End-to-end Quality Assurance

Command Quality & Food Safety Throughout Your Process

From pre-sorting to downstream sorting, sizing, and grading, through final product inspection, SiftAI® offers comprehensive and automated quality control that allows processors to achieve higher throughput with superior accuracy and reduced labor.

Intuitive, Powerful Software

The user interface is intuitive, and operators can monitor production lines in real time and provide analysis or health check reports.

Users can see a live feed summarizing the performance of each robotic system.
(Left): Adjust defect or product size tolerances of the AI model on-the-fly. (Right): Analyze rejected products down to their finest detail.
Review and export performance reports.

Specifications

Notes
Maximum Belt Width CoverageUp to 24 in.Up to 36 in.
Number of SiftAI Cameras NeededOne Camera: Belt width up to 18 in. Two Cameras: Belt width up to 24 in.
Picks per RobotOne-Robot Systems: 40-60 picks per minute Two-Robot Systems: 60-90 picks per minute

Related Resources

No items found.