LUCAi™ Smart sorting and grading for apples

Deep learning for precision apple sorting and grading
  • UniqueSellingPoint1
  • UniqueSellingPoint2
  • UniqueSellingPoint3

LUCAi™ for Apples

Deep learning technology for optical sorting and grading for unparalleled precision across the apple industry 

Provide your operators with unparalleled precision and the confidence that they'll get the right product in the right pack, season after season

Maximize your crop yields and minimize fruit loss like never before with the game-changing Spectrim with LUCAi® for apples, powered by TOMRA’s groundbreaking LUCAi® deep learning technology. Correctly classify those traditionally hard to detect defects with confidence, and empower your operator to control your operations season after season.

Spectrim with LUCAi™ is a game changer. We had to work a lot harder around the stem area mainly for splits and punches. Now you don' have to do the work anymore, LUCAi™ does it for you.

How LUCAi™ reduces the work required for stem splits and stem bowls

Traditional apple grading with machine learning inspects small sections of the fruit — a few pixels at a time — and predicts based on limited, local information. This can lead to confusion when interpreting complex textures, especially around the stem and stem bowl.

LUCAi™ is different. As a full deep learning classification model, it mimics the human brain’s visual cortex — the part that understands what the eyes see. Instead of making isolated guesses, LUCAi™ views the entire apple holistically.

By assessing the whole fruit, LUCAi™ accurately recognizes natural features like the stem and stem bowl without mistakenly identifying them as defects. It sees these areas in context, avoiding false positives caused by normal textural variations.

By analyzing up to 40,000 images a minute provided by Spectrim, LUCAi™ then assigns:

  • Defect class/category: Identifying specific defect types without misclassifying natural features
  • Confidence: How certain LUCAi™ is that the image belongs to the assigned category
  • Severity rating: A 0–100 score measuring how significant the defect is — allowing your operators to stay fully in control of the final sort

LUCAi™ benefits apple growers packers

Superior defect detection

Detects and eliminates over 95% of apple stem-related imperfections. Correctly classify those previously hard to detect defects, including stem splits.

User-friendly interface

Our simple user interface ensures operators of all levels will maintain complete control of each model and maximize the value of each and every apple.

Premium performance

Prevent accidental downgrades and eliminates manual grading delays. Reduce the number of good-in-bad and bad-in-good to maximize your pack out.

Effortlessly consistent

Reap the commercial benefits of performance consistency across seasons, batches, varieties, and operators. 

Customer Story

FreshCo, New Zealand

The Fresh Fruit Company of New Zealand (Freshco) is a leading fruit and vegetable exporter based in Auckland.

As expert apple packers, Freshco looked to TOMRA Food to optimize their process by increasing the efficiency and accuracy of their packing operation. These are achieved by both superior performance powered by deep learning and the ease of use of the system.

"The  new  UI  that  I  have on Spectrim made  it  so  easy  to  run.  It's  really  beneficial  for  tracking  your  grades and  just  overall  very easy to use. With the UI, you  can  see  more  data.  When  you  make  changes,  you  can  see  the  percentage of that  change  that  you're  making. That makes it really user-friendly.  The  AI  return  on  investment will  be  probably  achieved  within  two  years,  if  not  sooner."

 

Video poster

Talk to a TOMRA Spectrim with LUCAi™ expert today

Operating expenses are on the rise, and every efficiency and cost-cutting advancement has a significant impact. Let the Spectrim with LUCAi™ enhance your operations. Connect with us by using the form below to discover how you can make Every Resource Count.™

<script charset="utf-8" type="text/javascript" src="//js.hsforms.net/forms/embed/v2.js"></script> <script> hbspt.forms.create({ region: "na1", portalId: "1607495", formId: "2c80cde1-eb31-4e71-b6a6-6cbee755a317" }); </script> <style type="text/css"> :root { --main-text-color: #4c4c4c; --input-text-color: #757575; --main-font-bold: "proximanova-bold"; --main-font-semibold: "proximanova-semibold"; --main-font-medium: "proximanova-medium"; --main-font-regular: "proximanova-regular"; --main-border-color: #8f9ea9; } .component.type-plainHtml { padding-top: 0; padding-bottom: 0; } .hbspt-form { box-shadow: 4px 8px 32px rgba(69, 93, 111, 0.2); width: 100%; overflow: hidden; padding: 40px 20px; color: var(--main-text-color); font-size: 1.6rem; font-family: var(--main-font-regular); line-height: 2.4rem; display: inline-block; max-width: 100%; margin-bottom: 1.6rem; } @media (min-width: 1024px) { .hbspt-form { font-size: 1.8rem; line-height: 2.8rem; } } /*** labels **/ .hbspt-form label { margin-bottom: 1.1rem; display: inline-block; font-family: var(--main-font-semibold); } /*** fieldsets ***/ .hbspt-form fieldset.form-columns-1 { margin-bottom: 2rem; max-width: 580px; } .hbspt-form fieldset.form-columns-2 { max-width: 580px; } .hbspt-form fieldset.form-columns-2 > .hs-form-field { margin-bottom: 2rem; } /*** select and input ***/ .hbspt-form select, .hbspt-form .hs-input { cursor: pointer; color: var(--main-text-color); font-family: var(--main-font-medium); padding: 1.6rem 2.4rem 1.9rem 2.4rem; border: 1px solid var(--main-border-color); font-size: 1.6rem; line-height: 2.4rem; width: 100%; } .hbspt-form .hs-input { color: var(--input-text-color); } @media (min-width: 1024px) { .hbspt-form select, .hbspt-form .hs-input { font-size: 1.8rem; line-height: 2.8rem; } } .hbspt-form select:focus, .hbspt-form .hs-input:focus { outline: 1px solid #007cad; border: 1px solid #007cad; } .hbspt-form fieldset.form-columns-1 select, .hbspt-form fieldset.form-columns-1 .hs-input { width: 100%; } .hbspt-form fieldset.form-columns-1 .input { margin-right: 0; } .hbspt-form select { appearance: none; padding: 1.6rem 6.2rem 1.9rem 2.4rem; } .hbspt-form .hs-fieldtype-select .input { position: relative; } .hbspt-form .hs-fieldtype-select .input:before { content: ""; pointer-events: none; position: absolute; background-image: url("data:image/svg+xml,%0A%3Csvg xmlns='http://www.w3.org/2000/svg' width='14.828' height='8.414'%3E%3Cpath data-name='Path 8138' d='M1.414 1.414l6 6 6-6' fill='none' stroke='%234c4c4c' stroke-linecap='round' stroke-linejoin='round' stroke-width='2'/%3E%3C/svg%3E"); width: 1.2rem; height: 1rem; right: 20px; top: calc(50% - 0.5rem); background-repeat: no-repeat; background-position: center center; } /*** error messages and error on select and input ***/ .hbspt-form select.error, .hbspt-form .hs-input.error { display: block !important; padding: 1.6rem 6.2rem 1.9rem 2.4rem; box-shadow: none; border: 1px solid #ff0033; } .hbspt-form .hs-error-msgs { padding: 5px 0; color: #ff0033; font-size: 1.6rem; line-height: 2.4rem; } .hbspt-form .hs-error-msgs label { margin-bottom: 0; } /*** columns ***/ @media (min-width: 1024px) { .hbspt-form fieldset.form-columns-2 { display: flex; gap: 10px; } } .hbspt-form fieldset.form-columns-2 .input { margin-right: 0px; } .hbspt-form fieldset.form-columns-2 .hs-form-field { float: none; width: 100%; } .hbspt-form fieldset.form-columns-2 .hs-form-field .hs-input { width: 100% !important; } @media (min-width: 1024px) { .hbspt-form fieldset.form-columns-2 .hs-form-field { float: none; width: 49%; } } /*** checkboxes ***/ .hbspt-form label.hs-form-booleancheckbox-display { display: grid; grid-template-columns: 1.25em auto; gap: 0.5em; font-family: var(--main-font-regular); } .hbspt-form fieldset.form-columns-1 input[type="radio"], .hbspt-form fieldset.form-columns-1 input[type="checkbox"] { width: 20px; height: 20px; accent-color: #003768; margin-top: 2px; } .hbspt-form label.hs-form-booleancheckbox-display > span { display: block; margin-left: 0px; font-family: var(--main-font-regular); } .legal-consent-container > div { margin-bottom: 10px; } /** submit button ***/ .hbspt-form input[type="submit"] { cursor: pointer; text-transform: uppercase; color: #fff; font-family: var(--main-font-bold); background: #e3510d; padding: 1.8rem 2.4rem 2rem; border-radius: 5rem; border: none; font-size: 1.6rem; position: relative; float: right; margin-top: 20px; } .hbspt-form input[type="submit"]:hover { background: #f9dccf; color: #e3510d; } </style>