WIMPLE COUNT

More accurate and efficient than the human eye
An AI-powered waste recognition and counting solution WIMPLE COUNT is an AI-powered solution that accurately identifies and classifies hundreds of recyclable items in real time.
Using advanced object recognition technology, it precisely counts items eligible for recycling deposits
and calculates the corresponding reward amount.

Supports High-Volume,
Multi-Item Input

Unlike traditional methods that accept only a single item or deposit container at a time, this system allows all waste to be input at once, automatically identifying and counting the recyclable items.

AI-Powered
Automatic Recognition

In addition to being trained on both domestic and international recycling separation standards, the system accurately identifies hundreds of recyclable items, including detecting foreign substances and multi-layer packaging materials.

Automatic Counting
and Reward Calculation

WIMPLE COUNT can classify and calculate according to any deposit standard, and even distinguish between alcoholic beverage containers and soft drink containers.

Hardware

WIMPLE COUNT features a streamlined hardware design optimized
for autonomous operation, enabling the device to recognize and
count items automatically as soon as they are inserted into the slot.

Server Rack
600mm X 600mm X 750mm

A product that contains the PC
the most essential component for AI operation.

Inlet
1,080mm X 600mm X 1,200mm

A chain installed at the input area prevents recyclables from piling up and ensures they are evenly distributed when placed inside.

Left and Right Conveyor Belts
1,000mm X 1,200mm X 800mm

Sorts recognized recyclable materials to direct them to either the left or right receiving area.

AI BOX
700mm X 900mm X 1,700mm

A camera is installed inside to photograph recyclables as they move along the conveyor belt, with built-in lighting to ensure a consistent imaging environment.

Straight Conveyor Belt
600mm X 2,000mm X 800mm

It is installed at the base of the AI BOX, it enables recyclables to move at a steady pace synchronized with the AI’s recognition speed.

Smarter Counting with AI

WIMPLE COUNT is equipped with AI that has learned various waste conditions
and regional recycling standards, making it usable anywhere.

Wimple Count meets the complex and varying DRS standards
across Canadian provinces with 99% accuracy

Wimple Count features Seoreu’s advanced AI ‘Wee’,

capable of sorting up to 300 recyclable items per minute
with unmatched accuracy, no matter how complex the classification criteria.

Use Case

Our product, WIMPLE COUNT, is exported to the Canadian company
‘WOORI’ and is utilized as a system for accurately classifying and
counting deposit-eligible items at Canadian Return-It stores.

Listed as an ITU Best Practice

Seoreu's recycling AI use case has been recognized as a best practice
by the International Telecommunication Union (ITU), earning
international acclaim for its innovation and effectiveness.

Who Made This

Our goal is to create a platform

that supports full resource circulation

The volume of waste and the cost of disposal continue to rise globally each year. However, in many countries, proper waste separation and disposal are still lacking, and the actual recycling rate of collected waste remains very low. Relying on human eyes and hands for sorting makes it difficult to improve recycling efficiency. We address this challenge by enhancing traditional sorting and separation processes with an innovative AI-powered automation solution that replaces manual labor.

Publication

The technology from Seoreu integrated into WIMPLE COUNT
has been validated as both reliable and innovative through extensive
research papers and academic publications.

01

Implementation of Recyclable Trash Object Detection Model Based on YOLOv5

2021.11
Haejin Lee, Jonghyuk Lee, Heechul Jung

02

Recyclable Objects Detection via Bounding Box CutMix and Standardized Distance-based IoU

2022.10
Haejin Lee, Heechul Jung

03

Wimplebin: an AI-based recycle bin for a better waste management

2024.12
Jiacang Ho, Jonghyuk Lee, HyoungSuk Kim