Grün HDF-G Series AI Optical Sorting System

Glass recycling presents a sorting challenge that intensifies when moisture enters the equation — damp glass surfaces refract light unpredictably, obscure color signatures, and cling to contaminants, making reliable identification far more difficult than in dry conditions. Mixed container glass, MRF glass, and flat glass streams contain a complex blend of color variants, opaque contaminants, ceramics, stones, and heat-resistant glass across the 4–70 mm particle range, and when damp, these materials become even harder to classify with consistency. For glass recyclers processing both dry and damp feedstocks, the ability to maintain high-precision sorting performance regardless of moisture conditions is the key to producing clean, furnace-ready cullet that meets the strict quality specifications of glass manufacturers and commands premium market pricing. Intelligent vision-based sorting is the key to delivering reliable results across the full spectrum of real-world glass recycling conditions.

The Grün HDF-G Series is an intelligent vision-based sorting system custom-designed for both dry and damp glass recycling. Leveraging high-resolution RGB (HR-RGB) imaging technology powered by proprietary AI algorithms, the system accurately recovers targeted glass from both dry and damp container glass, MRF glass, or flat glass in the particle range of 4–70 mm through visual analysis and impurity detection — evaluating color, shape, brightness, and transparency to enhance recycling purity and material value.

AI-driven surface feature recognition algorithms — trained on real-world production data using deep learning, with sorting precision refined continuously — ensure consistently high sorting accuracy across diverse glass compositions, moisture levels, and contamination profiles — helping processors maximize cullet purity, eliminate costly impurities, and deliver recycled glass that meets the exacting quality standards of downstream manufacturing with unmatched confidence.

Technical Features

● High-Precision Inspection System

Powerful combination of High-Resolution RGB Camera System and optional NIR Camara System accurately detect glass colors (flint, amber, green, etc.) and contaminants (ceramics, stones, porcelain, metals, and plastics).

● Proprietary AI Deep Learning Algorithms

Innovative simulation of manual picking operation by teaching and training the intelligent sorting system all the standards and features of identification and sorting. AI algorithms continuously improve sorting accuracy, with models retrained on new production samples via deep learning.

● Wear-Resistant Belt Design

Wear-resistant conveyor belt ensures stable feeding for both dry and damp glass particles.

● High-Intensity LED Illumination System

Enhanced detection of characteristics and features of cullet materials and contaminants minimizes false rejects.

● Reliable High-Performance Air Ejector System

Powerful high-speed air ejectors designed and produced in-house ensure accurate rejection and reliable performance at high-capacity sorting.

Typical Application Scenarios

✦ Glass Recycling Plants

✦ Bottle-to-Bottle Recyclers

✦ Solar Glass Producers

Specifications