Plastic Color Sorting | AISORT
Application Overview — Plastic Recycling
Plastic Color Sorting in Modern Recycling Facilities
Color sorting is the most widespread optical sorting application in plastics recycling. While NIR handles polymer identification, RGB (and increasingly hyperspectral) cameras handle color-based decisions: separating clear/light-blue PET from green, amber, and opaque bottles; sorting natural HDPE from colored HDPE; and removing discolored or degraded resin particles from otherwise pure polymer streams. Color purity directly determines the market value of recycled plastic — clear fractions command 30-50% premiums over colored fractions across all polymer types.
Material Characteristics and Sorting Challenges
Plastic color sorting challenges: distinguishing subtle color differences (light blue PET is acceptable in the clear stream at up to 5-10%; dark blue is not); handling printed and labeled surfaces where the print color differs from the base polymer color; sorting translucent and opaque materials that scatter light differently; and maintaining color consistency across batches for buyers who require tight color specifications for their products.
Recommended Sorting Technology Stack
RGB camera (visible spectrum, 400-700nm) — the standard for color sorting. Higher-end systems use hyperspectral or multi-spectral cameras for finer color discrimination. AI vision systems can identify printed labels vs. base polymer color and make sorting decisions based on the underlying material color.
Performance Benchmarks
| Metric | Target |
|---|---|
| Color Separation Accuracy | >98% |
| Clear PET Purity | >99% clear/lt blue |
| Throughput | 5-20 t/h |
| Min Color Difference | ΔE > 2 detectable |
These benchmarks represent achievable performance with modern sensor-based sorting equipment, assuming properly sized, well-maintained equipment operating on representative feedstock. Actual results depend on specific material composition, throughput, and operating conditions.