HPB‑Series Intelligent SWIR Sorter Single‑layer Belt‑type Model Custom‑developed

Single‑layer belt‑type model custom‑developed for whole‑plastic‑bottle sorting. The short‑wave infrared camera recovers PET and other plastic bottles by polymer type, while the AI‑powered RGB camera sorts by color. Deep learning enables precise visual sorting.

Sorting whole plastic bottles is challenging due to the complexity of mixed waste streams. PET, HDPE and other polymers are mixed together with varied colors, labels and inconsistent compositions. Manual sorting or single‑sensor sorting is inefficient and unreliable. For recycling plants running PET whole‑bottle recovery lines and sorting centres handling municipal solid waste, accurate separation of target materials by polymer type and color in a single pass directly determines output purity, downstream recycling efficiency and the commercial value of recovered materials. Intelligent multi‑spectral sorting is the key to converting mixed waste into high‑purity, market‑ready material streams — delivering high efficiency, reliability and scalable processing capacity.

The HPB‑Series is an intelligent multi‑spectral sorting system custom‑designed for recycling plants (PET whole‑bottle recovery lines) and waste‑sorting centres (municipal solid‑waste treatment facilities). Integrating advanced InGaAs Short‑Wave Infrared (SWIR) and high‑resolution RGB (HR‑RGB) imaging technologies, equipped with self‑developed AI algorithms, it simultaneously identifies polymer types and colors within mixed waste streams. High‑precision automatic sorting of PET, HDPE and other plastics can be completed in one single pass.

The deep‑learning‑based AI surface‑defect recognition algorithm is trained on real‑world production data for continuous iterative improvement of visual‑sorting accuracy. It maintains superior sorting accuracy under diverse waste compositions and contaminated conditions. It helps processors maximise material recovery rates, deliver high‑purity material streams and reliably meet the stringent requirements of modern recycling operations.

• Dual‑sensor Synchronous Detection
  • Advanced InGaAs‑SWIR short‑wave infrared camera system accurately identifies polymer types (PET/HDPE/PP, etc.) by analysing unique short‑wave infrared spectral signatures of different polymers.
  • High‑resolution RGB camera system classifies bottles by color differences (transparent / blue / green / contaminated, etc.) with precise color‑discrimination capability.
• Self‑developed AI Deep‑learning Algorithm
  • Visual‑feature recognition automatically classifies bottles, caps, labels and contaminants (metal, silicone), complementing sensor‑based material identification.
  • Deep learning continuously enhances visual‑sorting performance and adapts to new material types and complex forms such as bottles with printed labels.
  • Capable of handling challenging scenarios including reflective bottles, black bottles and flattened bottles, and maintains stable sorting accuracy under variable feed‑in conditions.
• High‑performance Belt‑type Sorting
  • Modular belt system available in 1200 / 2000 / 2800 mm widths, with throughput ranging from 2‑6 t/h. Flexible configuration matches diverse production‑capacity requirements.
  • Three precision nozzle configurations: 88 / 160 / 224 units. Response time ≤1 ms, blow accuracy ±2 mm to guarantee accurate and reliable material separation at high running speeds.
• Higher Output Purity
  • Synergised SWIR polymer identification and RGB color classification produce cleaner material streams and effectively minimise cross‑contamination among PET, HDPE, PP and different color fractions.
• Higher Recovery Rate
  • Precise detection of bottles, caps, labels and contaminants maximises recoverable output while minimising loss of qualified products.
• Reduced Reliance on Manual Sorting
  • Automatic classification of difficult‑to‑sort materials such as reflective bottles, black bottles, flattened bottles and printed‑label bottles cuts labour costs and improves sorting consistency.
• Improved Downstream Processing Efficiency
  • Purified and finely‑sorted material fractions reduce re‑processing steps and equipment wear, supporting production of higher‑quality recycled materials.