Grading blueberries is about far more than simply removing defective fruit. Subtle quality differences in color, size, firmness, and dehydration level directly determine the market destination of each berry. The gap between top-tier returns from premium fresh fruit packaging for supermarket shelves and low-value-added applications such as IQF (Individually Quick Frozen) fruit and processing-grade raw materials often hinges on sorting precision. For global blueberry processors serving diverse market channels, sorting is not merely about “picking out bad fruit,” but about “placing every good berry in its most suitable position.“
The Demeter B Series is an intelligent dual-sensor sorting system tailored for this exact scenario. It features 360° inspection and a multi-array (4 or 8 groups) roller sorting structure. By deeply integrating hyperspectral imaging (HSI) with high-resolution RGB (HR-RGB) VIS imaging, and leveraging proprietary AI deep learning algorithms, the system performs real-time, dual non-destructive assessments of both internal and external blueberry quality. Coupled with high-performance air ejectors, it achieves fine classification into up to 16 grades based on indicators such as color, size, softness, and dehydration in a single pass, significantly boosting sorting efficiency while ensuring precision.
Empowered by AI-driven surface feature recognition, the system continuously learns from real production data, steadily improving its detection accuracy. This means the equipment's performance is not static after deployment; instead, it continuously adapts and improves as production data accumulates, delivering stable and reliable sorting performance. This enables blueberry processors consistently deliver batches with extremely low surface defect rates. Meanwhile, hyperspectral imaging (HSI) efficiently identifies hidden defects, such as softness issues and dehydration levels, that are difficult to detect with conventional methods, breaking through the technical limitations of traditional VIS imaging. This intelligent dual-sensor fusion machine-vision-based sorting system helps blueberry processors reduce false rejection rates and labor dependency while maintaining high rejection precision, enabling more reliable and intelligent quality control for premium blueberries.
Technical Features
● Tailored Sensor-Based Grading System for Fresh Blueberry
✦ High-Performance Uniform Distribution
Ensures blueberries enter the sorting system evenly and stably, minimizing inspection interference, and providing consistent input conditions for subsequent high-precision imaging.
✦ Steady-Flow Transmission
Achieves gentle conveying and reliable positioning on customized rollers, maximizing throughput while minimizing mechanical damage and fruit powder loss, and safeguarding the visual quality of fresh berries.
✦ 360° Precision Grading
Performs individual, all-around scanning and analysis of each blueberry for efficient grading based on preset standards, significantly reducing missed detections or misjudgments caused by posture occlusion.
AI Powered Ultra-High-Precision Defect Inspection System
✦ Hyperspectral Imaging
Captures spectral features pixel-by-pixel to efficiently detect hidden defects, such as bruises, rot, unripe fruit, and foreign material that standard industrial cameras struggle to identify, enabling in-depth quality assessment from the surface to the internal structure.
✦ High-Stability Halogen Illumination
Delivers consistent full-spectrum illumination, with an air-cooled design that maintains stable light output during operation and extends lamp lifespan.
✦ High-Resolution RGB (HR-RGB) VIS Imaging
Clearly renders surface characteristics to provide reliable image input for AI algorithms. Complements hyperspectral imaging to build a more comprehensive quality evaluation system.
✦ Proprietary AI Deep Learning Algorithms
Enhances surface feature recognition and supports model retraining with real-world production data, continuously improving the detection accuracy through deep learning.
✦ Ring Shadowless Illumination
Provides uniform, shadowless lighting from all directions to effectively eliminate dark areas caused by fruit contours, ensuring stable and consistent imaging quality, and delivering high-quality image input for AI algorithms.
Based on multi-dimensional features, the system achieves ultra-high-precision inspection for a wide variety of defects and foreign material:
✩ Color grading: Separation by ripeness (dark/light blue/unripe) for uniform appearance
✩ Size grading: Sorting by diameter
✩ Defect detection: Mold, shriveled, cracked, or scarred berries
✩ Soft berry detection: Internal water/sugar analysis
● In-House High-Performance Air Ejector System
✦ Optimized Ejector Valve Design
Balances durability, response speed, and air consumption efficiency, maintaining stable performance even under long-term, high-frequency operation.
✦ High-Speed, High-Precision Ejection
Delivers millisecond-level response and precise material rejection, ensuring optimal sorting accuracy even under high-throughput conditions.
✦ Reduced Operating Costs
Significantly lowers the false reject rate, minimizing material waste from good products being incorrectly ejected, while optimizing air consumption to reduce overall operational expenses.
● 15-Inch Industrial Touchscreen Control System
✦ Intuitive Interaction
User-friendly human-machine interface (HMI) with multi-language support.
✦ One-Click Switching
Quickly recalls sorting parameters for different products.
✦ Comprehensive Remote Support
✩ Remote access and real-time monitoring
✩ Performance data dashboard
✩ Online expert diagnostics
✩ Interactive troubleshooting
