KETT RN-300 Rice Quality Analyzer

Overview

The KETT RN-300 Rice Quality Analyzer is a dedicated optical inspection system engineered for non-destructive, high-throughput assessment of milled and brown rice grain appearance quality. Unlike conventional moisture analyzers or general-purpose imaging systems, the RN-300 employs a fixed-geometry, calibrated illumination setup combined with a high-resolution color CCD camera and embedded real-time image processing firmware to quantify morphological and chromatic attributes of individual grains. Its core functionality is rooted in computer vision–based classification—not gravimetric or thermal moisture measurement—making it a specialized tool for rice grading, breeding program support, milling process validation, and regulatory compliance in food quality assurance laboratories. The system operates without sample preparation beyond gentle leveling on the measurement tray, delivering objective, operator-independent results traceable to visual standards defined by JIS, ISO 7301, and Codex Alimentarius rice specifications.

Key Features

• Multi-tier classification architecture: Supports three hierarchical modes—Basic (3–4 classes), Standard (6 classes), and Detailed (up to 21 classes for brown rice; 17 for polished rice)—enabling granular differentiation of grain defects including chalkiness (white belly, heart-white, milky-white), discoloration (red rice, insect-stained, fungal blemishes), physical damage (cracks, broken kernels, sprouted grains), and developmental anomalies (immature, dead, malformed kernels).
• Embryo retention quantification: Uniquely measures and reports embryo presence level (0%, 50%, or 100%) per grain and calculates overall embryo retention rate—a critical parameter for nutritional labeling and germination potential assessment in parboiled or specialty rice.
• Geometric and photometric metrology: Computes grain length, width, projected area, and shape factor for each kernel; integrates typicality scoring to rank similarity against reference morphology profiles.
• Dual-mode operation: Optimized optical path and lighting configuration for both brown rice (husk-intact) and polished rice (de-husked), with dedicated optional accessories—including TZ-700 grain aligner, polished-rice tray, and polishing-specific light hood—to ensure consistent focus and contrast across sample types.
• Embedded processing architecture: All image acquisition, segmentation, feature extraction, and classification occur onboard via proprietary firmware—eliminating dependency on external PCs during routine analysis and ensuring deterministic processing latency and data integrity.

Sample Compatibility & Compliance

The RN-300 is validated for use with paddy-derived Oryza sativa cultivars commonly processed in East Asian, Southeast Asian, and North American rice supply chains. It accepts bulk samples of brown rice (unmilled) and, with optional trays and hoods, polished rice (milled to ≥90% degree of milling). Classification logic adheres to Japanese Agricultural Standards (JAS) for rice grading, aligns with ISO 7301:2017 (Rice — Specification) defect definitions, and supports internal QA protocols compliant with HACCP prerequisite programs and ISO/IEC 17025-accredited testing workflows. While not an analytical balance or moisture meter, its output metrics are routinely referenced in GLP-compliant rice varietal release dossiers and mill certification audits.

Software & Data Management

Data output includes tabular CSV exports (grain ID, class label, dimensions, typicality score, embryo status) and annotated JPEG image sequences. The included “KETT High-Quality Print” software enables report generation with configurable templates—supporting bilingual (Japanese/English) labeling, statistical summaries (count, % composition, coefficient of variation), and histogram visualization of size distribution. USB-to-RS-232C adapter enables integration into legacy LIMS environments. Audit trail functionality is limited to timestamped measurement logs; for full 21 CFR Part 11 compliance, integration with validated third-party laboratory information management systems is recommended.

Applications

• Rice breeding programs: Objective phenotyping of chalkiness, translucency, and kernel integrity across F₂/F₃ populations.
• Milling line optimization: Real-time feedback on breakage rates, whiteness uniformity, and embryo loss during polishing stages.
• Import/export inspection labs: Automated verification of contractual grade specifications per national standards (e.g., Thai Rice Standard TIS 279, Vietnamese TCVN 4278).
• Research on post-harvest physiology: Correlation of visual defect patterns (e.g., fissuring, discoloration) with storage conditions, humidity history, and drying parameters.
• Quality control in branded rice packaging: Batch-level consistency monitoring for premium white, brown, germinated, or parboiled rice products.

FAQ

Does the RN-300 measure moisture content?
No. The RN-300 is an optical grain quality analyzer—not a moisture analyzer. It does not employ infrared drying, halogen heating, or loss-on-drying principles. Moisture determination requires a separate instrument such as the KETT PX-300 or PM-300 series.
Can it analyze cooked or parboiled rice?
No. The system is calibrated exclusively for dry, uncooked brown and polished rice kernels. Hydrated or thermally treated samples cause condensation, surface gloss, and dimensional swelling that invalidate classification algorithms.
Is calibration traceable to national standards?
Yes. KETT provides factory calibration using NIST-traceable color targets and geometric reference grids. Annual recalibration services include verification against JAS-certified rice reference sets.
What is the minimum detectable crack width?
Under optimal lighting and focus, the system reliably identifies surface fissures ≥40 µm in width, corresponding to approximately 1.5 pixels at native resolution.
Can raw image data be exported for custom algorithm development?
No. Raw sensor frames are processed in real time and discarded after classification. Only annotated JPEGs and structured CSV reports are exportable.