Search for a command to run...
Background: Cocoa tree seeds (cocoa beans) represent a high-value agricultural commodity with multiple end uses. Compliance with contractual quality specifications and with national and interstate standards underpins their commercial value. At present, cocoa-bean quality assessment remains a complex and labour-intensive procedure performed largely by manual inspection. Moreover, the standard grading approach is destructive, leading to disposal of the tested sample and, consequently, loss of valuable raw material. Purpose: To develop a rapid, informative method for cocoa-bean quality control based on microfocus radiography. Materials and Methods: X-ray images of cocoa beans were obtained using a multifunctional mobile X-ray system of the PRDU family and a portable X-ray unit. Microfocus imaging parameters were: tube voltage 20 kV, tube current 80 μA, exposure time 2 s. The study examined five cocoa varieties (Criollo, Forastero, Trinitario, Nacional, and Sambirano) originating from Colombia, Venezuela, Côte d’Ivoire, Ghana, Honduras, the Philippines, Peru, Ecuador, and Madagascar (2023 harvest). Radiographic projections were visually compared with visible-light images (reference) of longitudinally cut beans prepared from the same samples in accordance with GOST ISO 1114-2014 ( Cocoa beans. Cut test ). Results: The study reports radiographic evaluations of cocoa-bean consignments imported into the Russian Federation from multiple countries. Optimal exposure regimes were established for both the portable device and the laboratory-based system. A relationship was identified between the radiographic appearance of an individual cocoa bean and its quality attributes. Morphological features and defects visualised on radiographs showed a high degree of correspondence with the visible-light images of the respective longitudinal sections. In addition, defect identification based on radiographic image matrices demonstrated a high level of agreement between the portable and stationary imaging platforms. Conclusion: Matrices of visual characteristics of cocoa beans were compiled by origin and defect type; microfocus radiographs and corresponding visible-light photographs were obtained for comparative assessment; and radiographs acquired with a laboratory X-ray system and a portable unit were systematically compared. The findings provide a methodological basis for developing specialised hardware and software tools to enable automated quality control of commercial cocoa-bean batches intended for industrial processing.