Application of X-ray sorting and priming improves the germination performance of low-quality seed fractions of the Papaccella pepper landrace
High seed quality, important for seed storage, germination performance and crop production, is required for vegetable production including pepper (Capsicum annum). Many vegetable landraces such as the yellow Papaccella pepper cultivar (PY) from the Campania region in South Italy, have low seedlot qu...
| Autores: | , , , , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2025 |
| País: | España |
| Institución: | Consejo Superior de Investigaciones Científicas (CSIC) |
| Repositorio: | DIGITAL.CSIC. Repositorio Institucional del CSIC |
| OAI Identifier: | oai:digital.csic.es:10261/403126 |
| Acceso en línea: | http://hdl.handle.net/10261/403126 https://api.elsevier.com/content/abstract/scopus_id/105014822592 |
| Access Level: | acceso abierto |
| Palabra clave: | Convolutional neural network (CNN) models Germination performance Postharvest seed storage Seed technology Vegetable seed quality X-ray imaging |
| Sumario: | High seed quality, important for seed storage, germination performance and crop production, is required for vegetable production including pepper (Capsicum annum). Many vegetable landraces such as the yellow Papaccella pepper cultivar (PY) from the Campania region in South Italy, have low seedlot quality. Using a PY seedlot with poor seed quality we demonstrate that it can be improved to high-quality level by a combination of X-ray imaging, targeted seed priming of identified medium-quality fractions, combined with elimination of the identified low-quality fractions. Non-destructive soft X-ray imaging combined with classical (ImageJ) and Convolutional Neural Network (CNN) deep learning models was used to identify and fractionate the PY pepper seedlot into distinct groups based on the relative proportion of the embryo+endosperm (EE) compartment in relation to the entire seed. Comparative analysis of these morphologically distinct groups for their germination performance and ageing resilience revealed that they represent physiologically distinct quality groups. Group 1 seeds (EE >90 %) were high-quality seeds with excellent performance which did not require any post-harvest treatment. Group 2 seeds (EE 75–90 %) were good quality seeds for which a post-harvest priming treatment improved their germination performance to the level of group 1. Groups 3 (EE<75 %), 4 (EE abnormalities) were low-quality seeds which can't be improved, but can be identified by X-ray imaging and discarded. We propose that the established method and criteria can be transferred easily to seedlots of other horticultural species. |
|---|