Automatic Counting and Individual Size and Mass Estimation of Olive-Fruits Through Computer Vision Techniques
Fruit grading is an essential post-harvest task in the olive industry, where size-and-mass based fruit classi cation is especially important when processing high-quality table olives.Within this context, this research presents a new methodology aimed at supporting accurate automatic olive-fruit grad...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2019 |
| País: | España |
| Institución: | Universidad de Huelva (UHU) |
| Repositorio: | Arias Montano. Repositorio Institucional de la Universidad de Huelva |
| Idioma: | inglés |
| OAI Identifier: | oai:ariasmontano.uhu.es:10272/16492 |
| Acceso en línea: | http://hdl.handle.net/10272/16492 |
| Access Level: | acceso abierto |
| Palabra clave: | Computer vision Feature modeling Food industry Fruit grading Image analysis Olive |
| Sumario: | Fruit grading is an essential post-harvest task in the olive industry, where size-and-mass based fruit classi cation is especially important when processing high-quality table olives.Within this context, this research presents a new methodology aimed at supporting accurate automatic olive-fruit grading by using computer vision techniques and feature modeling. For its development, a total of 3600 olive-fruits from nine varieties were photographed, stochastically distributing the individuals on the scene, using an ad-hoc designed an imaging chamber. Then, an image analysis algorithm, based on mathematical morphology, was designed to individually segment olives and extract descriptive features to estimate their major and minor axes and their mass. Regarding its accuracy for the individual segmentation of olive-fruits, the algorithm was proven through 117 captures containing 11 606 fruits, producing only six fruit-segmentation mistakes. Furthermore, by linearly correlating the data obtained by image analysis and the corresponding reference measurements, models for estimating the three features were computed. Then, the models were tested on 2700 external validation samples, giving relative errors below 0.80% and 1.05% for the estimation of the major and minor axis length for all varieties, respectively. In the case of estimating olive-fruit mass, the models provided relative errors never exceeding 1.16%. The ability of the developed algorithm to individually segment olives stochastically positioned, along with the lowerror rates of around 1% reported by the estimation models for the three features, makes the methodology a promising alternative to be integrated into a newgeneration of improved and non-invasive olive classi cation machines. The newdeveloped system has been registered in the Spanish Patent and Trademark Of ce with the number P201930297. |
|---|