Image processing in automated pattern classification of oranges.
No hay miniatura disponible
Fecha
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
Descripción
Abstract: A rapid method for automated classification of oranges in living trees by size has been developed. It is based on image processing with correlation analysis in the frequency domain. This technique has the advantage of being a direct measurement method that automatically identifies and counts oranges for quality control. Calibration was performed using standard image with known orange sizes. Orange’s sizes, ranging from less then 2 cm to over 10 cm in diameter have been automatically recognized and successfully measured. Error was not larger than 1.4%. In addition, practical examples of use of the method for determining characteristics of oranges are presented.
Palabras clave
Processamento de imagem, Laranja