Image processing applied to classification of avocado variety hass (Persea Americana Mill) during their ripening process

Nowadays, image processing (IP) methods has been developed rapidly and that has allowed measuring on line of size, shape, color and texture features on food materials. Therefore, the work was undertaken with objective of evaluating the ripening process of avocados variety Hass (Persea Americana Mill) during their storage by image processing. A set of avocados (70) variety Hass of exportation quality was selected. A subset (10) was used to evaluate the maturity stages by means of a computer vision system and Principal Component Analysis (PCA). The rest of avocados were used to determinate ripening parameters: firmness, mass loss and L* a* b* colorimeter values. During ripening process the firmness of avocados were decreasing from 75.43 until 2.63 N and showed a sigmoid behavior and skin color values were constants during 6 days, after that was observed a falling in peel color values (-9.68 to 2.32 a*). Seven image features were extracted from images and during ripening the image features such as color parameters (L* a* b*), entropy (4.29 to 4.00), angular second moment (ASM 0.287 to 0.360) and fractal dimension (2.58 to 2.44) had a similar behavior to ripening parameters. Relationships between image features and ripening parameters were obtained. The parameter a* (from green to red) was digital feature most helpful to establish three stages of maturity with an acceptable percentage of classification. ? 80%Results obtained by means of IP and PCA could be useful to evaluate the ripening process of the avocados.