Physical characteristics of cherry ( Prunus avium ) varieties and shape analysis with the elliptic Fourier method

Knowing the physical properties of biological products is of great importance in terms of developing harvesting technologies, innovating crop processing technologies and determining product drying properties. The aim of this study was to determine the dimensional and shape attributes of two different cherry types (‘10-5’ and ‘10-22’) and six different cherry varieties (‘Ferbolus Verdel’, ‘Lapins’, ‘Merton Late’, ‘Noir de Guben’, ‘Starks Gold’ and ‘Telegal’) and to put forth the differences and similarities of cherry varieties. For dimensional parameters, the greatest averages were observed in ‘10-22’ type and smallest averages were observed in ‘Telegal’ variety. Dimension properties of ‘Merton Late’, ‘Noir de Guben’ and ‘Starks Gold’ varieties were similar. ‘Telegal’ variety had the greatest sphericity value with an average of 89.48%. The sphericity decreased as the shape index increased. Accordingly, the greatest shape index was observed in ‘Starks Gold’ and ‘Merton Late’ varieties. Principal component analysis (PCA) revealed that the first three components explained 98.0% of total variation in dimensional and shape attributes. PC1, PC2 and PC3 explained respectively 67.0%, 19.6% and 11.4% of total variance. PC1 included dimensional parameters, PC2 included sphericity, shape index and shape attributes at vertical orientation, and PC3 included shape attributes at horizontal orientation. Scatter plots were generated by using the factor loads of the first three principal components and dimensional and shape differences between the cherry varieties were clearly observed. In this study, the elliptical Fourier analysis was performed for the closed contour of the cherry varieties and discriminant scores were determined to put forth the differences of cherry varieties. The first three components explained 76.8% of total variation. Cavity or single-sided hill formation at stalk section constituted the greatest source of variance. For shape similarity, elliptic Fourier analysis revealed three groups with a high similarity. The 1st group included ‘Telegal’ and ‘10-22’; the 2nd group included ‘Noir de Guben’ and ‘Lapins’ and the 3rd group included ‘Merton Late’, ‘Ferbolus Verdel’, ‘Starks Gold’ and ‘10-5’ varieties. In this study, the dimensional and shape descriptors of eight different cherry varieties were determined and their physical properties were compared. Furthermore, elliptic Fourier analysis method was successfully used to put forth the differences in shape of cherry varieties.