Image processing and biometrical investigations for choosing the most ranked rice cultivars

Rice ( Oryza sativa L.) is one of the vital food staples in our daily life, amongst the oldest cultivated crops worldwide, ranks as the most widely grown crop. It serves as a vital nutritional material to more than half of the world increasing crowd. This experiment was conducted for evaluation and selection of the most ranked rice cultivars: Kamfirooz, Yasouj, Dom Siah, Gharib, Dollar, Hassan saraei, 304, Lenjan, and Musa Tarom (MTA) as the well-known and prevalent varieties in the 'Fars' and 'Kogilouyeh-o-BoyreAhmad' provinces, Iran, along with other countries. Rough rice grains of the varieties were randomly selected and their principal dimensions were imaged using camera and a special box for light controlling and including Perimeter, Area, MajorAxis Length, MinorAxis Length, Solidity, Eccentricity, and Equiv Diameter, following with image analyzing by image processing software, in a labratroy experiment. The data were statistically analyzed and graphically plotted using SPSS v.17 and MINITAB v.16 software programs. The results of the image analysis indicated that there were significantly differences between most cultivars in the case of physical traits. The cluster categorization of the cultivars showed also that 9 cultivars clustered in 5 groups, i.e. Yasouj and Lenjan cultivars were located in different groups individually, where cultivars: Kamfirrooz, Dom Siah and Dollar in a unique cluster, but Gharib and 304 in the 4th group and MTA and Hassan Saraei in the 5th group. The results of this experiment showed the separateness of rice most recent cultivars according to agronomic and physical status, and leads to picking the most valuable cultivars for upcoming food nutritional values and also experiments