Dimensionality reduction of morphological features of tomato leaves and fruiting habits

Tomato (Solanum lycopersicum L) belongs to the family Solanaceae, which is extensively grown around the world. In Agriculture and Horticulture, the genetic purity of cultivars is critical to farmers, plant breeders, seed producers and as well as regulatory agencies. The genetic and morphological shape based features are used to classify different tomato cultivars and species. However, the large variations present in the shapes of tomato leaves and fruits make it complex enough to classify. In this paper we have applied image processing techniques on tomato leaves and fruits to obtain enhanced binarized images which precede the most crucial part of shape based feature extraction. Different morphological features are obtained and analyzed for tomato leaf recognition prototyping model. Finally, features are transformed for dimension reduction to get a better visibility of the features through Principal Component Analysis.