Estimating Volume of the Tomato Fruit by 3D Reconstruction Technique

Abstract

Due to its nutritional and economic value, the tomato is considered one of the main vegetables in terms of production and consumption in the world. For this reason, an important study is the volume of the fruit related to the loss of mass. This process develops in the fruit mainly after the harvest. This parameter affects the economic value of the crop, the scientific community has been progressively addressing the problem. However, there is no non-contact method that allows volume estimation. In this work a quantitative method is developed that allows to estimate the volume of the tomato fruit by means of a 3D reconstruction technique. The method is based on a camera that acquires different views of the silhouette of the fruit and by means of image processing a 3D point cloud is generated and later a Delaunay triangulation technique, which allows the estimation of the 3D volume of the tomato fruit. This volume estimation allows the design of precise strategies focused on the prediction of tomato post-harvest by producers, distributors and consumers.