Vineyard dataset with anotation of leaf density - for precision spraying

PRYSM is a project from CRIIS - INESC TEC, which focuses on developing robotics solutions for vineyard smart and precision spraying. The main objective is to develop a smart and precision electrical sprayer that allows better management of pesticide use. This activity has a lower impact on the environment and represents a lower expenditure of resources in agriculture. Therefore, this project requires a reliable computer vision system to classify the leaf density in the vineyards. The current dataset consists of 475 images of size 640x480, and it was collected using a stereo camera assembled in a robotic platform. Every image was manually annotated using an application that defines three regions of interest and then asks for the leaf density on each of these regions. The value is inserted in the four classes: 0% leaves, 33% leaves, 66% leaves, or 100% leaves. This value is then saved in a text file, together with the name of the image. This way, to use this dataset, it is necessary to first consider the name of the image in each line of the text file and then three values corresponding to the leaf density of the three regions of interest. This dataset is for those who want to calculate the foliar density of the vines in a central area of the image and three different regions of interest.