Sensing Crop Geometry and Structure

This chapter provides an introduction to the most common 3D sensing systems used for the extraction of crop geometry and structure information. These systems have been grouped into four families based on their principle mode of operation and performance: photogrammetric techniques, ultrasound sensors, optical sensors and depth cameras. The uses and applications of these systems vary considerably, from the simple indication of the presence or absence of vegetation to the generation of high-density point clouds or 3D models of crops and the ability to provide simultaneous colour and depth data. Throughout this chapter, the operating principles and main applications of all these systems are presented, comparing their advantages and limitations. We concluded that it is not possible to select an optimal sensor for use in the broad field of precision agriculture, but rather that the most suitable device should be chosen for each particular application. The future of crop geometry sensing lies in the development of automated methods to analyse the huge amount of data provided, an area in which the application of artificial intelligence algorithms seems to have promising prospects.