Towards Detection of Sheep Onboard a UAV

Abstract

In this work we consider the task of detecting sheep onboard an unmanned aerial vehicle (UAV) flying at an altitude of 80 m. At this height, the sheep are relatively small, only about 15 pixels across. Although deep learning strategies have gained enormous popularity in the last decade and are now extensively used for object detection in many fields, state-of-the-art detectors perform poorly in the case of smaller objects. We develop a novel dataset of UAV imagery of sheep and consider a variety of object detectors to determine which is the most suitable for our task in terms of both accuracy and speed. Our findings indicate that a UNet detector using the weighted Hausdorff distance as a loss function during training is an excellent option for detection of sheep onboard a UAV.

Lightweight Sheep Head Detection and Dynamic Counting Method Based on Neural Network

“…In scenarios where the entirety or a substantial portion of livestock appears within a single image frame, the static counting approach proves effective for quantifying the livestock population. In the realm of static counting methods, Sarwar et al [11,12] conducted aerial surveillance using Unmanned Aerial Vehicles (UAVs) to acquire images of sheep at varying altitudes. Employing both single-stage and two-stage target detection networks, they adeptly detected diminutive target sheep and subsequently computed their numbers within the images, thereby facilitating the static counting process.…”