We present an Active Learning (AL) strategy for reusing a deep Convolutional Neural Network (CNN)-based object detector on a new data set. This is of particular interest for wildlife conservation: given a set of images acquired with an Unmanned Aerial Vehicle (UAV) and manually labeled ground truth, our goal is to train an animal detector that can be reused for repeated acquisitions, e.g., in follow-up years. Domain shifts between data sets typically prevent such a direct model application. We thus propose to bridge this gap using AL and introduce a new criterion called Transfer Sampling (TS). TS uses Optimal Transport (OT) to find corresponding regions between the source and the target data sets in the space of CNN activations. The CNN scores in the source data set are used to rank the samples according to their likelihood of being animals, and this ranking is transferred to the target data set. Unlike conventional AL criteria that exploit model uncertainty, TS focuses on very confident samples, thus allowing quick retrieval of true positives in the target data set, where positives are typically extremely rare and difficult to find by visual inspection. We extend TS with a new window cropping strategy that further accelerates sample retrieval. Our experiments show that with both strategies combined, less than half a percent of oracle-provided labels are enough to find almost 80% of the animals in challenging sets of UAV images, beating all baselines by a margin.

We address the task of automatically detecting and counting seabirds in unmanned aerial vehicle (UAV) imagery using deep convolutional neural networks (CNNs). Our study area, the coast of West Africa, harbours significant breeding colonies of terns and gulls, which as top predators in the food web function as important bioindicators for the health of the marine ecosystem. Surveys to estimate breeding numbers have hitherto been carried out on foot, which is tedious, imprecise and causes disturbance. By using UAVs and CNNs that allow localizing tens of thousands of birds automatically, we show that all three limitations can be addressed elegantly. As we employ a lightweight CNN architecture and incorporate prior knowledge about the spatial distribution of birds within the colonies, we were able to reduce the number of bird annotations required for CNN training to just 200 examples per class. Our model obtains good accuracy for the most abundant species of royal terns (90% precision at 90% recall), but is less accurate for the rarer Caspian terns and gull species (60% precision at 68% recall, respectively 20% precision at 88% recall), which amounts to around 7% of all individuals present. In sum, our results show that we can detect and classify the majority of 21 000 birds in just 4.5 h, start to finish, as opposed to about 3 weeks of tediously identifying and labelling all birds by hand.

Ecological surveys increasingly rely on large‐scale image datasets, typically terabytes of imagery for a single survey. The ability to collect this volume of data allows surveys of unprecedented scale, at the cost of expansive volumes of photo‐interpretation labour. We present Annotation Interface for Data‐driven Ecology (AIDE), an open‐source web framework designed to alleviate the task of image annotation for ecological surveys. AIDE employs an easy‐to‐use and customisable labelling interface that supports multiple users, database storage and scalability to the cloud and/or multiple machines. Moreover, AIDE closely integrates users and machine learning models into a feedback loop, where user‐provided annotations are employed to re‐train the model, and the latter is applied over unlabelled images to e.g. identify wildlife. These predictions are then presented to the users in optimised order, according to a customisable active learning criterion. AIDE has a number of deep learning models built‐in, but also accepts custom model implementations. Annotation Interface for Data‐driven Ecology has the potential to greatly accelerate annotation tasks for a wide range of researches employing image data. AIDE is open‐source and can be downloaded for free at https://github.com/microsoft/aerial_wildlife_detection.

Many species aggregate in dense colonies. Species-specific spatial patterns provide clues about how colonies are shaped by various (a)biotic factors, including predation, temperature regulation or disease transmission. Using aerial imagery, we examined these patterns in colonies on land of two sympatric seal species: the harbour seal and grey seal. Results show that the density of grey seals on land is twice as high as that of harbour seals. Furthermore, the nearest neighbour distance (NND) of harbour seals (median = 1.06 m) is significantly larger than that of grey seals (median = 0.53 m). Avoidance at small distances (i.e. social distancing) was supported by spatial simulation: when the observed seal locations were shuffled slightly, the frequency of the smallest NNDs (0–25 cm) increased, while the most frequently observed NNDs decreased. As harbour seals are more prone to infectious diseases, we hypothesize that the larger NNDs might be a behavioural response to reduce pathogen transmission. The approach presented here can potentially be used as a practical tool to differentiate between harbour and grey seals in remote sensing applications, particularly in low to medium resolution imagery (e.g. satellite imagery), where morphological characteristics alone are insufficient to differentiate between species.