Fine-grained population mapping from coarse census counts and open geodata.

Fine-grained population maps are needed in several domains, like urban planning, environmental monitoring, public health, and humanitarian operations. Unfortunately, in many countries only aggregate census counts over large spatial units are collected, moreover, these are not always up-to-date. We present POMELO, a deep learning model that employs coarse census counts and open geodata to estimate fine-grained population maps with [Formula: see text]m ground sampling distance. Moreover, the model can also estimate population numbers when no census counts at all are available, by generalizing across countries. In a series of experiments for several countries in sub-Saharan Africa, the maps produced with POMELO are in good agreement with the most detailed available reference counts: disaggregation of coarse census counts reaches [Formula: see text] values of 85-89%; unconstrained prediction in the absence of any counts reaches 48-69%.

Perspectives in machine learning for wildlife conservation.

Inexpensive and accessible sensors are accelerating data acquisition in animal ecology. These technologies hold great potential for large-scale ecological understanding, but are limited by current processing approaches which inefficiently distill data into relevant information. We argue that animal ecologists can capitalize on large datasets generated by modern sensors by combining machine learning approaches with domain knowledge. Incorporating machine learning into ecological workflows could improve inputs for ecological models and lead to integrated hybrid modeling tools. This approach will require close interdisciplinary collaboration to ensure the quality of novel approaches and train a new generation of data scientists in ecology and conservation.

Counting using deep learning regression gives value to ecological surveys.

Many ecological studies rely on count data and involve manual counting of objects of interest, which is time-consuming and especially disadvantageous when time in the field or lab is limited. However, an increasing number of works uses digital imagery, which opens opportunities to automatise counting tasks. In this study, we use machine learning to automate counting objects of interest without the need to label individual objects. By leveraging already existing image-level annotations, this approach can also give value to historical data that were collected and annotated over longer time series (typical for many ecological studies), without the aim of deep learning applications. We demonstrate deep learning regression on two fundamentally different counting tasks: (i) daily growth rings from microscopic images of fish otolith (i.e., hearing stone) and (ii) hauled out seals from highly variable aerial imagery. In the otolith images, our deep learning-based regressor yields an RMSE of 3.40 day-rings and an [Formula: see text] of 0.92. Initial performance in the seal images is lower (RMSE of 23.46 seals and [Formula: see text] of 0.72), which can be attributed to a lack of images with a high number of seals in the initial training set, compared to the test set. We then show how to improve performance substantially (RMSE of 19.03 seals and [Formula: see text] of 0.77) by carefully selecting and relabelling just 100 additional training images based on initial model prediction discrepancy. The regression-based approach used here returns accurate counts ([Formula: see text] of 0.92 and 0.77 for the rings and seals, respectively), directly usable in ecological research.

Deep learning for automated detection of Drosophila suzukii: potential for UAV-based monitoring.

BACKGROUND: The fruit fly Drosophila suzukii, or spotted wing drosophila (SWD), is a serious pest worldwide, attacking many soft-skinned fruits. An efficient monitoring system that identifies and counts SWD in crops and their surroundings is therefore essential for integrated pest management (IPM) strategies. Existing methods, such as catching flies in liquid bait traps and counting them manually, are costly, time-consuming and labour-intensive. To overcome these limitations, we studied insect trap monitoring using image-based object detection with deep learning. RESULTS: Based on an image database with 4753 annotated SWD flies, we trained a ResNet-18-based deep convolutional neural network to detect and count SWD, including sex prediction and discrimination. The results show that SWD can be detected with an area under the precision recall curve (AUC) of 0.506 (female) and 0.603 (male) in digital images taken from a static position. For images collected using an unmanned aerial vehicle (UAV), the algorithm detected SWD individuals with an AUC of 0.086 (female) and 0.284 (male). The lower AUC for the aerial imagery was due to lower image quality caused by stabilisation manoeuvres of the UAV during image collection. CONCLUSION: Our results indicate that it is possible to monitor SWD using deep learning and object detection. Moreover, the results demonstrate the potential of UAVs to monitor insect traps, which could be valuable in the development of autonomous insect monitoring systems and IPM. © 2020 The Authors. Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.