Robust and simplified machine learning identification of pitfall trap-collected ground beetles at the continental scale

A gentle introduction to computer vision-based specimen classification in ecological datasets

Classifying specimens is a critical component of ecological research, biodiversity monitoring and conservation. However, manual classification can be prohibitively time-consuming and expensive, limiting how much data a project can afford to process. Computer vision, a form of machine learning, can help overcome these problems by rapidly, automatically and accurately classifying images of specimens. Given the diversity of animal species and contexts in which images are captured, there is no universal classifier for all species and use cases. As such, ecologists often need to train their own models. While numerous software programs exist to support this process, ecologists need a fundamental understanding of how computer vision works to select appropriate model workflows based on their specific use case, data types, computing resources and desired performance capabilities. Ecologists may also face characteristic quirks of ecological datasets, such as long-tail distributions, 'unknown' species, similarity between species and polymorphism within species, which impact the efficacy of computer vision. Despite growing interest in computer vision for ecology, there are few resources available to help ecologists face the challenges they are likely to encounter. Here, we present a gentle introduction for species classification using computer vision. In this manuscript and associated GitHub repository, we demonstrate how to prepare training data, basic model training procedures, and methods for model evaluation and selection. Throughout, we explore specific considerations ecologists should make when training classification models, such as data domains, feature extractors and class imbalances. With these basics, ecologists can adjust their workflows to achieve research goals and/or account for uncertainty in downstream analysis. Our goal is to provide guidance for ecologists for getting started in or improving their use of machine learning for visual classification tasks.

Accurate image-based identification of macroinvertebrate specimens using deep learning-How much training data is needed?

Image-based methods for species identification offer cost-efficient solutions for biomonitoring. This is particularly relevant for invertebrate studies, where bulk samples often represent insurmountable workloads for sorting, identifying, and counting individual specimens. On the other hand, image-based classification using deep learning tools have strict requirements for the amount of training data, which is often a limiting factor. Here, we examine how classification accuracy increases with the amount of training data using the BIODISCOVER imaging system constructed for image-based classification and biomass estimation of invertebrate specimens. We use a balanced dataset of 60 specimens of each of 16 taxa of freshwater macroinvertebrates to systematically quantify how classification performance of a convolutional neural network (CNN) increases for individual taxa and the overall community as the number of specimens used for training is increased. We show a striking 99.2% classification accuracy when the CNN (EfficientNet-B6) is trained on 50 specimens of each taxon, and also how the lower classification accuracy of models trained on less data is particularly evident for morphologically similar species placed within the same taxonomic order. Even with as little as 15 specimens used for training, classification accuracy reached 97%. Our results add to a recent body of literature showing the huge potential of image-based methods and deep learning for specimen-based research, and furthermore offers a perspective to future automatized approaches for deriving ecological data from bulk arthropod samples.

Pollution impacts on water bugs (Nepomorpha, Gerromorpha): state of the art and their biomonitoring potential

As water pollution poses an increasing risk worldwide, it is timely to assess the achievements of the aquatic macroinvertebrate ecotoxicology to provide a sound basis for the discipline's future and support the development of biomonitoring. Aquatic and semi-aquatic bugs (Hemiptera: Nepomorpha, Gerromorpha) are ubiquitous in almost all water types, sometimes in high densities, and play a significant role in organic material turnover and energy flow. Nevertheless, they are ignored in the water pollution biomonitoring schemes. Here, based on 300 papers, we review and evaluate the effects of chemical pesticides, microorganism-derived pesticides, insecticides of plant origin, heavy metals, eutrophication, salinisation and light pollution which are summarised for the first time. Our review encompasses the results of 100 laboratory and 39 semi-field/field experiments with 47 pesticides and 70 active ingredients. Pyrethroids were found to be more toxic than organochlorine, organophosphate and neonicotinoid insecticides to water bugs, like other macroinvertebrate groups. Additionally, in 10 out of 17 cases, the recommended field concentration of the pesticide was higher than the LC50 values, indicating potential hazards to water bugs. The recommended field concentrations of pesticides used in mosquito larvae control were found non-toxic to water bugs. As very few replicated studies are available, other findings on the effects of pesticides cannot be generalised. The microorganism-derived pesticide Bti appears to be safe when used at the recommended field concentration. Data indicates that plant-derived pesticides are safe with a high degree of certainty. We have identified three research areas where water bugs could be better involved in water biomonitoring. First, some Halobates spp. are excellent, and Gerris spp. are promising sentinels for Cd contamination. Second, Micronecta and, to a certain extent, Corixidae species composition is connected to and the indicator of eutrophication. Third, the species composition of the Corixidae is related to salinisation, and a preliminary method to quantify the relationship is already available. Our review highlights the potential of water bugs in water pollution monitoring.

Towards Machine Vision for Insect Welfare Monitoring and Behavioural Insights

Machine vision has demonstrated its usefulness in the livestock industry in terms of improving welfare in such areas as lameness detection and body condition scoring in dairy cattle. In this article, we present some promising results of applying state of the art object detection and classification techniques to insects, specifically Black Soldier Fly (BSF) and the domestic cricket, with the view of enabling automated processing for insect farming. We also present the low-cost “Insecto” Internet of Things (IoT) device, which provides environmental condition monitoring for temperature, humidity, CO₂, air pressure, and volatile organic compound levels together with high resolution image capture. We show that we are able to accurately count and measure size of BSF larvae and also classify the sex of domestic crickets by detecting the presence of the ovipositor. These early results point to future work for enabling automation in the selection of desirable phenotypes for subsequent generations and for providing early alerts should environmental conditions deviate from desired values.

Activity density at a continental scale: What drives invertebrate biomass moving across the soil surface?

Activity density (AD), the rate that an individual taxon or its biomass moves through the environment, is used both to monitor communities and quantify the potential for ecosystem work. The Abundance Velocity Hypothesis posited that AD increases with aboveground net primary productivity (ANPP) and is a unimodal function of temperature. Here we show that, at continental extents, increasing ANPP may have nonlinear effects on AD: increasing abundance, but decreasing velocity as accumulating vegetation interferes with movement. We use 5 yr of data from the NEON invertebrate pitfall trap arrays including 43 locations and four habitat types for a total of 77 habitat-site combinations to evaluate continental drivers of invertebrate AD. ANPP and temperature accounted for one-third to 92% of variation in AD. As predicted, AD was a unimodal function of temperature in forests and grasslands but increased linearly in open scrublands. ANPP yielded further nonlinear effects, generating unimodal AD curves in wetlands, and bimodal curves in forests. While all four habitats showed no AD trends over 5 yr of sampling, these nonlinearities suggest that trends in AD, often used to infer changes in insect abundance, will vary qualitatively across ecoregions.