Deep learning and computer vision will transform entomology

Abstract

Most animal species on Earth are insects, and recent reports suggest that their abundance is in drastic decline. Although these reports come from a wide range of insect taxa and regions, the evidence to assess the extent of the phenomenon is sparse. Insect populations are challenging to study, and most monitoring methods are labor intensive and inefficient. Advances in computer vision and deep learning provide potential new solutions to this global challenge. Cameras and other sensors can effectively, continuously, and noninvasively perform entomological observations throughout diurnal and seasonal cycles. The physical appearance of specimens can also be captured by automated imaging in the laboratory. When trained on these data, deep learning models can provide estimates of insect abundance, biomass, and diversity. Further, deep learning models can quantify variation in phenotypic traits, behavior, and interactions. Here, we connect recent developments in deep learning and computer vision to the urgent demand for more cost-efficient monitoring of insects and other invertebrates. We present examples of sensor-based monitoring of insects. We show how deep learning tools can be applied to exceptionally large datasets to derive ecological information and discuss the challenges that lie ahead for the implementation of such solutions in entomology. We identify four focal areas, which will facilitate this transformation: 1) validation of image-based taxonomic identification; 2) generation of sufficient training data; 3) development of public, curated reference databases; and 4) solutions to integrate deep learning and molecular tools.

The effect of resource limitation on the temperature dependence of mosquito population fitness

“…Such systems are generally being used to test the effectiveness of novel biocontrol strategies, such as transgenic fungi [79], but they also could allow for the effects of temperature × resource interactions on fitness and abundance to be explored under conditions which more closely resemble natural environments. State of the art insect traps and geospatial mapping of microclimates and vegetation indices could also be used to study the effects of variation in temperature and resource availability on vector populations in the field [80,81].…”

Is it time to describe new species without diagnoses?—A comment on Sharkey et al. (2021)

“…This would be extremely detrimental to taxonomy as a science. Last but not least, deep learning and automated image recognition (Høye et al 2021;Gerovichev et al 2021) provide great potential even in small insects to employ morphology in taxonomy and rapid automatized biodiversity assessment, as well as for fast and automatized morphological trait extraction (e.g., Klasen et al 2020). Thus, future biodiversity research will rely on all available diagnostic data (morphology and DNA) for species identification and delimitation.…”

Revision and annotation of DNA barcode records for marine invertebrates: report of the 8thiBOL conference hackathon

“…The outstanding capabilities of machine learning (ML) and artificial intelligence (AI) systems have been extensively demonstrated in various research fields (Davenport and Kalakota 2019) including in image-based species identification (e.g. Ärje et al 2020Ärje et al , Høye et al 2021 and in the analysis of metabarcoding data for ecological status assessment (e.g. Cordier et al 2019, Frühe et al 2020).…”