Deep learning and computer vision will transform entomology

A Swin Transformer-based model for mosquito species identification

Mosquito transmit numbers of parasites and pathogens resulting in fatal diseases. Species identification is a prerequisite for effective mosquito control. Existing morphological and molecular classification methods have evitable disadvantages. Here we introduced Deep learning techniques for mosquito species identification. A balanced, high-definition mosquito dataset with 9900 original images covering 17 species was constructed. After three rounds of screening and adjustment-testing (first round among 3 convolutional neural networks and 3 Transformer models, second round among 3 Swin Transformer variants, and third round between 2 images sizes), we proposed the first Swin Transformer-based mosquito species identification model (Swin MSI) with 99.04% accuracy and 99.16% F1-score. By visualizing the identification process, the morphological keys used in Swin MSI were similar but not the same as those used by humans. Swin MSI realized 100% subspecies-level identification in Culex pipiens Complex and 96.26% accuracy for novel species categorization. It presents a promising approach for mosquito identification and mosquito borne diseases control.

An overview of remote monitoring methods in biodiversity conservation

Conservation of biodiversity is critical for the coexistence of humans and the sustenance of other living organisms within the ecosystem. Identification and prioritization of specific regions to be conserved are impossible without proper information about the sites. Advanced monitoring agencies like the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) had accredited that the sum total of species that are now threatened with extinction is higher than ever before in the past and are progressing toward extinct at an alarming rate. Besides this, the conceptualized global responses to these crises are still inadequate and entail drastic changes. Therefore, more sophisticated monitoring and conservation techniques are required which can simultaneously cover a larger surface area within a stipulated time frame and gather a large pool of data. Hence, this study is an overview of remote monitoring methods in biodiversity conservation via a survey of evidence-based reviews and related studies, wherein the description of the application of some technology for biodiversity conservation and monitoring is highlighted. Finally, the paper also describes various transformative smart technologies like artificial intelligence (AI) and/or machine learning algorithms for enhanced working efficiency of currently available techniques that will aid remote monitoring methods in biodiversity conservation.

A Review of Automated Bioacoustics and General Acoustics Classification Research

Automated bioacoustics classification has received increasing attention from the research community in recent years due its cross-disciplinary nature and its diverse application. Applications in bioacoustics classification range from smart acoustic sensor networks that investigate the effects of acoustic vocalizations on species to context-aware edge devices that anticipate changes in their environment adapt their sensing and processing accordingly. The research described here is an in-depth survey of the current state of bioacoustics classification and monitoring. The survey examines bioacoustics classification alongside general acoustics to provide a representative picture of the research landscape. The survey reviewed 124 studies spanning eight years of research. The survey identifies the key application areas in bioacoustics research and the techniques used in audio transformation and feature extraction. The survey also examines the classification algorithms used in bioacoustics systems. Lastly, the survey examines current challenges, possible opportunities, and future directions in bioacoustics.

Pest-YOLO: A model for large-scale multi-class dense and tiny pest detection and counting

Frequent outbreaks of agricultural pests can reduce crop production severely and restrict agricultural production. Therefore, automatic monitoring and precise recognition of crop pests have a high practical value in the process of agricultural planting. In recent years, pest recognition and detection have been rapidly improved with the development of deep learning-based methods. Although certain progress has been made in the research on pest detection and identification technology based on deep learning, there are still many problems in the production application in a field environment. This work presents a pest detector for multi-category dense and tiny pests named the Pest-YOLO. First, the idea of focal loss is introduced into the loss function using weight distribution to improve the attention of hard samples. In this way, the problems of hard samples arose from the uneven distribution of pest populations in a dataset and low discrimination features of small pests are relieved. Next, a non-Intersection over Union bounding box selection and suppression algorithm, the confluence strategy, is used. The confluence strategy can eliminate the errors and omissions of pest detection caused by occlusion, adhesion and unlabeling among tiny dense pest individuals to the greatest extent. The proposed Pest-YOLO model is verified on a large-scale pest image dataset, the Pest24, which includes more than 20k images with over 190k pests labeled by agricultural experts and categorized into 24 classes. Experimental results show that the Pest-YOLO can obtain 69.59% for mAP and 77.71% for mRecall on the 24-class pest dataset, which is 5.32% and 28.12% higher than the benchmark model YOLOv4. Meanwhile, our proposed model is superior to other several state-of-the-art methods, including the SSD, RetinaNet, Faster RCNN, YOLOv3, YOLOv4, YOLOv5s, YOLOv5m, YOLOX, DETR, TOOD, YOLOv3-W, and AF-RCNN detectors. The code of the proposed algorithm is available at: https://github.com/chr-secrect/Pest-YOLO.

Measuring Plant Attractiveness to Pollinators: Methods and Considerations

Global pollinator declines have fostered increased public interest in creating pollinator-friendly gardens in human-managed landscapes. Indeed, studies on urban pollinator communities suggest that flower-rich greenspaces can serve as promising sites for conservation. Ornamental flowers, which are readily available at most commercial garden centers, are ubiquitous in these landscapes. These varieties are often non-native and highly bred, and their utility to pollinators is complex. In this study, we used observational data and citizen science to develop a methods framework that will assist stakeholders in the floriculture industry to incorporate metrics of pollinator health into existing breeding and evaluation protocols. The results of this study support how plant attractiveness to pollinators is often dependent on variables such as climate and plant phenology, which should be considered when developing an assessment tool. Furthermore, we found that some cultivars were consistently attractive across all observations while for other cultivars, pollinator visitation was apparently conditional. We determine using multiple statistical tests that 10 min is a sufficient length of time for observation of most plant types to broadly estimate three measures of plant attractiveness: visitor abundance, primary visitors attracted, and cultivar rank attractiveness, without sacrificing efficiency or accuracy. Additionally, we demonstrate that properly trained non-expert observers can collect accurate observational data, and our results suggest that protocols may be designed to maximize consistency across diverse data collectors.

Reference bioimaging to assess the phenotypic trait diversity of bryophytes within the family Scapaniaceae

Macro- and microscopic images of organisms are pivotal in biodiversity research. Despite that bioimages have manifold applications such as assessing the diversity of form and function, FAIR bioimaging data in the context of biodiversity are still very scarce, especially for difficult taxonomic groups such as bryophytes. Here, we present a high-quality reference dataset containing macroscopic and bright-field microscopic images documenting various phenotypic characters of the species belonging to the liverwort family of Scapaniaceae occurring in Europe. To encourage data reuse in biodiversity and adjacent research areas, we annotated the imaging data with machine-actionable metadata using community-accepted semantics. Furthermore, raw imaging data are retained and any contextual image processing like multi-focus image fusion and stitching were documented to foster good scientific practices through source tracking and provenance. The information contained in the raw images are also of particular interest for machine learning and image segmentation used in bioinformatics and computational ecology. We expect that this richly annotated reference dataset will encourage future studies to follow our principles.

How does IPM 3.0 look like (and why do we need it in Africa)?

The concept of Integrated Pest Management (IPM) was introduced sixty years ago to curb the overuse of agricultural pesticides, whereby its simplest version (IPM 1.0) was aiming at reducing the frequency of applications. Gradually, agro-ecological principles, such as biological control and habitat management, were included in IPM 2.0. However, throughout this time, smallholder farmers did not improve their decision-making skills and continue to use hazardous pesticides as their first control option. We are therefore proposing a new paradigm - IPM 3.0 - anchored on 3 pillars: 1) real-time farmer access to decision-making, 2) pest-management options relying on science-driven and nature-based approaches, and 3) the integration of genomic approaches, biopesticides, and habitat-management practices. We are convinced that this new paradigm based on technological advances, involvement of youth, gender-responsiveness, and climate resilience will be a game changer. However, this can only become effective through redeployment of public funding and stronger policy support.

Towards intelligent and integrated pest management through an AIoT-based monitoring system

BACKGROUND

Main bottleneck in facilitating integrated pest management (IPM) is the unavailability of reliable and immediate crop damage data. Without sufficient insect pest and plant disease information, farm managers are unable to make proper decisions to prevent crop damage. This work aims to present how an integrated system was able to drive farm managers towards sustainable and data-driven IPM.

RESULTS

A system called Intelligent and Integrated Pest and Disease Management (I² PDM) system was developed. Edge computing devices were developed to automatically detect and recognize major greenhouse insect pests such as thrips (Frankliniella intonsa, Thrips hawaiiensis, and Thrips tabaci), and whiteflies (Bemisia argentifolii and Trialeurodes vaporariorum), to name a few, and measure environmental conditions including temperature, humidity, and light intensity, and send data to a remote server. The system has been installed in greenhouses producing tomatoes and orchids for gathering long-term spatiotemporal insect pest count and environmental data, for as long as 1368 days. The findings demonstrated that the proposed system supported the farm managers in performing IPM-related tasks. Significant yearly reductions in insect pest count as high as 50.7% were observed on the farms.

CONCLUSION

It was concluded that novel and efficient strategies can be achieved by using an intelligent IPM system, opening IPM to potential benefits that cannot be easily realized with a traditional IPM program. This is the first work that reports the development of an intelligent strategic model for IPM based on actual automatically collected long-term data. The work presented herein can help in encouraging farm managers, researchers, experts, and industries to work together in implementing sustainable and data-driven IPM. © 2022 Society of Chemical Industry.

Herbivores in Arctic ecosystems: Effects of climate change and implications for carbon and nutrient cycling

Arctic terrestrial herbivores influence tundra carbon and nutrient dynamics through their consumption of resources, waste production, and habitat-modifying behaviors. The strength of these effects is likely to change spatially and temporally as climate change drives shifts in herbivore abundance, distribution, and activity timing. Here, we review how herbivores influence tundra carbon and nutrient dynamics through their consumptive and nonconsumptive effects. We also present evidence for herbivore responses to climate change and discuss how these responses may alter the spatial and temporal distribution of herbivore impacts. Several current knowledge gaps limit our understanding of the changing functional roles of herbivores; these include limited characterization of the spatial and temporal variability in herbivore impacts and of how herbivore activities influence the cycling of elements beyond carbon. We conclude by highlighting approaches that will promote better understanding of herbivore effects on tundra ecosystems, including their integration into existing biogeochemical models, new applications of remote sensing techniques, and the continued use of distributed experiments.

A Taxonomic Review of the Genus Telsimia Casey (Coleoptera, Coccinellidae) from China, with Descriptions of Eight New Species

Twenty species of the genus Telsimia from China are reviewed herein. Among them, eight species are described as new to science: Telsimia chayuensis, T. forcipata, T. latus, T. lobatus, T. lunata, T. menglaensis, T. parascymnoides, and T. parvus spp. nov.; two species are reported from China for the first time: T. darjeelingensis Kapur, 1969 and T. elongate Hoàng, 1985. All species are provided with nomenclatural history, diagnoses, detailed descriptions (except for the 10 previously described species), colored illustrations, and distributions. The female genitalia of five described species are provided for the first time. A distribution map and a key to all the Chinese species are also provided.

Diel Periodicity in Males of the Navel Orangeworm (Lepidoptera: Pyralidae) as Revealed by Automated Camera Traps

Navel orangeworm, Amyelois transitella (Walker), is a key pest of walnuts, pistachio, and almonds in California. Pheromone mating disruption using timed aerosol dispensers is an increasingly common management technique. Dispenser efficiency may be increased by timing releases with the active mating period of navel orangeworm. Past work found that the peak time of sexual activity for navel orangeworm females is 2 h before sunrise when temperatures are above 18°C. Inference of male responsiveness from data collected in that study was limited by the necessity of using laboratory-reared females as a source of sex pheromone emission to attract males and the inherent limitations of human observers for nocturnal events. Here we used camera traps baited with artificial pheromone to observe male navel orangeworm mating response in the field over two field seasons. Male response to synthetic pheromone exhibited diel patterns broadly similar to females, i.e., they were active for a brief period of 2-3 h before dawn under summer conditions and began responding to pheromone earlier and over a longer period of time during spring and fall. But contrary to the previous findings with females, some males were captured at all hours of the day and night, and there was no evidence of short-term change of pheromone responsiveness in response to temperature. Environmental effects on the response of navel orangeworm males to an artificial pheromone source differ in important ways from the environmental effects on female release of sex pheromone.

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.

Machine learning for stone artifact identification: Distinguishing worked stone artifacts from natural clasts using deep neural networks

Stone artifacts are often the most abundant class of objects found in archaeological sites but their consistent identification is limited by the number of experienced analysts available. We report a machine learning based technology for stone artifact identification as part of a solution to the lack of such experts directed at distinguishing worked stone objects from naturally occurring lithic clasts. Three case study locations from Egypt, Australia, and New Zealand provide a data set of 6769 2D images, 3868 flaked artifact and 2901 rock images used to train and test a machine learning model based on an openly available PyTorch implementation of Faster R-CNN ResNet 50. Results indicate 100% agreement between the model and original human derived classifications, a better performance than the results achieved independently by two human analysts who reassessed the 2D images available to the machine learning model. Machine learning neural networks provide the potential to consistently assess the composition of large archaeological assemblages composed of objects modified in a variety of ways.

Periodically taken photographs reveal the effect of pollinator insects on seed set in lotus flowers

Understanding of pollination systems is an important topic for evolutionary ecology, food production, and biodiversity conservation. However, it is difficult to grasp the whole picture of an individual system, because the activity of pollinators fluctuates depending on the flowering period and time of day. In order to reveal effective pollinator taxa and timing of visitation to the reproductive success of plants under the complex biological interactions and fluctuating abiotic factors, we developed an automatic system to take photographs at 5-s intervals to get near-complete flower visitation by pollinators during the entire flowering period of selected flowers of Nelumbo nucifera and track the reproductive success of the same flowers until fruiting. Bee visits during the early morning hours of 05:00-07:59 on the second day of flowering under optimal temperatures with no rainfall or strong winds contributed strongly to seed set, with possible indirect negative effects by predators of the pollinators. Our results indicate the availability of periodic and consecutive photography system in clarifying the plant-pollinator interaction and its consequence to reproductive success of the plant. Further development is required to build a monitoring system to collect higher-resolution time-lapse images and automatically identify visiting insect species in the natural environment.

Emerging technologies revolutionise insect ecology and monitoring

Insects are the most diverse group of animals on Earth, but their small size and high diversity have always made them challenging to study. Recent technological advances have the potential to revolutionise insect ecology and monitoring. We describe the state of the art of four technologies (computer vision, acoustic monitoring, radar, and molecular methods), and assess their advantages, current limitations, and future potential. We discuss how these technologies can adhere to modern standards of data curation and transparency, their implications for citizen science, and their potential for integration among different monitoring programmes and technologies. We argue that they provide unprecedented possibilities for insect ecology and monitoring, but it will be important to foster international standards via collaboration.

Assessing the effect of complex ground types on ground-dwelling arthropod movements with video monitoring: Dealing with concealed movements under a layer of plant residues

Understanding the effect of ground types on foraging movements of ground-dwelling arthropods is a key step to managing their spatial distribution as required for successful conservation biological control. Indeed, fine movements at the centimeter scale can strongly influence the foraging ability of pest predators. However, because radio frequency identification or harmonic tracking techniques are not yet suitable for small species and video tracking focuses on uniform and light backgrounds, foraging movements have rarely been studied in relation to ground types. We present a method to track a ground-dwelling arthropod (the earwig Euborellia caraibea) at night, walking on two contrasted ground types: bare soil and soil partly covered with a stratum of banana plant residues allowing individuals to hide periodically. The tracking of individuals within these ground types was achieved by infrared light, tagging individuals, video treatments, and semi-automatic cleaning of trajectories. We tested different procedures to obtain segments with identical durations to quantify speeds and sinuosities. These procedures were characterized by the junction time gap between trajectory fragments, the rediscretization time of trajectories, and whether or not to use interpolation to fill in missing points in the trajectories. Earwigs exhibited significantly slower and more sinuous movements on soil with banana plant residues than on bare soil. Long time gaps for trajectory junction, extended rediscretization times, and interpolation were complementary means to integrate concealed movements in the trajectories. The highest slowdown in plant residues was detected when the procedure could account for longer periods under the residues. These results suggest that earwigs spent a significant amount of time concealed by the residues. Additionally, the residues strongly decreased the earwigs' movement. Since the technical solutions presented in this study are inexpensive, easy to set up, and replicate, they represent valuable contributions to the emerging field of video monitoring.

Sticky Pi is a high-frequency smart trap that enables the study of insect circadian activity under natural conditions

In the face of severe environmental crises that threaten insect biodiversity, new technologies are imperative to monitor both the identity and ecology of insect species. Traditionally, insect surveys rely on manual collection of traps, which provide abundance data but mask the large intra- and interday variations in insect activity, an important facet of their ecology. Although laboratory studies have shown that circadian processes are central to insects' biological functions, from feeding to reproduction, we lack the high-frequency monitoring tools to study insect circadian biology in the field. To address these issues, we developed the Sticky Pi, a novel, autonomous, open-source, insect trap that acquires images of sticky cards every 20 minutes. Using custom deep learning algorithms, we automatically and accurately scored where, when, and which insects were captured. First, we validated our device in controlled laboratory conditions with a classic chronobiological model organism, Drosophila melanogaster. Then, we deployed an array of Sticky Pis to the field to characterise the daily activity of an agricultural pest, Drosophila suzukii, and its parasitoid wasps. Finally, we demonstrate the wide scope of our smart trap by describing the sympatric arrangement of insect temporal niches in a community, without targeting particular taxa a priori. Together, the automatic identification and high sampling rate of our tool provide biologists with unique data that impacts research far beyond chronobiology, with applications to biodiversity monitoring and pest control as well as fundamental implications for phenology, behavioural ecology, and ecophysiology. We released the Sticky Pi project as an open community resource on https://doc.sticky-pi.com.

Moths complement bumblebee pollination of red clover: a case for day-and-night insect surveillance

Recent decades have seen a surge in awareness about insect pollinator declines. Social bees receive the most attention, but most flower-visiting species are lesser known, non-bee insects. Nocturnal flower visitors, e.g. moths, are especially difficult to observe and largely ignored in pollination studies. Clearly, achieving balanced monitoring of all pollinator taxa represents a major scientific challenge. Here, we use time-lapse cameras for season-wide, day-and-night pollinator surveillance of Trifolium pratense (L.; red clover) in an alpine grassland. We reveal the first evidence to suggest that moths, mainly Noctua pronuba (L.; large yellow underwing), pollinate this important wildflower and forage crop, providing 34% of visits (bumblebees: 61%). This is a remarkable finding; moths have received no recognition throughout a century of T. pratense pollinator research. We conclude that despite a non-negligible frequency and duration of nocturnal flower visits, nocturnal pollinators of T. pratense have been systematically overlooked. We further show how the relationship between visitation and seed set may only become clear after accounting for moth visits. As such, population trends in moths, as well as bees, could profoundly affect T. pratense seed yield. Ultimately, camera surveillance gives fair representation to non-bee pollinators and lays a foundation for automated monitoring of species interactions in future.

Evolutionary Ecology of Plant-Arthropod Interactions in Light of the "Omics" Sciences: A Broad Guide

Aboveground plant-arthropod interactions are typically complex, involving herbivores, predators, pollinators, and various other guilds that can strongly affect plant fitness, directly or indirectly, and individually, synergistically, or antagonistically. However, little is known about how ongoing natural selection by these interacting guilds shapes the evolution of plants, i.e., how they affect the differential survival and reproduction of genotypes due to differences in phenotypes in an environment. Recent technological advances, including next-generation sequencing, metabolomics, and gene-editing technologies along with traditional experimental approaches (e.g., quantitative genetics experiments), have enabled far more comprehensive exploration of the genes and traits involved in complex ecological interactions. Connecting different levels of biological organization (genes to communities) will enhance the understanding of evolutionary interactions in complex communities, but this requires a multidisciplinary approach. Here, we review traditional and modern methods and concepts, then highlight future avenues for studying the evolution of plant-arthropod interactions (e.g., plant-herbivore-pollinator interactions). Besides promoting a fundamental understanding of plant-associated arthropod communities' genetic background and evolution, such knowledge can also help address many current global environmental challenges.

Wild Bee Nutritional Ecology: Integrative Strategies to Assess Foraging Preferences and Nutritional Requirements

Bees depend on flowering plants for their nutrition, and reduced availability of floral resources is a major driver of declines in both managed and wild bee populations. Understanding the nutritional needs of different bee species, and how these needs are met by the varying nutritional resources provided by different flowering plant taxa, can greatly inform land management recommendations to support bee populations and their associated ecosystem services. However, most bee nutrition research has focused on the three most commonly managed and commercially reared bee taxa—honey bees, bumble bees, and mason bees—with fewer studies focused on wild bees and other managed species, such as leafcutting bees, stingless bees, and alkali bees. Thus, we have limited information about the nutritional requirements and foraging preferences of the vast majority of bee species. Here, we discuss the approaches traditionally used to understand bee nutritional ecology: identification of floral visitors of selected focal plant species, evaluation of the foraging preferences of adults in selected focal bee species, evaluation of the nutritional requirements of focal bee species (larvae or adults) in controlled settings, and examine how these methods may be adapted to study a wider range of bee species. We also highlight emerging technologies that have the potential to greatly facilitate studies of the nutritional ecology of wild bee species, as well as evaluate bee nutritional ecology at significantly larger spatio-temporal scales than were previously feasible. While the focus of this review is on bee species, many of these techniques can be applied to other pollinator taxa as well.

Edge Computing for Vision-Based, Urban-Insects Traps in the Context of Smart Cities

Our aim is to promote the widespread use of electronic insect traps that report captured pests to a human-controlled agency. This work reports on edge-computing as applied to camera-based insect traps. We present a low-cost device with high power autonomy and an adequate picture quality that reports an internal image of the trap to a server and counts the insects it contains based on quantized and embedded deep-learning models. The paper compares different aspects of performance of three different edge devices, namely ESP32, Raspberry Pi Model 4 (RPi), and Google Coral, running a deep learning framework (TensorFlow Lite). All edge devices were able to process images and report accuracy in counting exceeding 95%, but at different rates and power consumption. Our findings suggest that ESP32 appears to be the best choice in the context of this application according to our policy for low-cost devices.

Sharing taxonomic expertise between natural history collections using image recognition

Natural history collections play a vital role in biodiversity research and conservation by providing a window to the past. The usefulness of the vast amount of historical data depends on their quality, with correct taxonomic identifications being the most critical. The identification of many of the objects of natural history collections, however, is wanting, doubtful or outdated. Providing correct identifications is difficult given the sheer number of objects and the scarcity of expertise. Here we outline the construction of an ecosystem for the collaborative development and exchange of image recognition algorithms designed to support the identification of objects. Such an ecosystem will facilitate sharing taxonomic expertise among institutions by offering image datasets that are correctly identified by their in-house taxonomic experts. Together with openly accessible machine learning algorithms and easy to use workbenches, this will allow other institutes to train image recognition algorithms and thereby compensate for the lacking expertise.

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.

Long-term abundance trends of insect taxa are only weakly correlated

Changes in the abundances of animals, such as with the ongoing concern about insect declines, are often assumed to be general across taxa. However, this assumption is largely untested. Here, we used a database of assemblage-wide long-term insect and arachnid monitoring to compare abundance trends among co-occurring pairs of taxa. We show that 60% of co-occurring taxa qualitatively showed long-term trends in the same direction-either both increasing or both decreasing. However, in terms of magnitude, temporal trends were only weakly correlated (mean freshwater r = 0.05 (±0.03), mean terrestrial r = 0.12 (±0.09)). The strongest correlation was between trends of beetles and those of moths/butterflies (r = 0.26). Overall, even though there is some support for directional similarity in temporal trends, we find that changes in the abundance of one taxon provide little information on the changes of other taxa. No clear candidate for umbrella or indicator taxa emerged from our analysis. We conclude that obtaining a better picture of changes in insect abundances will require monitoring of multiple taxa, which remains uncommon, especially in the terrestrial realm.

A Computer Vision-Based Approach for Tick Identification Using Deep Learning Models

A wide range of pathogens, such as bacteria, viruses, and parasites can be transmitted by ticks and can cause diseases, such as Lyme disease, anaplasmosis, or Rocky Mountain spotted fever. Landscape and climate changes are driving the geographic range expansion of important tick species. The morphological identification of ticks is critical for the assessment of disease risk; however, this process is time-consuming, costly, and requires qualified taxonomic specialists. To address this issue, we constructed a tick identification tool that can differentiate the most encountered human-biting ticks, Amblyomma americanum, Dermacentor variabilis, and Ixodes scapularis, by implementing artificial intelligence methods with deep learning algorithms. Many convolutional neural network (CNN) models (such as VGG, ResNet, or Inception) have been used for image recognition purposes but it is still a very limited application in the use of tick identification. Here, we describe the modified CNN-based models which were trained using a large-scale molecularly verified dataset to identify tick species. The best CNN model achieved a 99.5% accuracy on the test set. These results demonstrate that a computer vision system is a potential alternative tool to help in prescreening ticks for identification, an earlier diagnosis of disease risk, and, as such, could be a valuable resource for health professionals.

Classification of Fruit Flies by Gender in Images Using Smartphones and the YOLOv4-Tiny Neural Network

The fruit fly Drosophila melanogaster is a classic research object in genetics and systems biology. In the genetic analysis of flies, a routine task is to determine the offspring size and gender ratio in their populations. Currently, these estimates are made manually, which is a very time-consuming process. The counting and gender determination of flies can be automated by using image analysis with deep learning neural networks on mobile devices. We proposed an algorithm based on the YOLOv4-tiny network to identify Drosophila flies and determine their gender based on the protocol of taking pictures of insects on a white sheet of paper with a cell phone camera. Three strategies with different types of augmentation were used to train the network. The best performance (F1 = 0.838) was achieved using synthetic images with mosaic generation. Females gender determination is worse than that one of males. Among the factors that most strongly influencing the accuracy of fly gender recognition, the fly’s position on the paper was the most important. Increased light intensity and higher quality of the device cameras have a positive effect on the recognition accuracy. We implement our method in the FlyCounter Android app for mobile devices, which performs all the image processing steps using the device processors only. The time that the YOLOv4-tiny algorithm takes to process one image is less than 4 s.

Novel CropdocNet Model for Automated Potato Late Blight Disease Detection from Unmanned Aerial Vehicle-Based Hyperspectral Imagery

The accurate and automated diagnosis of potato late blight disease, one of the most destructive potato diseases, is critical for precision agricultural control and management. Recent advances in remote sensing and deep learning offer the opportunity to address this challenge. This study proposes a novel end-to-end deep learning model (CropdocNet) for accurate and automated late blight disease diagnosis from UAV-based hyperspectral imagery. The proposed method considers the potential disease-specific reflectance radiation variance caused by the canopy’s structural diversity and introduces multiple capsule layers to model the part-to-whole relationship between spectral–spatial features and the target classes to represent the rotation invariance of the target classes in the feature space. We evaluate the proposed method with real UAV-based HSI data under controlled and natural field conditions. The effectiveness of the hierarchical features is quantitatively assessed and compared with the existing representative machine learning/deep learning methods on both testing and independent datasets. The experimental results show that the proposed model significantly improves accuracy when considering the hierarchical structure of spectral–spatial features, with average accuracies of 98.09% for the testing dataset and 95.75% for the independent dataset, respectively.

A Computer Vision Approach to Identifying Ticks Related to Lyme Disease

Background: Lyme disease (caused by Borrelia burgdorferi) is an infectious disease transmitted to humans by a bite from infected blacklegged ticks (Ixodes scapularis) in eastern North America. Lyme disease can be prevented if antibiotic prophylaxis is given to a patient within 72 hours of a blacklegged tick bite. Therefore, recognizing a blacklegged tick could facilitate the management of Lyme disease. Methods: In this work, we build an automated detection tool that can differentiate blacklegged ticks from other tick species using advanced computer vision approaches in real-time. Specially, we use convolution neural network models, trained end-to-end, to classify tick species. Also, advanced knowledge transfer techniques are adopted to improve the performance of convolution neural network models. Results: Our best convolution neural network model achieves 92% accuracy on unseen tick species. Conclusion: Our proposed vision-based approach simplifies tick identification and contributes to the emerging work on public health surveillance of ticks and tick-borne diseases. In addition, it can be integrated with the geography of exposure and potentially be leveraged to inform the risk of Lyme disease infection. This is the first report of using deep learning technologies to classify ticks, providing the basis for automation of tick surveillance, and advancing tick-borne disease ecology and risk management.

Improving the Accuracy of Species Identification by Combining Deep Learning With Field Occurrence Records

Citizen science is essential for nationwide ecological surveys of species distribution. While the accuracy of the information collected by beginner participants is not guaranteed, it is important to develop an automated system to assist species identification. Deep learning techniques for image recognition have been successfully applied in many fields and may contribute to species identification. However, deep learning techniques have not been utilized in ecological surveys of citizen science, because they require the collection of a large number of images, which is time-consuming and labor-intensive. To counter these issues, we propose a simple and effective strategy to construct species identification systems using fewer images. As an example, we collected 4,571 images of 204 species of Japanese dragonflies and damselflies from open-access websites (i.e., web scraping) and scanned 4,005 images from books and specimens for species identification. In addition, we obtained field occurrence records (i.e., range of distribution) of all species of dragonflies and damselflies from the National Biodiversity Center, Japan. Using the images and records, we developed a species identification system for Japanese dragonflies and damselflies. We validated that the accuracy of the species identification system was improved by combining web-scraped and scanned images; the top-1 accuracy of the system was 0.324 when trained using only web-scraped images, whereas it improved to 0.546 when trained using both web-scraped and scanned images. In addition, the combination of images and field occurrence records further improved the top-1 accuracy to 0.668. The values of top-3 accuracy under the three conditions were 0.565, 0.768, and 0.873, respectively. Thus, combining images with field occurrence records markedly improved the accuracy of the species identification system. The strategy of species identification proposed in this study can be applied to any group of organisms. Furthermore, it has the potential to strike a balance between continuously recruiting beginner participants and updating the data accuracy of citizen science.

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

The accuracy of specimen identification through DNA barcoding and metabarcoding relies on reference libraries containing records with reliable taxonomy and sequence quality. The considerable growth in barcode data requires stringent data curation, especially in taxonomically difficult groups such as marine invertebrates. A major effort in curating marine barcode data in the Barcode of Life Data Systems (BOLD) was undertaken during the 8th International Barcode of Life Conference (Trondheim, Norway, 2019). Major taxonomic groups (crustaceans, echinoderms, molluscs, and polychaetes) were reviewed to identify those which had disagreement between Linnaean names and Barcode Index Numbers (BINs). The records with disagreement were annotated with four tags: a) MIS-ID (misidentified, mislabeled, or contaminated records), b) AMBIG (ambiguous records unresolved with the existing data), c) COMPLEX (species names occurring in multiple BINs), and d) SHARE (barcodes shared between species). A total of 83,712 specimen records corresponding to 7,576 species were reviewed and 39% of the species were tagged (7% MIS-ID, 17% AMBIG, 14% COMPLEX, and 1% SHARE). High percentages (>50%) of AMBIG tags were recorded in gastropods, whereas COMPLEX tags dominated in crustaceans and polychaetes. The high proportion of tagged species reflects either flaws in the barcoding workflow (e.g., misidentification, cross-contamination) or taxonomic difficulties (e.g., synonyms, undescribed species). Although data curation is essential for barcode applications, such manual attempts to examine large datasets are unsustainable and automated solutions are extremely desirable.

Development of a deep learning model from breeding substrate images: a novel method for estimating the abundance of house fly (Musca domestica L.) larvae

BACKGROUND

The application of computer vision and deep learning to pest monitoring has recently received much attention. Although several studies have demonstrated the application of object detection to the number of pests on a substrate, for house flies (Musca domestica L.), in which the larvae were aggregated and overlapped together, the object detection technique was difficult to implement. We demonstrate a novel method for estimating larval abundance by using computer vision on larval breeding substrate, in which the reflective color and topography are affected by the size of the population.

RESULTS

We demonstrate a method using a web-based tool to construct a deep learning model and later export the model for deployment. We train the model by using breeding substrate images with different spectra of illumination on known densities of larvae and evaluate the training model in both the test set and field-collected samples. In general, the model was able to predict the larval abundance by the laboratory-prepared breeding substrate with 87.56% to 94.10% accuracy, precision, recall, and F-score on the unseen test set, and white and green illumination performed significantly higher compared to other illuminations. For field samples, the model was able to obtain at least 70% correct predictions by using white and infrared illumination.

CONCLUSION

Larval abundance can be monitored with computer vision and deep learning, and the monitoring can be improved by using more biochemistry parameters as the predictors and examples of field samples included building a more robust model. © 2021 Society of Chemical Industry.

A Special Collection: Drones to Improve Insect Pest Management

The Special Collection Drones to Improve Insect Pest Management presents research and development of unmanned (or uncrewed) aircraft system (UAS, or drone) technology to improve insect pest management. The articles bridge from more foundational studies (i.e., evaluating and refining abilities of drones to detect pest concerns or deliver pest management materials) to application-oriented case studies (i.e., evaluating opportunities and challenges of drone use in pest management systems). The collection is composed of a combination of articles presenting information first-time published, and a selection of articles previously published in Journal of Economic Entomology (JEE). Articles in the Collection, as well as selected citations of articles in other publications, reflect the increase in entomology research using drones that has been stimulated by advancement in drone structural and software engineering such as autonomous flight guidance; in- and post-flight data storage and processing; and companion advances in spatial data management and analyses including machine learning and data visualization. The Collection is also intended to stimulate discussion on the role of JEE as a publication venue for future articles on drones as well as other cybernectic-physical systems, big data analyses, and deep learning processes. While these technologies have their genesis in fields arguably afar from the discipline of entomology, we propose that interdisciplinary collaboration is the pathway for applications research and technology transfer leading to an acceleration of research and development of these technologies to improve pest management.

Do amateurs and citizen science fill the gaps left by scientists?

The diversity of insects is tremendous and so is the effort needed to assess it in order to better understand insect ecology as well as their role for the functioning of ecosystems. While the interest of academics and naturalists for these species has always existed, it is only recently that such interest started to reach society more generally. From insect taxonomy and distribution to the collection of large range and long scale monitoring data, the involvement of non-academics in research outputs is growing. Is this a sign of scientists not being able to meet expectations or of science getting more and more entrenched in society? We argue for the latter, highlighting the opportunities that such involvement of amateurs in insect science represent for insect conservation.

Image-based Automated Species Identification: Can Virtual Data Augmentation Overcome Problems of Insufficient Sampling?

Automated species identification and delimitation is challenging, particularly in rare and thus often scarcely sampled species, which do not allow sufficient discrimination of infraspecific versus interspecific variation. Typical problems arising from either low or exaggerated interspecific morphological differentiation are best met by automated methods of machine learning that learn efficient and effective species identification from training samples. However, limited infraspecific sampling remains a key challenge also in machine learning. In this study, we assessed whether a data augmentation approach may help to overcome the problem of scarce training data in automated visual species identification. The stepwise augmentation of data comprised image rotation as well as visual and virtual augmentation. The visual data augmentation applies classic approaches of data augmentation and generation of artificial images using a Generative Adversarial Networks (GAN) approach. Descriptive feature vectors are derived from bottleneck features of a VGG-16 convolutional neural network (CNN) that are then stepwise reduced in dimensionality using Global Average Pooling and PCA to prevent overfitting. Finally, data augmentation employs synthetic additional sampling in feature space by an oversampling algorithm in vector space (SMOTE). Applied on four different image datasets, which include scarab beetle genitalia (Pleophylla, Schizonycha) as well as wing patterns of bees (Osmia) and cattleheart butterflies (Parides), our augmentation approach outperformed a deep learning baseline approach by means of resulting identification accuracy with non-augmented data as well as a traditional 2D morphometric approach (Procrustes analysis of scarab beetle genitalia).

Automatic ladybird beetle detection using deep-learning models

Fast and accurate taxonomic identification of invasive trans-located ladybird beetle species is essential to prevent significant impacts on biological communities, ecosystem functions, and agricultural business economics. Therefore, in this work we propose a two-step automatic detector for ladybird beetles in random environment images as the first stage towards an automated classification system. First, an image processing module composed of a saliency map representation, simple linear iterative clustering superpixels segmentation, and active contour methods allowed us to generate bounding boxes with possible ladybird beetles locations within an image. Subsequently, a deep convolutional neural network-based classifier selects only the bounding boxes with ladybird beetles as the final output. This method was validated on a 2, 300 ladybird beetle image data set from Ecuador and Colombia obtained from the iNaturalist project. The proposed approach achieved an accuracy score of 92% and an area under the receiver operating characteristic curve of 0.977 for the bounding box generation and classification tasks. These successful results enable the proposed detector as a valuable tool for helping specialists in the ladybird beetle detection problem.

High Throughput Data Acquisition and Deep Learning for Insect Ecoinformatics

Ecology documents and interprets the abundance and distribution of organisms. Ecoinformatics addresses this challenge by analyzing databases of observational data. Ecoinformatics of insects has high scientific and applied importance, as insects are abundant, speciose, and involved in many ecosystem functions. They also crucially impact human well-being, and human activities dramatically affect insect demography and phenology. Hazards, such as pollinator declines, outbreaks of agricultural pests and the spread insect-borne diseases, raise an urgent need to develop ecoinformatics strategies for their study. Yet, insect databases are mostly focused on a small number of pest species, as data acquisition is labor-intensive and requires taxonomical expertise. Thus, despite decades of research, we have only a qualitative notion regarding fundamental questions of insect ecology, and only limited knowledge about the spatio-temporal distribution of insects. We describe a novel high throughput cost-effective approach for monitoring flying insects as an enabling step toward “big data” entomology. The proposed approach combines “high tech” deep learning with “low tech” sticky traps that sample flying insects in diverse locations. As a proof of concept we considered three recent insect invaders of Israel’s forest ecosystem: two hemipteran pests of eucalypts and a parasitoid wasp that attacks one of them. We developed software, based on deep learning, to identify the three species in images of sticky traps from Eucalyptus forests. These image processing tasks are quite difficult as the insects are small (<5 mm) and stick to the traps in random poses. The resulting deep learning model discriminated the three focal organisms from one another, as well as from other elements such as leaves and other insects, with high precision. We used the model to compare the abundances of these species among six sites, and validated the results by manually counting insects on the traps. Having demonstrated the power of the proposed approach, we started a more ambitious study that monitors these insects at larger spatial and temporal scales. We aim at building an ecoinformatics repository for trap images and generating data-driven models of the populations’ dynamics and morphological traits.

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

Laboratory-derived temperature dependencies of life-history traits are increasingly being used to make mechanistic predictions for how climatic warming will affect vector-borne disease dynamics, partially by affecting abundance dynamics of the vector population. These temperature-trait relationships are typically estimated from juvenile populations reared on optimal resource supply, even though natural populations of vectors are expected to experience variation in resource supply, including intermittent resource limitation. Using laboratory experiments on the mosquito Aedes aegypti, a principal arbovirus vector, combined with stage-structured population modelling, we show that low-resource supply in the juvenile life stages significantly depresses the vector's maximal population growth rate across the entire temperature range (22-32°C) and causes it to peak at a lower temperature than at high-resource supply. This effect is primarily driven by an increase in juvenile mortality and development time, combined with a decrease in adult size with temperature at low-resource supply. Our study suggests that most projections of temperature-dependent vector abundance and disease transmission are likely to be biased because they are based on traits measured under optimal resource supply. Our results provide compelling evidence for future studies to consider resource supply when predicting the effects of climate and habitat change on vector-borne disease transmission, disease vectors and other arthropods.

Automated Applications of Acoustics for Stored Product Insect Detection, Monitoring, and Management

Simple Summary

A variety of different acoustic devices has been commercialized for detection of hidden insect infestations in stored products, trees, and soil, including a recently introduced device demonstrated in this report to successfully detect rice weevil immatures and adults in grain. Several of the systems have incorporated digital signal processing and statistical analyses such as neural networks and machine learning to distinguish targeted pests from each other and from background noise, enabling automated monitoring of the abundance and distribution of pest insects in stored products, and potentially reducing the need for chemical control. Current and previously available devices are reviewed in the context of the extensive research in stored product insect acoustic detection since 2011. It is expected that further development of acoustic technology for detection and management of stored product insect pests will continue, facilitating automation and decreasing detection and management costs.

Abstract

Acoustic technology provides information difficult to obtain about stored insect behavior, physiology, abundance, and distribution. For example, acoustic detection of immature insects feeding hidden within grain is helpful for accurate monitoring because they can be more abundant than adults and be present in samples without adults. Modern engineering and acoustics have been incorporated into decision support systems for stored product insect management, but with somewhat limited use due to device costs and the skills needed to interpret the data collected. However, inexpensive modern tools may facilitate further incorporation of acoustic technology into the mainstream of pest management and precision agriculture. One such system was tested herein to describe Sitophilus oryzae (Coleoptera: Curculionidae) adult and larval movement and feeding in stored grain. Development of improved methods to identify sounds of targeted pest insects, distinguishing them from each other and from background noise, is an active area of current research. The most powerful of the new methods may be machine learning. The methods have different strengths and weaknesses depending on the types of background noise and the signal characteristic of target insect sounds. It is likely that they will facilitate automation of detection and decrease costs of managing stored product insects in the future.