Remote Insects Trap Monitoring System Using Deep Learning Framework and IoT.

Insect detection and control at an early stage are essential to the built environment (human-made physical spaces such as homes, hotels, camps, hospitals, parks, pavement, food industries, etc.) and agriculture fields. Currently, such insect control measures are manual, tedious, unsafe, and time-consuming labor dependent tasks. With the recent advancements in Artificial Intelligence (AI) and the Internet of things (IoT), several maintenance tasks can be automated, which significantly improves productivity and safety. This work proposes a real-time remote insect trap monitoring system and insect detection method using IoT and Deep Learning (DL) frameworks. The remote trap monitoring system framework is constructed using IoT and the Faster RCNN (Region-based Convolutional Neural Networks) Residual neural Networks 50 (ResNet50) unified object detection framework. The Faster RCNN ResNet 50 object detection framework was trained with built environment insects and farm field insect images and deployed in IoT. The proposed system was tested in real-time using four-layer IoT with built environment insects image captured through sticky trap sheets. Further, farm field insects were tested through a separate insect image database. The experimental results proved that the proposed system could automatically identify the built environment insects and farm field insects with an average of 94% accuracy.

Evaluation of the effect of agroclimatic variables on the probability and timing of olive fruit fly attack.

Agroclimatic variables may affect insect and plant phenology, with unpredictable effects on pest populations and crop losses. Bactrocera oleae Rossi (Diptera: Tephritidae) is a specific pest of Olea europaea plants that can cause annual economic losses of more than one billion US dollars in the Mediterranean region. In this study, we aimed at understanding the effect of olive tree phenology and other agroclimatic variables on B. oleae infestation dynamics in the Umbria region (Central Italy). Analyses were carried out on B. oleae infestation data collected in 79 olive groves during a 7-year period (from 2015 to 2021). In July-August, B. oleae infestation (1% attack) was negatively affected by altitude and spring mean daily temperatures and positively by higher winter mean daily temperatures and olive tree cumulative degree days. In September-October, infestation was negatively affected by a positive soil water balance and high spring temperatures. High altitude and cumulative plant degree days were related to delayed attacks. In contrast, high winter and spring temperatures accelerated them. Our results could be helpful for the development of predictive models and for increasing the reliability of decision support systems currently used in olive orchards.

An annotated image dataset of pests on different coloured sticky traps acquired with different imaging devices.

The sticky trap is probably the most cost-effective tool for catching insect pests, but the identification and counting of insects on sticky traps is very labour-intensive. When investigating the automatic identification and counting of pests on sticky traps using computer vision and machine learning, two aspects can strongly influence the performance of the model - the colour of the sticky trap and the device used to capture the images of the pests on the sticky trap. As far as we know, there are no available image datasets to study these two aspects in computer vision and deep learning algorithms. Therefore, this paper presents a new dataset consisting of images of two pests commonly found in post-harvest crops - the red flour beetle (Tribolium castaneum) and the rice weevil (Sitophilus oryzae) - captured with three different devices (DSLR, webcam and smartphone) on blue, yellow, white and transparent sticky traps. The images were sorted by device, colour and species and divided into training, validation and test parts for the development of the deep learning model.

Diversity and composition of beetle assemblages attracted to dung in cloud forests under active and passive restoration practices.

Ecosystem loss and degradation has become a worldwide concern. The implementation of ecological restoration plans has been proposed to facilitate the recovery of ecosystems. It is imperative that once restoration strategies have been implemented, the effects of these actions in the medium and long term be evaluated, particularly the structure and functioning of the ecosystem. Diversity (α- and ß-diversity) of beetles attracted to dung was assessed and compared in 3 habitat conditions (conserved forest, passive restoration, and active restoration) at 2 different seasons during the year (dry vs. rainy season) in cloud forest in San Luis Potosí (central Mexico). We found that the dry season was slightly richer than the rainy season, but the latter was significantly more diverse. Species diversity and composition in active restoration were more similar to passive restoration, and both differed greatly from the conserved forest. In contrast, conserved and passive restoration conditions exhibited similar patterns in ß-diversity of insects likely because they maintain more species associated with the original vegetation of the cloud forest. Beetle assemblages could be of more habitat generalists, as they actively distribute across the restoration sites. Beetles attracted to dung provide an overview of the effect of restoration in early faunal recovery, even though we monitored this entomofauna for a short period (31 months after the restoration plots were established). These beetles can be a useful indicator for exploring the main forces driving species diversity for the management and conservation status of cloud forests, a threatened ecosystem.

Pan trapping is an effective method to trap adults of the jujube gall midge, Dasineura jujubifolia (Diptera: Cecidomyiidae).

Jujube gall midge (Dasineura jujubifolia Jiao & Bu) (Diptera: Cecidomyiidae) is an important pest in jujube (Ziziphus jujuba Mill.) orchards in Aksu, Xinjiang, China. Yellow sticky traps are the main device used for monitoring jujube gall midge adults, but their efficacy is low. Here, we compared the effectiveness of yellow sticky traps with water pan traps (are commonly used for trapping Diptera insects) to monitor jujube gall midge adults. Yellow sticky traps and pan traps were deployed for 2 consecutive years in jujube orchards in Aksu, Xinjiang, China. The midge's population dynamics as revealed by these 2 trap types were consistent, but the effectiveness of pan traps was about 5 times greater than that of the yellow sticky traps. In addition, pan traps captured fewer non-target species (e.g., parasitic wasps, lacewings, and lady beetles) than yellow sticky traps. Our study suggests that pan trap is an effective device to monitor jujube gall midge adults with minimal harm to natural enemies.

A New Approach for Detecting Sublethal Effects of Neonicotinoids on Bumblebees Using Optical Sensor Technology.

Among insects, bees are important pollinators, providing many vital ecosystem services. The recent pollinator decline is threatening both their diversity and abundance. One of the main drivers of this decline is the extensive use of pesticides. Neonicotinoids, one of the most popular groups of pesticides, can be toxic to bees. In fact, numerous studies have found that neonicotinoids can cause sublethal effects, which can impair the biology, physiology, and colony survival of the bees. Yet, there are still knowledge gaps, and more research is needed to better understand the interaction between neonicotinoids and bees, especially in the field. A new optical sensor, which can automatically identify flying insects using machine learning, has been created to continuously monitor insect activity in the field. This study investigated the potential use of this sensor as a tool for monitoring the sublethal effects of pesticides on bumblebees. Bombus terrestris workers were orally exposed to field-realistic doses of imidacloprid. Two types of exposures were tested: acute and chronic. The flight activity of pesticide-exposed and non-exposed bumblebees was recorded, and the events of the insect flights recorded by the sensor were used in two ways: to extract the values of the wingbeat frequency and to train machine learning models. The results showed that the trained model was able to recognize differences between the events created by pesticide-exposed bumblebees and the control bumblebees. This study demonstrates the possibility of the optical sensor for use as a tool to monitor bees that have been exposed to sublethal doses of pesticides. The optical sensor can provide data that could be helpful in managing and, ideally, mitigating the decline of pollinators from one of their most major threats, pesticides.

Abundance of Halyomorpha halys (Hemiptera: Pentatomidae) and Megacopta cribraria (Hemiptera: Plataspidae) in Soybean in Areas with Few Previous Sightings in Tennessee.

Halyomorpha halys (Stål) and Megacopta cribraria (Fabricius) are two exotic invasive pests that have invaded the United States in recent years. Halyomorpha halys can damage various fruits, vegetables, and field crops, such as soybean and corn, while Megacopta cribraria only attacks soybean and kudzu, a weed species. They are currently found in southeastern states and threaten soybean and other crops grown in the region. This study evaluated the seasonal abundance of H. halys and M. cribraria in soybeans in 2016 and 2017 in two counties in the central region of Tennessee, where both species had either a few sightings or none that were recorded when this research was being planned. Lures and sweep sampling were used to monitor H. halys, and sweep sampling was used to monitor M. cribraria. Halyomorpha halys was first detected in samples in late July. Their numbers increased in early to mid-September, reached the economic threshold in late Sept, and then started to decline. Megacopta cribraria was first detected in mid to late July, increased their populations in September, but did not reach the economic threshold and declined mid-October. Our results showed the seasonal abundances of H. halys and M. cribraria and their establishment in the central region of Tennessee.

Arthropod Surveillance Programs: Basic Components, Strategies, and Analysis.

Effective entomological surveillance planning stresses a careful consideration of methodology, trapping technologies, and analysis techniques. Herein, the basic principles and technological components of arthropod surveillance plans are described, as promoted in the symposium "Advancements in arthropod monitoring technology, techniques, and analysis" presented at the 58th annual meeting of the Entomological Society of America in San Diego, CA. Interdisciplinary examples of arthropod monitoring for urban, medical, and veterinary applications are reviewed. Arthropod surveillance consists of the three components: 1) sampling method, 2) trap technology, and 3) analysis technique. A sampling method consists of selecting the best device or collection technique for a specific location and sampling at the proper spatial distribution, optimal duration, and frequency to achieve the surveillance objective. Optimized sampling methods are discussed for several mosquito species (Diptera: Culicidae) and ticks (Acari: Ixodidae). The advantages and limitations of novel terrestrial and aerial insect traps, artificial pheromones and kairomones are presented for the capture of red flour beetle (Coleoptera: Tenebrionidae), small hive beetle (Coleoptera: Nitidulidae), bed bugs (Hemiptera: Cimicidae), and Culicoides (Diptera: Ceratopogonidae) respectively. After sampling, extrapolating real world population numbers from trap capture data are possible with the appropriate analysis techniques. Examples of this extrapolation and action thresholds are given for termites (Isoptera: Rhinotermitidae) and red flour beetles.

Semi-Quantitative Detection of Drosophila suzukii (Diptera: Drosophilidae) From Bulk Trap Samples Using PCR Technology.

Drosophila suzukii (Matsumura) (Diptera: Drosophilidae) is a ubiquitous global pest of several fruit crops. Trapped adult numbers are used to monitor populations and make control decisions, but differentiating D. suzukii from other trapped Drosophila spp. is laborious. We developed a real-time PCR method for specific detection and semi-quantification of D. suzukii from trap samples. The PCR assay did not amplify DNA from 29 other Drosophilidae species tested. Drosophila suzukii was detected from ≥0.96 pg target DNA and from laboratory samples containing one D. suzukii in 2000 other Drosophila spp. flies. We tested DNA stability of one D. suzukii in 100 Drosophila spp. flies in water or ethanol at 20, 25, or 30°C for 1, 4, or 7 d. Only water at 30°C for 7 d fully impaired D. suzukii DNA detectability. Substituting mouthwash for water resulted in D. suzukii detection in all samples held for 7 d at 30°C or daily fluctuating temperatures of 33/23°C. Traps with mouthwash as a drowning liquid had D. suzukii captures equal to traps with water. A calibration curve was established using samples in mouthwash containing 1/1,000-100/1,000 D. suzukii/total Drosophila spp. flies and incubated at 25°C for 7 d. The curve had a coefficient of determination (R2) of 0.9279 between D. suzukii numbers from the PCR and the true D. suzukii numbers in samples prepared in 70% ethanol. Collecting samples in mouthwash is expected to improve detection accuracy, and the qPCR method can be a useful tool to support D. suzukii monitoring and management.

Camera traps are an effective tool for monitoring insect-plant interactions.

Insect and pollinator populations are vitally important to the health of ecosystems, food production, and economic stability, but are declining worldwide. New, cheap, and simple monitoring methods are necessary to inform management actions and should be available to researchers around the world. Here, we evaluate the efficacy of a commercially available, close-focus automated camera trap to monitor insect-plant interactions and insect behavior. We compared two video settings-scheduled and motion-activated-to a traditional human observation method. Our results show that camera traps with scheduled video settings detected more insects overall than humans, but relative performance varied by insect order. Scheduled cameras significantly outperformed motion-activated cameras, detecting more insects of all orders and size classes. We conclude that scheduled camera traps are an effective and relatively inexpensive tool for monitoring interactions between plants and insects of all size classes, and their ease of accessibility and set-up allows for the potential of widespread use. The digital format of video also offers the benefits of recording, sharing, and verifying observations.

Using Malaise Traps and Metabarcoding for Biodiversity Assessment in Vineyards: Effects of Weather and Trapping Effort.

Metabarcoding is a powerful tool for ecological studies and monitoring that might provide a solution to the time-consuming taxonomic identification of the vast diversity of insects. Here, we assess how ambient weather conditions during Malaise trap exposure and the effort of trapping affect biomass and taxa richness in vineyards. Biomass varied by more than twofold with weather conditions. It increased with warmer and drier weather but was not significantly related with wind or precipitation. Taxa richness showed a saturating relationship with increasing trapping duration and was influenced by environmental and seasonal effects. Taxa accumulation was high, increasing fourfold from three days of monthly trap exposure compared to continuous trapping and nearly sixfold from sampling at a single site compared to 32 sites. The limited saturation was mainly due to a large number of singletons, such as rare species, in the metabarcoding dataset. Metabarcoding can be key for long-term insect monitoring. We conclude that single traps operated for up to ten days per month are suitable to monitor the presence of common species. However, more intensive trapping is necessary for a good representation of rare species in biodiversity monitoring. The data collected here can potentially guide the design of monitoring studies.

Sensor-based outdoor monitoring of insects in arable crops for their precise control.

The implementation of precision farming technologies into agricultural practice requires, among other things, precise determination of the extent and intensity of insect infestation in the farmer' fields. Manual insect identification is time-consuming and has low efficiency, especially for large fields. Therefore, scientists and practitioners devote much effort to the automatization of this process. There are two complementary approaches to insect identification: (i) direct, in which the insect (ultimately the species) is determined, and (ii) indirect, in which the damage caused by the insects is monitored and forms the basis on which to formulate the information about insect infestation. A mini-review of both approaches is presented in this work. Additionally, the advantages and disadvantages of each are briefly described. Methods of insect identification are still characterized by relatively small selectivity and efficiency, therefore it is necessary to keep searching for new methods and improve the development of existing ones. The goal of such systems should be to work in real time and be inexpensive to run, enabling widespread use amongst farmers. A possible solution seems to be integrating various techniques (sensor fusion) into a single measurement system. © 2020 Society of Chemical Industry.

A Novel CO2-Based Insect Sampling Device and Associated Field Method Evaluated in a Strawberry Agroecosystem.

There is high demand for accurate insect sampling methods to inform integrated pest management strategies. Despite widespread application, existing sampling methods, such as portable aspirating and sweep netting, can result in overrepresentation of prominent pests, underrepresentation of natural enemies, and damage to plants. In this study, we test a novel device for insect sampling via anesthetization. Specifically, we test the effect of CO2 (application pressure and duration of exposure) on Lygus hesperus Knight (Hemiptera: Miridae) anesthetization in the laboratory and on insect community density in a strawberry agroecosystem. Carbon dioxide application proves an effective means of anesthetization compared to negative controls, and an increase in net CO2 exposure results in a decrease in time until L. hesperus anesthetization. Field results indicate the CO2 method collects more parasitoids and thrips than a portable aspirator, and at the 50 PSI application pressure and 15-s exposure, the CO2 method results in a comparable number of pests collected as the research standard, a portable aspirator with 8-s aspiration time. Benefits of the CO2 method include minimal plant damage, highly explicit spatial and temporal data, and scalability.

Summer Development and Survivorship of the Weed Biocontrol Agent, Mecinus janthiniformis (Coleoptera: Curculionidae), Within Stems of Its Host, Dalmatian Toadflax (Lamiales: Plantaginaceae), in Utah.

The summer phenology and survivorship of the stem-mining weevil, Mecinus janthiniformis Tosevski and Caldara, a biocontrol agent of Dalmatian toadflax, Linaria dalmatica (L.) Miller, was studied in 2015-2016 as it developed within host plant stems at a low elevation, open rangeland site in northern Utah. Hatching from eggs in spring and early summer, weevils occurred as larvae within stems in June. Earliest maturing adults occurred in mid-July, and the majority of individuals had completed pupal development by early August. Survivorship within stems was high, with two-thirds or more of individuals surviving from egg hatch to adulthood as assessed in mid-September. Mortality rates within stems were highest during larval development, with parasitism accounting for the majority of deaths. At least three parasitoid species (Chalcidoidea: Pteromalidae and Eupelmidae), including both endoparasitoids and ectoparasitoids, were found attacking weevils within stems. Although most surviving weevils remained as adults within stems to overwinter, some adults were found to have chewed exit holes, and in some cases had exited from stems, beginning in July; the fate of these prematurely exiting adults is unknown. Low summer mortality rates within stems should promote weevil establishment under the hot, dry conditions of northern Utah, but parasitism and premature exiting of adults from host stems merit further investigation concerning their potential to reduce biocontrol efficacy. The results presented here for M. janthiniformis phenology within host stems will contribute to the development of standardized, summer monitoring for this biocontrol agent by stem dissection.

Colored Sticky Traps to Selectively Survey Thrips in Cowpea Ecosystem.

The bean flower thrips, Megalurothrips usitatus (Bagrall) (Thysanoptera: Thripidae), is an important pest of legume crops in South China. Yellow, blue, or white sticky traps are currently recommended for monitoring and controlling thrips, but it is not known whether one is more efficient than the other or if selectivity could be optimized by trap color. We investigated the response of thrips and beneficial insects to different-colored sticky traps on cowpea, Vigna unguiculata. More thrips were caught on blue, light blue, white, and purple traps than on yellow, green, pink, gray, red, or black traps. There was a weak correlation on the number of thrips caught on yellow traps and survey from flowers (r = 0.139), whereas a strong correlation was found for blue traps and thrips' survey on flowers (r = 0.929). On commercially available sticky traps (Jiaduo®), two and five times more thrips were caught on blue traps than on white and yellow traps, respectively. Otherwise, capture of beneficial insects was 1.7 times higher on yellow than on blue traps. The major natural enemies were the predatory ladybird beetles (63%) and pirate bugs Orius spp. (29%), followed by a number of less representative predators and parasitoids (8%). We conclude the blue sticky trap was the best to monitor thrips on cowpea in South China.

Multistate Comparison of Attractants for Monitoring Drosophila suzukii (Diptera: Drosophilidae) in Blueberries and Caneberries.

Drosophila suzukii Matsumara, also referred to as the spotted wing drosophila, has recently expanded its global range with significant consequences for its primary host crops: blueberries, blackberries, raspberries, cherries, and strawberries. D. suzukii populations can increase quickly, and their infestation is difficult to predict and prevent. The development of effective tools to detect D. suzukii presence in new areas, to time the beginning of activity within a crop, to track seasonal activity patterns, and to gauge the effectiveness of management efforts has been a key research goal. We compared the efficiency, selectivity, and relationship to fruit infestation of a range of commonly used homemade baits and a synthetic formulated lure across a wide range of environments in 10 locations throughout the United States. Several homemade baits were more efficient than apple cider vinegar, a commonly used standard, and a commercially formulated lure was, in some configurations and environments, comparable with the most effective homemade attractant as well as potentially more selective. All alternative attractants also captured flies between 1 and 2 wk earlier than apple cider vinegar, and detected the presence of D. suzukii prior to the development of fruit infestation. Over half the Drosophila spp. flies captured in traps baited with any of the attractants were not D. suzukii, which may complicate their adoption by nonexpert users. The alternative D. suzukii attractants tested are improvement on apple cider vinegar and may be useful in the development of future synthetic lures.

Multiscale approach to pest insect monitoring: random walks, pattern formation, synchronization, and networks.

Pest insects pose a significant threat to food production worldwide resulting in annual losses worth hundreds of billions of dollars. Pest control attempts to prevent pest outbreaks that could otherwise destroy a sward. It is good practice in integrated pest management to recommend control actions (usually pesticides application) only when the pest density exceeds a certain threshold. Accurate estimation of pest population density in ecosystems, especially in agro-ecosystems, is therefore very important, and this is the overall goal of the pest insect monitoring. However, this is a complex and challenging task; providing accurate information about pest abundance is hardly possible without taking into account the complexity of ecosystems' dynamics, in particular, the existence of multiple scales. In the case of pest insects, monitoring has three different spatial scales, each of them having their own scale-specific goal and their own approaches to data collection and interpretation. In this paper, we review recent progress in mathematical models and methods applied at each of these scales and show how it helps to improve the accuracy and robustness of pest population density estimation.