From identification to forecasting: the potential of image recognition and artificial intelligence for aphid pest monitoring

Frontiers in Plant Science, Journal Year: 2023, Volume and Issue: 14

Published: July 19, 2023

Insect monitoring has gained global public attention in recent years the context of insect decline and biodiversity loss. Monitoring methods that can collect samples over a long period time independently human influences are particular importance. While these passive collection methods, e.g. suction traps, provide standardized comparable data sets, required to analyze large number trapped specimens is high. Another challenge necessary high level taxonomic expertise for accurate specimen processing. These factors create bottleneck In this context, machine learning, image recognition artificial intelligence have emerged as promising tools address shortcomings manual identification quantification analysis such trap catches. Aphids important agricultural pests pose significant risk several crops cause economic losses through feeding damage transmission plant viruses. It been shown long-term migrating aphids using traps be used make, adjust improve predictions their abundance so viruses spreading more accurately predicted. With increasing demand alternatives conventional pesticide use crop protection, need predictive models growing, basis resistance development measure management. advancing climate change strong influence on total well peak occurrences within year. Using model organism, we demonstrate possibilities systematic potential future technical developments subsequent automated individuals case intelligent forecasting models. an example, show from static images (i.e. advances software). We discuss applications with regard automatic processing prediction

Language: Английский

---

Insect detect: An open-source DIY camera trap for automated insect monitoring

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(4), P. e0295474 - e0295474

Published: April 3, 2024

Insect monitoring is essential to design effective conservation strategies, which are indispensable mitigate worldwide declines and biodiversity loss. For this purpose, traditional methods widely established can provide data with a high taxonomic resolution. However, processing of captured insect samples often time-consuming expensive, limits the number potential replicates. Automated facilitate collection at higher spatiotemporal resolution comparatively lower effort cost. Here, we present Detect DIY (do-it-yourself) camera trap for non-invasive automated flower-visiting insects, based on low-cost off-the-shelf hardware components combined open-source software. Custom trained deep learning models detect track insects landing an artificial flower platform in real time on-device subsequently classify cropped detections local computer. Field deployment solar-powered confirmed its resistance temperatures humidity, enables autonomous during whole season. On-device detection tracking estimate activity/abundance after metadata post-processing. Our classification model achieved top-1 accuracy test dataset generalized well real-world images. The software highly customizable be adapted different use cases. With custom models, as accessible programming, many possible applications surpassing our proposed method realized.

Language: Английский

---

Factors affecting catches of bark beetles and woodboring beetles in traps

Journal of Pest Science, Journal Year: 2024, Volume and Issue: 97(4), P. 1767 - 1793

Published: April 29, 2024

Abstract The use of semiochemical-baited traps for detection, monitoring, and sampling bark beetles woodboring (BBWB) has rapidly increased since the early 2000s. Semiochemical-baited survey are used in generic (broad community level) specific (targeted toward a species or group) surveys to detect nonnative potentially invasive BBWB, monitor established populations damaging native species, as tool natural communities various purposes. Along with expansion use, much research on ways improve efficacy trapping detection pests well BBWB general been conducted. In this review, we provide information intrinsic extrinsic factors how they influence detecting traps. Intrinsic factors, such trap type color, other described, important habitat selection, horizontal vertical placement, disturbance. When developing surveys, consideration these should increase richness and/or abundance captured probability that may be present. During deploying more than one using an array lures, at different positions is beneficial can number captured. Specific generally rely predetermined protocols recommendations type, lure, placement.

Language: Английский

---

Unravelling the use of artificial intelligence in management of insect pests

Smart Agricultural Technology, Journal Year: 2024, Volume and Issue: 8, P. 100517 - 100517

Published: July 29, 2024

As per the FAO, insect pest causes 30 to 40 percent loss every year across globe. The identification, classification and management of is very important avoid significant loss. Practicing above process by adopting manual methods are time consuming less effective achieve task. traditional often fall short in addressing dynamic behaviours, resulting crop losses increased chemical usage. Therefore, adoption Artificial Intelligence (AI) techniques identification act as a good substitute that arises from challenges posed evolving populations desire for sustainable agricultural practices. AI offers transformative approach utilizing advanced algorithms analyse intricate data patterns numerous sources like sensors imagery. This enables accurate early detection, predictive modelling, enhancing decision-making control, minimizing indiscriminate pesticide application optimizing interventions. not only reduces economic but also promotes eco-friendly strategies efficient resilient systems. present review an endeavour explain intermingling future scope management.

Language: Английский

---

Exploring the potential of saponins from Dicotyledonous plants in sustainable pest management: innovations and challenges; A reveiw

Journal of Natural Pesticide Research, Journal Year: 2025, Volume and Issue: unknown, P. 100111 - 100111

Published: Jan. 1, 2025

Language: Английский

---

Development of an autonomous smart trap for precision monitoring of hematophagous flies on cattle

Smart Agricultural Technology, Journal Year: 2025, Volume and Issue: unknown, P. 100842 - 100842

Published: Feb. 1, 2025

Language: Английский

---

Does insect trapping truly measure insect populations?

Agricultural and Forest Entomology, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 19, 2025

Abstract The measurement process has a well‐known structure and requires tools with proper characteristics, depending on the physical quantities their magnitude. While most fields of research have reliable to support experiments, this is not properly case for measurements in population dynamics insects animals, more general. Monitoring insect species common practice agriculture forest environments, above all develop pest control or biodiversity conservation strategies validate feed decision system tools. Besides development several monitoring techniques, an explicit connection between entomology metrology (the science measurements) still missing. We may ask if current involved populations, as traps instance, respect standard features that should have, they provide ‘proper measurements’ just ‘estimation’. This work analyses pros cons trapping by connecting provides some perspectives which communities focus interact answer questions open.

Language: Английский

---

Automatic Pest Monitoring Systems in Apple Production under Changing Climatic Conditions

Horticulturae, Journal Year: 2022, Volume and Issue: 8(6), P. 520 - 520

Published: June 14, 2022

Apple is one of the most important economic fruit crops in world. Despite all strategies integrated pest management (IPM), insecticides are still frequently used its cultivation. In addition, phenology extremely influenced by changing climatic conditions. The frequent spread invasive species, unexpected outbreaks, and development additional generations some problems posed climate change. adopted IPM therefore need to be changed as do current monitoring techniques, which increasingly unreliable outdated. for more sophisticated, accurate, efficient techniques leading increasing automated systems. this paper, we summarize automatic methods (image analysis systems, smart traps, sensors, decision support etc.) monitor major apple production (Cydia pomonella L.) other pests (Leucoptera maifoliella Costa, Grapholita molesta Busck, Halyomorpha halys Stål, flies—Tephritidae Drosophilidae) improve sustainable under

Language: Английский

---

An efficient mobile model for insect image classification in the field pest management

Engineering Science and Technology an International Journal, Journal Year: 2023, Volume and Issue: 39, P. 101335 - 101335

Published: Jan. 28, 2023

Accurately recognizing insect pest in their larva phase is significant to take the early treatment on infected crops, thus helping timely reduce yield loss agricultural products. The convolutional neural networks (CNNs)-based classification methods have become most competitive address many technical challenges related image recognition field. Focusing accurate and small models carried mobile devices, this study proposed a novel method PCNet (Pest Classification Network) based lightweight CNNs embedded attention mechanism. was designed with EfficientNet V2 as backbone, coordinate mechanism (CA) incorporated architecture learn inter-channel information positional of input images. Moreover, combining feature maps output by inverted bottleneck (MBConv) average pooling develop fusion module, which implements between shallow layers deep features down-sampling procedures. In addition, stochastic, pipeline-based data augmentation approach adopted randomly enhance diversity avoid model overfitting. experimental results show that achieved accuracy 98.4 % self-built dataset consisting 30 classes larvae, outperforms three classic CNN (AlexNet, VGG16, ResNet101), four (ShuffleNet V2, MobileNet V3, V1 V2). To further verify robustness different datasets, also tested two other public datasets: IP102 miniImageNet. 73.7 dataset, outperforming 94.0 miniImageNet only lower than ResNet101 V3. number parameters 20.7 M, less those traditional models. satisfactory size makes it suitable for real-time field resource constrained devices. Our code will be available at https://github.com/pby521/PCNet/tree/master.

Language: Английский

---

Advanced biosensing technologies for monitoring of agriculture pests and diseases: A review

Journal of Semiconductors, Journal Year: 2023, Volume and Issue: 44(2), P. 023104 - 023104

Published: Feb. 1, 2023

Abstract The threat posed to crop production by pests and diseases is one of the key factors that could reduce global food security. Early detection critical importance make accurate predictions, optimize control strategies prevent losses. Recent technological advancements highlight opportunity revolutionize monitoring diseases. Biosensing methodologies offer potential solutions for real-time automated monitoring, which allow in early thus support sustainable protection. Herein, advanced biosensing technologies including image-based technologies, electronic noses, wearable sensing methods are presented. Besides, challenges future perspectives widespread adoption these discussed. Moreover, we believe it necessary integrate through interdisciplinary cooperation further exploration, may provide unlimited possibilities innovations applications agriculture monitoring.

Language: Английский

---