Agricultural and Forest Entomology · 2024 · 20 citations · 48 references
AgricultureComputer ScienceSustainable AgricultureAgricultural EntomologyArtificial IntelligenceIntegrated Pest ManagementDecision SupportInsect Pest SurveillancePest DetectionPest IdentificationPest ManagementPest SurveillanceImage-based Pest DetectionImage-based Pest RecognitionInsect Pest DetectionInsect Pest IdentificationIntegrated Pest Management SupportsIntelligent Decision Support SystemsPest Monitoring IntegrationTechnological Expertise
Abstract Recent years have seen significant advances in artificial intelligence (AI) technology. This advancement has enabled the development of decision support systems that support farmers with herbivorous pest identification and pest monitoring. In these systems, the AI supports farmers through the detection, classification and quantification of herbivorous pests. However, many of the systems under development fall short of meeting the demands of the end user, with these shortfalls acting as obstacles that impede the integration of these systems into integrated pest management (IPM) practices. There are four common obstacles that restrict the uptake of these AI‐driven decision support systems. Namely: AI technology effectiveness, functionality under field conditions, the level of computational expertise and power required to use and run the system and system mobility. We propose four criteria that AI‐driven systems need to meet in order to overcome these challenges: (i) The system should be based on effective and efficient AI; (ii) The system should be adaptable and capable of handling ‘real‐world’ image data collected from the field; (iii) Systems should be user‐friendly, device‐driven and low‐cost; (iv) Systems should be mobile and deployable under multiple weather and climate conditions. Systems that meet these criteria are likely to represent innovative and transformative systems that successfully integrate AI technology with IPM principles into tools that can support farmers.
48
Feature Pyramid Networks for Object Detection
Tsung-Yi Lin, Piotr Dollár, Ross Girshick et al. · 2017 · 28.7K citations
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Shaoqing Ren, Kaiming He, Ross Girshick et al. · arXiv (Cornell University) · 2015 · 18.2K citations · Full text
Proceedings of IEEE Conference on Computer Vision and Pattern Recognition
1993 · 14.3K citations
The global burden of pathogens and pests on major food crops
Serge Savary, Laetitia Willocquet, Sarah J. Pethybridge et al. · Nature Ecology & Evolution · 2019 · 4K citations · Full text