Abstract:
This paper presents SMARTRAPS, an innovative device designed for automated monitoring and identification of insect pests, aimed at enhancing food security. Insect pests, such as fruit flies and fall armyworm (FAW), pose significant challenges to agricultural productivity and food sustainability. Fruit flies are known for their detrimental impact on fruits and vegetables, while FAW is a major threat to maize crops. Conventional pest monitoring methods, such as pheromone traps and sticky tapes, often prove time- consuming, labor-intensive, and ineffective for large fields. To address this challenge, SMARTRAPS leverages artificial intelligence techniques and real-time image analysis collected by a camera, along with Internet of Things (IoT) sensors. The device aims to automate pest monitoring, detect attacks, and identify hotspots of pest growth in fields and indoor storage spaces. By providing farmers with real-time data on pest infestations, SMARTRAPS empowers efficient and reliable decision-making for timely pest control interventions. The device operates on solar power and integrates seamlessly with a mobile application, facilitating efficient pest management practices. The dataset used to train and validate the object detection model was collected from insect-rearing laboratories and local fields. Our implementation of the object detection model yielded promising results in identifying different fruit fly species and FAW in captured images. This model leverages a 24-layer Convolutional Neural Network (CNN) for robust feature extraction and accurately classifies fruit fly species and FAW based on various visual attributes, including color, size, shape, and wing, abdomen, and thorax patterns. In field trials conducted in mango orchards and maize fields, SMARTRAPS demonstrated exceptional efficacy, achieving an impressive 84% detection accuracy even in challenging environmental conditions. These findings were further validated through comparison with ground truth data labeled by expert entomologists. SMARTRAPS represents a significant advancement in pest management strategies, offering a proactive solution to mitigate crop losses and promote sustainable agriculture. By integrating AI, IoT, and image analysis technologies, this device plays a vital role in strengthening food security by safeguarding crop yields and minimizing the impact of insect pests on agricultural production.
Page(s):
82-82
DOI:
DOI not available
Published:
Journal: Abstract Book on International Conference on Food and Applied Sciences (ICFAS-23) 3-5 August 23, Volume: 0, Issue: 0, Year: 2023
Keywords:
detection
,
Food security
,
Artificial Intelligence
,
IoT
,
Accuracy
,
Sustainable agriculture
,
pest monitoring
,
realtime data