Agricultural Robot-Centered Recognition of Early-Developmental Pest Stage Based on Deep Learning: A Case Study on Fall Armyworm (Spodoptera frugiperda)

Abstract

Accurately detecting early developmental stages of insect pests (larvae) from off-the-shelf stereo camera sensor data using deep learning holds several benefits for farmers, from simple robot configuration to early neutralization of this less agile but more disastrous stage. Machine vision technology has advanced from bulk spraying to precise dosage to directly rubbing on the infected crops. However, these solutions primarily focus on adult pests and post-infestation stages. This study suggested using a front-pointing red-green-blue (RGB) stereo camera mounted on a robot to identify pest larvae using deep learning. The camera feeds data into our deep-learning algorithms experimented on eight ImageNet pre-trained models. The combination of the insect classifier and the detector replicates the peripheral and foveal line-of-sight vision on our custom pest larvae dataset, respectively. This enables a trade-off between the robot’s smooth operation and localization precision in the pest captured, as it first appeared in the farsighted section. Consequently, the nearsighted part utilizes our faster region-based convolutional neural network-based pest detector to localize precisely. Simulating the employed robot dynamics using CoppeliaSim and MATLAB/SIMULINK with the deep-learning toolbox demonstrated the excellent feasibility of the proposed system. Our deep-learning classifier and detector exhibited 99% and 0.84 accuracy and a mean average precision, respectively.

Advancing horizons in vegetable cultivation: a journey from ageold practices to high-tech greenhouse cultivation—a review

“…This integration has given birth to innovative solutions for pest detection, classification, and localization leading to a significant improvement in agricultural efficiency. Deep learning models have been extensively applied for the detection and diagnosis of plant diseases and pests offering promising results and large potential in image processing and data analysis ( Ferentinos, 2018 ; Kamilaris and Prenafeta-Boldú, 2018 ; Liu and Wang, 2021 ), These models have been particularly effective in early pest detection, such as the recognition of insect pests at the larval stage before planting, allowing for precise localization and targeted intervention to minimize chemical usage ( Obasekore et al., 2023 ). This early identification allows for precise localization and targeted intervention minimizing the need for chemical pesticides.…”

Enhancing Crop Yield Predictions with PEnsemble 4: IoT and ML-Driven for Precision Agriculture

“…To the best of the authors' knowledge, this paper presents a novel and comprehensive approach that significantly differs from existing works in predicting crop yields. While previous studies have explored various methodologies, such as statistical modeling [7], and deep learning techniques [13,14], none have integrated multi-temporal images and machine learning algorithms in the manner proposed here.…”

Enhancing Crop Yield Predictions with PEnsemble 4: IoT and ML-Driven for Precision Agriculture

“…To the best of the authors' knowledge, this paper presents a novel and comprehensive approach that significantly differs from existing works in predicting crop yields. While previous studies have explored various methodologies, such as statistical modeling [26], and deep learning techniques [27,28], they have yet to integrate multi-temporal images and ML algorithms in the manner proposed here.…”