Weed detection in a Rice Crop through Image Processing and Classification Using Convolutional Neural Networks

Tojonirina Ranaivoson, Hantanirina Rosiane Rasamimanana, Andriniaina Narindra Rasoanaivo, Omer Andrianarimanana, Alfred Andriamamonjy and Dieudonné Razafimahatratra

MPRA Paper from University Library of Munich, Germany

Abstract: Artificial Intelligence (AI) today occupies a central ranking, especially in a context where technological progress is omnipresent. Among the most influential tools, deep learning has established itself in both professional and academic domains. This article focuses on the effectiveness of convolutional neural networks for detecting weeds competing with rice. To achieve this, an extension of the pre-trained Inception_V3 model was used for image classification, while MobileNet was employed for image processing. This innovative approach, tested on a rice field where distinguishing between rice and weeds is challenging, represents a significant advancement in the AI field. However, the training of both models revealed limitations: Inception_V3 exhibited overfitting after the 10th iteration, while MobileNet showed high volatility and overfitting from the first iteration. Despite these challenges, Inception_V3 stood out for its superior accuracy.

Keywords: Convolutional; Neural; Pre-trained; Detection (search for similar items in EconPapers)
JEL-codes: Q16 (search for similar items in EconPapers)
Date: 2024-11-28, Revised 2025-01-27
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://mpra.ub.uni-muenchen.de/123474/1/MPRA_paper_123474.pdf original version (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:pra:mprapa:123474

Access Statistics for this paper

More papers in MPRA Paper from University Library of Munich, Germany Ludwigstraße 33, D-80539 Munich, Germany. Contact information at EDIRC.
Bibliographic data for series maintained by Joachim Winter ().