Development of AI methods for seed yield prediction using RGB images: Progress Report

In the realm of agricultural advancement, the intersection of technology and farming practices has been a cornerstone for enhancing productivity and sustainability. As global food demands continue to rise, innovative solutions are crucial for maximizing agricultural outputs. This report focuses on o...

Full description

Bibliographic Details
Main Authors: Arrechea, Darwin A., Cardoso Arango, Juan Andrés
Format: Internal Document
Language:Inglés
Published: CGIAR Initiative on Accelerated Breeding 2023
Subjects:
Online Access:https://hdl.handle.net/10568/135079
_version_ 1855526629775769600
author Arrechea, Darwin A.
Cardoso Arango, Juan Andrés
author_browse Arrechea, Darwin A.
Cardoso Arango, Juan Andrés
author_facet Arrechea, Darwin A.
Cardoso Arango, Juan Andrés
author_sort Arrechea, Darwin A.
collection Repository of Agricultural Research Outputs (CGSpace)
description In the realm of agricultural advancement, the intersection of technology and farming practices has been a cornerstone for enhancing productivity and sustainability. As global food demands continue to rise, innovative solutions are crucial for maximizing agricultural outputs. This report focuses on one such innovative frontier: the application of artificial intelligence (AI) in agriculture, specifically in predicting seed yield. The focal point of our exploration is the automated counting of seeds in forages using DL algorithms. The current landscape of agricultural technology reveals a growing need for more accurate and efficient methods of predicting seed yields. Traditional manual-labor approaches, while beneficial, often fall short in terms of precision and scalability. The advent of AI promises a transformative shift in this domain. These technologies have already shown their prowess in various sectors and are now poised to revolutionize agricultural practices. The main goal of this project is to develop and refine AI methods that leverage RGB images for predicting seed yields. The initial choice of RGB images stems from their proven effectiveness in capturing vital plant health indicators, which are essential for accurate yield predictions. By enhancing the accuracy and efficiency of seed counting, is possible to have an efficient yield prediction. This project aims to empower farmers with actionable insights, optimize resource allocation, and ultimately contribute to higher and more sustainable agricultural outputs. This report will detail the methodologies employed, the experimental setup, data analysis, and the results obtained.
format Internal Document
id CGSpace135079
institution CGIAR Consortium
language Inglés
publishDate 2023
publishDateRange 2023
publishDateSort 2023
publisher CGIAR Initiative on Accelerated Breeding
publisherStr CGIAR Initiative on Accelerated Breeding
record_format dspace
spelling CGSpace1350792025-11-05T11:47:03Z Development of AI methods for seed yield prediction using RGB images: Progress Report Arrechea, Darwin A. Cardoso Arango, Juan Andrés seed characteristics crop modelling data collection artificial intelligence agriculture data In the realm of agricultural advancement, the intersection of technology and farming practices has been a cornerstone for enhancing productivity and sustainability. As global food demands continue to rise, innovative solutions are crucial for maximizing agricultural outputs. This report focuses on one such innovative frontier: the application of artificial intelligence (AI) in agriculture, specifically in predicting seed yield. The focal point of our exploration is the automated counting of seeds in forages using DL algorithms. The current landscape of agricultural technology reveals a growing need for more accurate and efficient methods of predicting seed yields. Traditional manual-labor approaches, while beneficial, often fall short in terms of precision and scalability. The advent of AI promises a transformative shift in this domain. These technologies have already shown their prowess in various sectors and are now poised to revolutionize agricultural practices. The main goal of this project is to develop and refine AI methods that leverage RGB images for predicting seed yields. The initial choice of RGB images stems from their proven effectiveness in capturing vital plant health indicators, which are essential for accurate yield predictions. By enhancing the accuracy and efficiency of seed counting, is possible to have an efficient yield prediction. This project aims to empower farmers with actionable insights, optimize resource allocation, and ultimately contribute to higher and more sustainable agricultural outputs. This report will detail the methodologies employed, the experimental setup, data analysis, and the results obtained. 2023-11 2023-12-06T15:26:58Z 2023-12-06T15:26:58Z Internal Document https://hdl.handle.net/10568/135079 en Open Access application/pdf CGIAR Initiative on Accelerated Breeding Arrechea, D.A.; Cardoso, J.A. (2023) Development of AI methods for seed yield prediction using RGB images: Progress Report. Cali (Colombia): CGIAR Initiative on Accelerated Breeding. 15 p.
spellingShingle seed characteristics
crop modelling
data collection
artificial intelligence
agriculture
data
Arrechea, Darwin A.
Cardoso Arango, Juan Andrés
Development of AI methods for seed yield prediction using RGB images: Progress Report
title Development of AI methods for seed yield prediction using RGB images: Progress Report
title_full Development of AI methods for seed yield prediction using RGB images: Progress Report
title_fullStr Development of AI methods for seed yield prediction using RGB images: Progress Report
title_full_unstemmed Development of AI methods for seed yield prediction using RGB images: Progress Report
title_short Development of AI methods for seed yield prediction using RGB images: Progress Report
title_sort development of ai methods for seed yield prediction using rgb images progress report
topic seed characteristics
crop modelling
data collection
artificial intelligence
agriculture
data
url https://hdl.handle.net/10568/135079
work_keys_str_mv AT arrecheadarwina developmentofaimethodsforseedyieldpredictionusingrgbimagesprogressreport
AT cardosoarangojuanandres developmentofaimethodsforseedyieldpredictionusingrgbimagesprogressreport