REVISTA
AI

   
Redirigiendo al acceso original de articulo en 17 segundos...
Inicio  /  AI  /  Vol: 2 Par: 1 (2021)  /  Artículo
ARTÍCULO
TITULO

Testing the Suitability of Automated Machine Learning for Weeds Identification

Borja Espejo-Garcia    
Ioannis Malounas    
Eleanna Vali and Spyros Fountas    

Resumen

In the past years, several machine-learning-based techniques have arisen for providing effective crop protection. For instance, deep neural networks have been used to identify different types of weeds under different real-world conditions. However, these techniques usually require extensive involvement of experts working iteratively in the development of the most suitable machine learning system. To support this task and save resources, a new technique called Automated Machine Learning has started being studied. In this work, a complete open-source Automated Machine Learning system was evaluated with two different datasets, (i) The Early Crop Weeds dataset and (ii) the Plant Seedlings dataset, covering the weeds identification problem. Different configurations, such as the use of plant segmentation, the use of classifier ensembles instead of Softmax and training with noisy data, have been compared. The results showed promising performances of 93.8% and 90.74% F1 score depending on the dataset used. These performances were aligned with other related works in AutoML, but they are far from machine-learning-based systems manually fine-tuned by human experts. From these results, it can be concluded that finding a balance between manual expert work and Automated Machine Learning will be an interesting path to work in order to increase the efficiency in plant protection.

 Artículos similares

       
 
Natnael Gonfa Berihun, Cyrille Dongmo and John Andrew Van der Poll    
Mobile applications are developed and released to the market every day. Due to the intense usage of mobile applications, their quality matters. End users? rejection of mobile apps increases from time to time due to their low quality and lack of proper mo... ver más
Revista: Computers

 
Elena Jasiuniene, Bengisu Yilmaz, Damira Smagulova, Gawher Ahmad Bhat, Vaidotas Cicenas, Egidijus ?ukauskas and Liudas Ma?eika    
The aim of this work is to achieve reliable nondestructive evaluation (NDE) of adhesively bonded aerospace components by developing novel multidimensional data fusion techniques, which would combine the information obtained by ultrasonic and X-ray NDE me... ver más
Revista: Applied Sciences

 
Sérgio D. Correia, João Fé, Slavisa Tomic and Marko Beko    
The present work addresses the development of a test-bench for the embedded implementation, validity, and testing of the recently proposed Improved Elephant Herding Optimization (iEHO) algorithm, applied to the acoustic localization problem. The implemen... ver más
Revista: Computers

 
Malwina Brozyna, Anna Zywicka, Karol Fijalkowski, Damian Gorczyca, Monika Oleksy-Wawrzyniak, Karolina Dydak, Pawel Migdal, Bartlomiej Dudek, Marzenna Bartoszewicz and Adam Junka    
Herein, we present a new test, dubbed AntiBioVol, to be used for the quantitative evaluation of antibiofilm activity of volatile compounds in vitro. AntiBioVol is performed in two 24-well plates using a basic microbiological laboratory equipment. To demo... ver más
Revista: Applied Sciences

 
Sittisak Rattanaprapawan,Apisak Sindhuphak,Krissana Kiddee,Sompol Dumrongsatien     Pág. pp. 196 - 214
The research focuses on digital sketching aspect through a depiction of Himavanta creatures and characters from Thailand, Laos, and Myanmar, where the details of paintings, sculptures, and architectural features as well as multicultural aspiration a... ver más