Yield prediction with machine learning algorithms and satellite images

Sharifi, Alireza

doi:10.1002/jsfa.10696

Cited by 103 publications

(49 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Still, the amount of empirical data included in the modeling remains a controversial issue. In many works, the authors of the models use a lot of independent variables [3,12,44,45] or use classical prognostic models developed exclusively for potato: SUBSTOR-Potato [46,47], LINTUL-Potato-DSS Model [48], etc. In that situation, when the model tries to estimate too many unknowns for the number of observations made, the model's ability to detect real relationships is severely limited [49].…”

Section: Discussionmentioning

confidence: 99%

The Application of Multiple Linear Regression and Artificial Neural Network Models for Yield Prediction of Very Early Potato Cultivars before Harvest

et al. 2021

View full text Add to dashboard Cite

Yield forecasting is a rational and scientific way of predicting future occurrences in agriculture—the level of production effects. Its main purpose is reducing the risk in the decision-making process affecting the yield in terms of quantity and quality. The aim of the following study was to generate a linear and non-linear model to forecast the tuber yield of three very early potato cultivars: Arielle, Riviera, and Viviana. In order to achieve the set goal of the study, data from the period 2010–2017 were collected, coming from official varietal experiments carried out in northern and northwestern Poland. The linear model has been created based on multiple linear regression analysis (MLR), while the non-linear model has been built using artificial neural networks (ANN). The created models can predict the yield of very early potato varieties on 20th June. Agronomic, phytophenological, and meteorological data were used to prepare the models, and the correctness of their operation was verified on the basis of separate sets of data not participating in the construction of the models. For the proper validation of the model, six forecast error metrics were used: i.e., global relative approximation error (RAE), root mean square error (RMS), mean absolute error (MAE), and mean absolute percentage error (MAPE). As a result of the conducted analyses, the forecast error results for most models did not exceed 15% of MAPE. The predictive neural model NY1 was characterized by better values of quality measures and ex post forecast errors than the regression model RY1.

show abstract

Section: Discussionmentioning

confidence: 99%

The Application of Multiple Linear Regression and Artificial Neural Network Models for Yield Prediction of Very Early Potato Cultivars before Harvest

et al. 2021

View full text Add to dashboard Cite

show abstract

“…After the thresholding step, a machine learning classification (MLC) based on the kernel is used to classify remotely sensed images. It uses a Bayesian formulation of a linear model with an appropriate prior which leads to model sparseness (Sharifi, 2020c). Sparsity means few non‐zero coefficients define the model which provides an accurate classification using the limited training data (Pal and Foody, 2012).…”

Section: Methodsmentioning

confidence: 99%

Development of a method for flood detection based on Sentinel‐1 images and classifier algorithms

Sharifi

2021

Water & Environment J

Self Cite

View full text Add to dashboard Cite

Floods are one of the most devastating natural disasters in the world, displacing millions of people each year and causing severe damage to people's lives and infrastructure. It is the most widespread hydrological hazard worldwide that affects water management, nature protection, economic activities, hydromorphological alterations on ecosystem services and human health. Real‐time monitoring systems play a key role in flood risk reduction and disaster response decisions. Studies have shown that using earth observation data for flood monitoring and timely actions based on good quality information reduces damages. In this paper, after thresholding, a machine learning algorithm and an object‐based classification method were used to classify the SAR data. Thresholding helps detect regions in the flooded areas. A comparison of the results showed that the machine learning algorithm obtained significant results. Because of the results obtained, the usefulness of Sentinel‐1 images as a baseline data for the improvement of the methodological guide is appreciated and should be used as a new source to monitor the flood risks.

show abstract

“…One can obtain efficient representations using techniques such as band selection [2], multi-modal learning [3] dimensionality reduction [4]. Also, advanced methods and technologies in the field of machine learning have provided the conditions to benefit from hyperspectral image information in various fields such as natural language processing [5], medical connection prediction [6], remote sensing image processing [4][5][6][7][8][9], etc. In the early stages, traditional classification methods are based on spectral information, which generally includes two main elements: feature engineering and classifiers [10].…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Image Classification Using a Hybrid 3D-2D Convolutional Neural Networks

Ghaderizadeh

Abbasi‐Moghadam

Sharifi

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

137

View full text Add to dashboard Cite

Due to the unique feature of the three -dimensional convolution neural network, it is used in image classification. For There are some problems such as noise, lack of labeled samples, the tendency to overfitting, a lack of extraction of spectral and spatial features, which has challenged the classification. Among the mentioned problems, the lack of experimental samples is the main problem that has been used to solve the methods in recent years. Among them, convolutional neural network-based algorithms have been proposed as a popular option for hyperspectral image analysis due to their ability to extract useful features and high performance. The traditional CNN-based methods mainly use the 2D-CNN for feature extraction, which makes the interband correlations of HS Is underutilized. The 3D-CNN extracts the joint spectral-spatial information representation, but it depends on a more complex model. To address these issues, the report uses a 3D fast learning block (depthwise separable convolution block and a fast convolution block) followed by a 2D convolutional neural network was introduced to extract spectral-spatial features. Using a hybrid CNN reduces the complexity of the model compared to using 3D-CNN alone and can also perform well against noise and a limited number of training samples. In addition, a series of optimization methods including batch normalization, dropout, exponential decay learning rate, and L2 regularization are adopted to alleviate the problem of overfitting and improve the classification results. To test the performance of this hybrid method, it is performed on the S alinas, University Pavia and Indian Pines datasets, and the results are compared with 2D-CNN and 3D-CNN deep learning models with the same number of layers.

show abstract

Yield prediction with machine learning algorithms and satellite images

Cited by 103 publications

References 37 publications

The Application of Multiple Linear Regression and Artificial Neural Network Models for Yield Prediction of Very Early Potato Cultivars before Harvest

The Application of Multiple Linear Regression and Artificial Neural Network Models for Yield Prediction of Very Early Potato Cultivars before Harvest

Development of a method for flood detection based on Sentinel‐1 images and classifier algorithms

Hyperspectral Image Classification Using a Hybrid 3D-2D Convolutional Neural Networks

Contact Info

Product

Resources

About