Corn and Wheat Plant Identification on Radar and Optical Image Data

Mustafa Öğretmen; Abdülkadir Gümüşçü

Makale detayları

Corn and Wheat Plant Identification on Radar and Optical Image Data

Yazarlar:

Mustafa Öğretmen | Abdülkadir Gümüşçü

Yayıncı:

Kastamonu University

Ana Başlıklar:

Makine Öğrenme (Diğer) | Yapay Zeka (Diğer) | Bilgisayar Yazılımı

Dergi:

Inspiring Technologies and Innovations Vol.4 Issue 1, pp.7-17

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001352105

pISSN:

2822-6062

Anahtar kelimeler:

Özet:

 In recent years, prediction, detection, and classification applications have been made in many fields such as agriculture, health, stock market, economy, cybersecurity, etc., in Machine Learning and Artificial Intelligence. These applications are user-friendly and provide fast, high-quality, and accurate results. The advancements in these fields have shown that machine learning and deep learning methods are very useful in classifying large and complex data, especially when human brain and physical power are insufficient. Todays findings suggest there have been promising studies using these models, focused on time- and cost-effective and high-quality products. These studies provide efficiency in agricultural areas, thereby guiding both farmers and policymakers. In addition, the development and widespread implementation of unmanned aerial vehicles UAVs accelerated the process of obtaining multispectral aerial images. With the combined use of these technologies and high-speed computer software and hardware for precise and high-quality production in agriculture, it was possible to determine plant species and increase product quality. In this study, a dataset consisting of radar and optical image data was used to classify corn and wheat crops cultivated in agricultural areas. Four different machine learning models, namely Decision Tree DT, K-Nearest Neighbors K-NN, Naive Bayes NB, and Support Vector Machines SVM, were trained and compared on the dataset consisting of 174 features from Winnipeg, Canada. The dataset has been divided into 80 for training and 20 for testing. According to the results, the SVM model performed the best with the highest accuracy 0.9998 and F1-Score 0.9996, while the NB model performed the worst accuracy 0.9895 and F1-Score 0.9835. The detection of wheat and corn crop types by processing radar and optical image data with machine learning models has shown that other crops in cultivated lands in the Southeastern Anatolia Project GAP region can be classified using the same method, which shows the importance of this study.

 In recent years, prediction, detection, and classification applications have been made in many fields such as agriculture, health, stock market, economy, cybersecurity, etc., in Machine Learning and Artificial Intelligence. These applications are user-friendly and provide fast, high-quality, and accurate results. The advancements in these fields have shown that machine learning and deep learning methods are very useful in classifying large and complex data, especially when human brain and physical power are insufficient. Todays findings suggest there have been promising studies using these models, focused on time- and cost-effective and high-quality products. These studies provide efficiency in agricultural areas, thereby guiding both farmers and policymakers. In addition, the development and widespread implementation of unmanned aerial vehicles UAVs accelerated the process of obtaining multispectral aerial images. With the combined use of these technologies and high-speed computer software and hardware for precise and high-quality production in agriculture, it was possible to determine plant species and increase product quality. In this study, a dataset consisting of radar and optical image data was used to classify corn and wheat crops cultivated in agricultural areas. Four different machine learning models, namely Decision Tree DT, K-Nearest Neighbors K-NN, Naive Bayes NB, and Support Vector Machines SVM, were trained and compared on the dataset consisting of 174 features from Winnipeg, Canada. The dataset has been divided into 80 for training and 20 for testing. According to the results, the SVM model performed the best with the highest accuracy 0.9998 and F1-Score 0.9996, while the NB model performed the worst accuracy 0.9895 and F1-Score 0.9835. The detection of wheat and corn crop types by processing radar and optical image data with machine learning models has shown that other crops in cultivated lands in the Southeastern Anatolia Project GAP region can be classified using the same method, which shows the importance of this study.

 In recent years, prediction, detection, and classification applications have been made in many fields such as agriculture, health, stock market, economy, cybersecurity, etc., in Machine Learning and Artificial Intelligence. These applications are user-friendly and provide fast, high-quality, and accurate results. The advancements in these fields have shown that machine learning and deep learning methods are very useful in classifying large and complex data, especially when human brain and physical power are insufficient. Todays findings suggest there have been promising studies using these models, focused on time- and cost-effective and high-quality products. These studies provide efficiency in agricultural areas, thereby guiding both farmers and policymakers. In addition, the development and widespread implementation of unmanned aerial vehicles UAVs accelerated the process of obtaining multispectral aerial images. With the combined use of these technologies and high-speed computer software and hardware for precise and high-quality production in agriculture, it was possible to determine plant species and increase product quality. In this study, a dataset consisting of radar and optical image data was used to classify corn and wheat crops cultivated in agricultural areas. Four different machine learning models, namely Decision Tree DT, K-Nearest Neighbors K-NN, Naive Bayes NB, and Support Vector Machines SVM, were trained and compared on the dataset consisting of 174 features from Winnipeg, Canada. The dataset has been divided into 80 for training and 20 for testing. According to the results, the SVM model performed the best with the highest accuracy 0.9998 and F1-Score 0.9996, while the NB model performed the worst accuracy 0.9895 and F1-Score 0.9835. The detection of wheat and corn crop types by processing radar and optical image data with machine learning models has shown that other crops in cultivated lands in the Southeastern Anatolia Project GAP region can be classified using the same method, which shows the importance of this study.

 In recent years, prediction, detection, and classification applications have been made in many fields such as agriculture, health, stock market, economy, cybersecurity, etc., in Machine Learning and Artificial Intelligence. These applications are user-friendly and provide fast, high-quality, and accurate results. The advancements in these fields have shown that machine learning and deep learning methods are very useful in classifying large and complex data, especially when human brain and physical power are insufficient. Todays findings suggest there have been promising studies using these models, focused on time- and cost-effective and high-quality products. These studies provide efficiency in agricultural areas, thereby guiding both farmers and policymakers. In addition, the development and widespread implementation of unmanned aerial vehicles UAVs accelerated the process of obtaining multispectral aerial images. With the combined use of these technologies and high-speed computer software and hardware for precise and high-quality production in agriculture, it was possible to determine plant species and increase product quality. In this study, a dataset consisting of radar and optical image data was used to classify corn and wheat crops cultivated in agricultural areas. Four different machine learning models, namely Decision Tree DT, K-Nearest Neighbors K-NN, Naive Bayes NB, and Support Vector Machines SVM, were trained and compared on the dataset consisting of 174 features from Winnipeg, Canada. The dataset has been divided into 80 for training and 20 for testing. According to the results, the SVM model performed the best with the highest accuracy 0.9998 and F1-Score 0.9996, while the NB model performed the worst accuracy 0.9895 and F1-Score 0.9835. The detection of wheat and corn crop types by processing radar and optical image data with machine learning models has shown that other crops in cultivated lands in the Southeastern Anatolia Project GAP region can be classified using the same method, which shows the importance of this study.

Makale detayları

Corn and Wheat Plant Identification on Radar and Optical Image Data

Yazarlar:

Mustafa Öğretmen | Abdülkadir Gümüşçü

Yayıncı:

Kastamonu University

Ana Başlıklar:

Makine Öğrenme (Diğer) | Yapay Zeka (Diğer) | Bilgisayar Yazılımı

Dergi:

Inspiring Technologies and Innovations Vol.4 Issue 1, pp.7-17

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001352105

pISSN:

2822-6062

machine learning

unmanned aerial vehicle

multispectral image

agriculture

plant classification

Sayı

Makale detayları

Corn and Wheat Plant Identification on Radar and Optical Image Data

Yazarlar: Mustafa Öğretmen | Abdülkadir Gümüşçü

Yayıncı: Kastamonu University

Ana Başlıklar: Makine Öğrenme (Diğer) | Yapay Zeka (Diğer) | Bilgisayar Yazılımı

Dergi: Inspiring Technologies and Innovations Vol.4 Issue 1, pp.7-17

Belge Tipi: Makale

Yayın Tarihi: 6/2025

Belge Dili: İngilizce

Document ID (Org No.): 2001352105

pISSN: 2822-6062

machine learning

unmanned aerial vehicle

multispectral image

agriculture

plant classification

Sayı

Yazarlar:

Mustafa Öğretmen | Abdülkadir Gümüşçü

Yayıncı:

Kastamonu University

Ana Başlıklar:

Makine Öğrenme (Diğer) | Yapay Zeka (Diğer) | Bilgisayar Yazılımı

Dergi:

Inspiring Technologies and Innovations Vol.4 Issue 1, pp.7-17

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001352105

pISSN:

2822-6062