Enhanced landslide susceptibility prediction with 3D ALOS PALSAR imagery and neural networks: A data-efficient framework

Sohaib K M Abujayyab

Makale detayları

Enhanced landslide susceptibility prediction with 3D ALOS PALSAR imagery and neural networks: A data-efficient framework

Yazarlar:

Sohaib K M Abujayyab

Yayıncı:

Marmara University

Ana Başlıklar:

Coğrafi Bilgi Sistemleri

Dergi:

International Journal of Geography and Geography Education Issue 51, pp.115-126

Belge Tipi:

Makale

Yayın Tarihi:

1/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001310476

pISSN:

2630-6336

Anahtar kelimeler:

Özet:

 Landslide susceptibility mapping LSM founded on DEM is a growing research field with profound implications for human safety and infrastructure preservation. Many existing methods rely on extensive input data to enhance predictive accuracy. This paper aims to introduce a remote sensing-data-requirement framework for LSM. Our approach exclusively leverages a single ALOS PALSAR image, comprising three key steps: 1 Pre-processing, 2 derivation of explanatory variables, and 3 neural network modeling. To begin, we extracted 22 input variables from the ALOS PALSAR image. These variables played a pivotal role in developing the Neural Network NN predictor. The predictor structure consists of 22 variables in the input layer, 150 neurons in the hidden layer, and a single output layer. Our model was trained using 5,829 sample points, and subsequently, it was employed to generate landslide susceptibility LS map with 745,810 points. Based on the Overall accuracy metric, the model exhibited impressive performance accuracy, achieving 89.3 training and 82.3 testing accuracies. Additionally, it demonstrated a strong performance of 95.22 during training and 84.7 during testing according to the ROC curve. In conclusion, the implementation of our proposed method underscores its ability to develop remarkable accuracy model with remote sensing-data-requirement. This framework offers valuable insights for future progress in regions with challenging conditions and extensive data coverage. Moreover, it effectively handles data quality inconsistencies and data updating issues.

 Landslide susceptibility mapping LSM founded on DEM is a growing research field with profound implications for human safety and infrastructure preservation. Many existing methods rely on extensive input data to enhance predictive accuracy. This paper aims to introduce a remote sensing-data-requirement framework for LSM. Our approach exclusively leverages a single ALOS PALSAR image, comprising three key steps: 1 Pre-processing, 2 derivation of explanatory variables, and 3 neural network modeling. To begin, we extracted 22 input variables from the ALOS PALSAR image. These variables played a pivotal role in developing the Neural Network NN predictor. The predictor structure consists of 22 variables in the input layer, 150 neurons in the hidden layer, and a single output layer. Our model was trained using 5,829 sample points, and subsequently, it was employed to generate landslide susceptibility LS map with 745,810 points. Based on the Overall accuracy metric, the model exhibited impressive performance accuracy, achieving 89.3 training and 82.3 testing accuracies. Additionally, it demonstrated a strong performance of 95.22 during training and 84.7 during testing according to the ROC curve. In conclusion, the implementation of our proposed method underscores its ability to develop remarkable accuracy model with remote sensing-data-requirement. This framework offers valuable insights for future progress in regions with challenging conditions and extensive data coverage. Moreover, it effectively handles data quality inconsistencies and data updating issues.

 Landslide susceptibility mapping LSM founded on DEM is a growing research field with profound implications for human safety and infrastructure preservation. Many existing methods rely on extensive input data to enhance predictive accuracy. This paper aims to introduce a remote sensing-data-requirement framework for LSM. Our approach exclusively leverages a single ALOS PALSAR image, comprising three key steps: 1 Pre-processing, 2 derivation of explanatory variables, and 3 neural network modeling. To begin, we extracted 22 input variables from the ALOS PALSAR image. These variables played a pivotal role in developing the Neural Network NN predictor. The predictor structure consists of 22 variables in the input layer, 150 neurons in the hidden layer, and a single output layer. Our model was trained using 5,829 sample points, and subsequently, it was employed to generate landslide susceptibility LS map with 745,810 points. Based on the Overall accuracy metric, the model exhibited impressive performance accuracy, achieving 89.3 training and 82.3 testing accuracies. Additionally, it demonstrated a strong performance of 95.22 during training and 84.7 during testing according to the ROC curve. In conclusion, the implementation of our proposed method underscores its ability to develop remarkable accuracy model with remote sensing-data-requirement. This framework offers valuable insights for future progress in regions with challenging conditions and extensive data coverage. Moreover, it effectively handles data quality inconsistencies and data updating issues.

 Üç boyutlu verilere dayalı heyelan duyarlılık haritalaması LSM, insan güvenliği ve altyapının korunması açısından derin etkileri olan, büyüyen bir çalışma alanıdır. Mevcut yöntemlerin çoğu, tahmin doğruluğunu artırmak için kapsamlı girdi verilerine dayanır. Bu makale, LSM için uzaktan algılama veri gereksinimi çerçevesini tanıtmayı amaçlamaktadır. Yaklaşımımız yalnızca üç temel aşamadan oluşan tek bir 3D ALOS PALSAR görüntüsünden yararlanıyor: A veri ön işleme, B açıklayıcı faktörlerin türetilmesi ve C sinir ağı modelleme. Başlamak için ALOS PALSAR görüntüsünden 22 giriş değişkenini çıkardık. Bu değişkenler, İleri Beslemeli Sinir Ağı FFNN tahmincisinin geliştirilmesinde çok önemli bir rol oynadı. Tahmin yapısı giriş katmanında 22 değişken, gizli katmanda 150 nöron ve tek çıkış katmanından oluşmaktadır. Modelimiz 5.829 örnek nokta kullanılarak eğitilmiş ve ardından 745.810 noktalı heyelan duyarlılık LS haritasının oluşturulmasında kullanılmıştır. Genel doğruluk metriğine dayalı olarak model, eğitim sırasında 89,3 ve test sırasında 82,3e ulaşarak etkileyici bir performans doğruluğu sergiledi. Ayrıca Alıcı Çalışma Karakteristiği eğrisine göre eğitim sırasında 95,22 ve test sırasında 84,7 gibi güçlü bir performans gösterdi. Sonuçlar, Karabükteki çalışma alanının 3,53 km²sinin 3,03 çok yüksek duyarlılık kategorisine girdiğini göstermiştir. Sonuç olarak, önerdiğimiz çerçevenin uygulanması, uzaktan algılama-veri gereksinimi tahminlerini dikkate değer bir doğrulukla geliştirme yeteneğinin altını çiziyor. Bu çerçeve, zorlu topografik koşullara ve kapsamlı veri kapsamına sahip bölgelerdeki gelecekteki gelişmeler için değerli bilgiler sunmaktadır. Ayrıca, veri kalitesi tutarsızlıklarını ve veri güncelleme sorunlarını etkili bir şekilde ele alır.

Makale detayları

Enhanced landslide susceptibility prediction with 3D ALOS PALSAR imagery and neural networks: A data-efficient framework

Yazarlar:

Sohaib K M Abujayyab

Yayıncı:

Marmara University

Ana Başlıklar:

Coğrafi Bilgi Sistemleri

Dergi:

International Journal of Geography and Geography Education Issue 51, pp.115-126

Belge Tipi:

Makale

Yayın Tarihi:

1/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001310476

pISSN:

2630-6336

Landslide Susceptibility

Data-Efficient Framework

ALOS PALSAR

Neural Networks

Topographic AttributesHeyelana Duyarlılık

Veri Verimli Çerçeve

ALOS PALSAR

Sinir Ağları

Topografik Nitelikler.

Sayı

Makale detayları

Yazarlar: Sohaib K M Abujayyab

Yayıncı: Marmara University

Ana Başlıklar: Coğrafi Bilgi Sistemleri

Dergi: International Journal of Geography and Geography Education Issue 51, pp.115-126

Belge Tipi: Makale

Yayın Tarihi: 1/2024

Belge Dili: İngilizce

Document ID (Org No.): 2001310476

pISSN: 2630-6336

Sayı

Yazarlar:

Sohaib K M Abujayyab

Yayıncı:

Marmara University

Ana Başlıklar:

Coğrafi Bilgi Sistemleri

Dergi:

International Journal of Geography and Geography Education Issue 51, pp.115-126

Belge Tipi:

Makale

Yayın Tarihi:

1/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001310476

pISSN:

2630-6336