Remote Sensing-Based Deep Learning Approach for Identifying Burned Forest Areas

Reha Paşaoğlu; Ahmet Ertuğrul Arık; Nuri Emrahaoğlu

Makale detayları

Remote Sensing-Based Deep Learning Approach for Identifying Burned Forest Areas

Yazarlar:

Reha Paşaoğlu | Ahmet Ertuğrul Arık | Nuri Emrahaoğlu

Yayıncı:

Cukurova University

Ana Başlıklar:

Yapay Zeka (Diğer) | Fotogrametri ve Uzaktan Algılama

Dergi:

Çukurova Üniversitesi Mühendislik Fakültesi Dergisi Vol.40 Issue 1, pp.33-48

Belge Tipi:

Makale

Yayın Tarihi:

3/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001343966

pISSN:

2757-9255

Anahtar kelimeler:

Özet:

 In this study, the burnt areas and intensity of forest fires that occurred in the Samandağ region of Hatay between September 5-10, 2020, are mapped. Analyses were carried out using deep learning, remote sensing, and satellite data from Sentinel 2. With Sentinel 2 satellite photos of the research locations, an image dataset for deep learning was constructed. Then, using deep learning approaches, a deep learning model was developed, trained using the photos in the dataset, and successfully tested. Images from Sentinel 2 were used to produce the Normalized Burn RatioNBR and Burnt Area Index for Sentinel 2 BAIS2 indices using the results of a new deep learning model. Calculating the Difference Normalized Burning Intensity dNBR and Burnt Area Index for Difference Sentinel-2 dBAIS2 values for the discrepancies between these indices before and after the fire allowed for categorization and determination of the fire area. The deep learning approach burnt area indexes, and General Directorate of Forestry GDF fire registration slips were compared, and it was established that the new deep learning model was more effective at locating burned forest areas than the indexes. In identifying the burnt forest areas, the new model has a proportionate accuracy of 98.36 in the Samandağ study region.

 In this study, the burnt areas and intensity of forest fires that occurred in the Samandağ region of Hatay between September 5-10, 2020, are mapped. Analyses were carried out using deep learning, remote sensing, and satellite data from Sentinel 2. With Sentinel 2 satellite photos of the research locations, an image dataset for deep learning was constructed. Then, using deep learning approaches, a deep learning model was developed, trained using the photos in the dataset, and successfully tested. Images from Sentinel 2 were used to produce the Normalized Burn RatioNBR and Burnt Area Index for Sentinel 2 BAIS2 indices using the results of a new deep learning model. Calculating the Difference Normalized Burning Intensity dNBR and Burnt Area Index for Difference Sentinel-2 dBAIS2 values for the discrepancies between these indices before and after the fire allowed for categorization and determination of the fire area. The deep learning approach burnt area indexes, and General Directorate of Forestry GDF fire registration slips were compared, and it was established that the new deep learning model was more effective at locating burned forest areas than the indexes. In identifying the burnt forest areas, the new model has a proportionate accuracy of 98.36 in the Samandağ study region.

 In this study, the burnt areas and intensity of forest fires that occurred in the Samandağ region of Hatay between September 5-10, 2020, are mapped. Analyses were carried out using deep learning, remote sensing, and satellite data from Sentinel 2. With Sentinel 2 satellite photos of the research locations, an image dataset for deep learning was constructed. Then, using deep learning approaches, a deep learning model was developed, trained using the photos in the dataset, and successfully tested. Images from Sentinel 2 were used to produce the Normalized Burn RatioNBR and Burnt Area Index for Sentinel 2 BAIS2 indices using the results of a new deep learning model. Calculating the Difference Normalized Burning Intensity dNBR and Burnt Area Index for Difference Sentinel-2 dBAIS2 values for the discrepancies between these indices before and after the fire allowed for categorization and determination of the fire area. The deep learning approach burnt area indexes, and General Directorate of Forestry GDF fire registration slips were compared, and it was established that the new deep learning model was more effective at locating burned forest areas than the indexes. In identifying the burnt forest areas, the new model has a proportionate accuracy of 98.36 in the Samandağ study region.

 Bu çalışmada, 5-10 Eylül 2020 tarihleri arasında Hatayın Samandağ bölgesinde meydana gelen orman yangınlarının yanık alanları ve şiddeti haritalandırılmıştır. Derin öğrenme, uzaktan algılama ve Sentinel 2 uydu verileri kullanılarak analizler yapılmıştır. Araştırma bölgelerine ait Sentinel 2 uydu fotoğrafları ile derin öğrenme için bir görüntü veri seti oluşturulmuştur. Ardından, derin öğrenme yaklaşımları kullanılarak bir model geliştirilmiş, bu model veri setindeki fotoğraflarla eğitilmiş ve başarıyla test edilmiştir. Sentinel 2den elde edilen görüntüler, yeni derin öğrenme modelinin sonuçları kullanılarak Normalleştirilmiş Yanma Yoğunluğu NBR ve Yanık Alan İndeksi BAIS2 değerleri hesaplanmıştır. Yangın öncesi ve sonrası bu indeksler arasındaki farklılıkların hesaplanmasıyla Farklı Normalleştirilmiş Yanma Yoğunluğu dNBR ve Farklı Yanık Alan İndeksi dBAIS2 değerleri elde edilerek yangın alanı kategorize edilmiş ve belirlenmiştir. Derin öğrenme yaklaşımı, yanık alan indeksleri ve Orman Genel Müdürlüğü yangın kayıt fişleri karşılaştırılmış ve yeni derin öğrenme modelinin yanmış orman alanlarını belirlemede indekslere göre daha etkili olduğu tespit edilmiştir. Samandağ çalışma bölgesinde, yanık orman alanlarının belirlenmesinde yeni modelin doğruluk oranı 98,36 olarak hesaplanmıştır.

Makale detayları

Remote Sensing-Based Deep Learning Approach for Identifying Burned Forest Areas

Yazarlar:

Reha Paşaoğlu | Ahmet Ertuğrul Arık | Nuri Emrahaoğlu

Yayıncı:

Cukurova University

Ana Başlıklar:

Yapay Zeka (Diğer) | Fotogrametri ve Uzaktan Algılama

Dergi:

Çukurova Üniversitesi Mühendislik Fakültesi Dergisi Vol.40 Issue 1, pp.33-48

Belge Tipi:

Makale

Yayın Tarihi:

3/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001343966

pISSN:

2757-9255

Deep learning

Sentinel 2

NBR-dNBR

BAIS2-dBAIS2

Remote sensingDerin öğrenme

Sentinel 2

NBR-dNBR

BAIS2-dBAIS2

Uzaktan algılama

Sayı

Makale detayları

Yazarlar: Reha Paşaoğlu | Ahmet Ertuğrul Arık | Nuri Emrahaoğlu

Yayıncı: Cukurova University

Ana Başlıklar: Yapay Zeka (Diğer) | Fotogrametri ve Uzaktan Algılama

Dergi: Çukurova Üniversitesi Mühendislik Fakültesi Dergisi Vol.40 Issue 1, pp.33-48

Belge Tipi: Makale

Yayın Tarihi: 3/2025

Belge Dili: İngilizce

Document ID (Org No.): 2001343966

pISSN: 2757-9255

Sayı

Yazarlar:

Reha Paşaoğlu | Ahmet Ertuğrul Arık | Nuri Emrahaoğlu

Yayıncı:

Cukurova University

Ana Başlıklar:

Yapay Zeka (Diğer) | Fotogrametri ve Uzaktan Algılama

Dergi:

Çukurova Üniversitesi Mühendislik Fakültesi Dergisi Vol.40 Issue 1, pp.33-48

Belge Tipi:

Makale

Yayın Tarihi:

3/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001343966

pISSN:

2757-9255