SAR Ship Detection Using Image Histograms and Machine Learning Approach

---

Hanbay K., Çalışan M., Özdemir T. B.

Türk Doğa ve Fen Dergisi, cilt.13, sa.3, ss.171-175, 2024 (Hakemli Dergi)

• Yayın Türü: Makale / Tam Makale
• Cilt numarası: 13 Sayı: 3
• Basım Tarihi: 2024
• Doi Numarası: 10.46810/tdfd.1528267
• Dergi Adı: Türk Doğa ve Fen Dergisi
• Derginin Tarandığı İndeksler: TR DİZİN (ULAKBİM)
• Sayfa Sayıları: ss.171-175
• İnönü Üniversitesi Adresli: Evet

Özet

Ship detection and classification in SAR images is an important and active research
area. It provides effective surveillance by facilitating applications such as surveillance and
tracking of commercial and military ships. In this study, SAR images were classified using
Hessian matrix and HOG algorithm. Using the eigenvalues of the Hessian matrix, the angle and
orientation information of the HOG method was calculated. Thus, distinctive pixel
characteristics were coded. Our method has obtained desired results in terms of classification
accuracy. The proposed method achieved 94.50% classification success.