Wave-CNN Method Approach of Archaeogeophysics Studies

Albora A. M.

24th Signal Processing and Communication Application Conference (SIU), Zonguldak, Türkiye, 16 - 19 Mayıs 2016, ss.173-176 identifier identifier

  • Cilt numarası:
  • Doi Numarası: 10.1109/siu.2016.7495705
  • Basıldığı Şehir: Zonguldak
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.173-176


In recent years, geophysical approaches are being commonly used in modeling and pre-processing of excavation of buried archeological structures. The methods to be applied for the archeological structures are decided due to the comments of the archeologists. Thus more information about the place and the type of the structure of the archaeological ruins can be obtained. Thus only the area that archaeological structures with the estimated depth and borders are excavated without distorting original historical ruin. In this algorithm it is integrated the achievements of Wavelet Method (in detecting edges and corners) and CNN (Cellular Neural Network) methods (in regional-residual separation). In this study, first the Wave-CNN method has been tested with synthetic data. Then Wave-CNN approach is applied to the total magnetic anomaly map of the Hittite Empire, established in Sivas-Altinyayla, Turkey and residual anomalies are found. The walls of the historical city and the borders and the characteristics of these walls are well evaluated by these residual anomaly outputs.