Automatic detection of illegally constructed buildings using single epoch UAV data


ACAR H.

Gumushane Universitesi Fen Bilimleri Dergisi, cilt.16, sa.2, ss.350-361, 2026 (Scopus, TRDizin) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 16 Sayı: 2
  • Basım Tarihi: 2026
  • Doi Numarası: 10.17714/gumusfenbil.1856457
  • Dergi Adı: Gumushane Universitesi Fen Bilimleri Dergisi
  • Derginin Tarandığı İndeksler: Scopus, TR DİZİN (ULAKBİM)
  • Sayfa Sayıları: ss.350-361
  • Anahtar Kelimeler: 3D Point cloud, Illegal building detection, UAV
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

Rapid urbanization has led to an increase in construction activities, making it necessary that buildings are constructed in compliance with approved projects and current legal regulations. Buildings constructed in accordance with expert approved designs can maintain their structural integrity, particularly during natural disasters. This plays a significant role in ensuring the safety of life and property. In addition, buildings constructed based on approved projects provide important economic contributions to the public budget through taxation systems. In this context, the regular monitoring of urban areas and keeping spatial data up to date are of great importance for planning, inspection, and legal property management. This study aims to automatically detect illegal buildings that are constructed in violation of approved projects or are not legally permitted. Within this scope, building detection was carried out using three dimensional (3D) data obtained by Unmanned Aerial Vehicles (UAVs), and the results were analyzed with respect to cadastral data. As input data, a 3D point cloud produced from optical images acquired during single period UAV flights and 1:1000 scale base maps obtained from the Arsin Municipality of Trabzon were used. After eliminating vegetation and ground points, building roof points were automatically identified. Then, building polygons and roof points are analyzed in a three dimensional environment using a rule based approach to automatically identify illegal structures. The proposed approach successfully detected all the buildings that were manually identified by the municipality within the study area and additionally identified 7 extra illegal structures or building extensions. Preventive measures based on automatic illegal building detection may help reduce the number of people affected by natural disasters and prevent or reduce property related tax losses faced by public institutions.