Helmi Zulhaidi Mohd Shafri / Universiti Putra Malaysia
Nurin Syuhada Suliman / Universiti Putra Malaysia
Aidi Hizami Ales @ Alias / Universiti Putra Malaysia
Building Information Modeling – Geographical Information System integration, for site assessment, is useful in evaluating the topographical information of the site by using unmanned aerial vehicle (UAV) and terrestrial data. Thus, this study aimed to produce a BIM-GIS integrated 3D model with high accuracy for site assessment. The objectives of this study are to produce the model by using UAV data which are aerial images and Light Detection and ranging (LiDAR) data, together with terrestrial images, in BIM software. The model was then imported into GIS. In GIS, the raster dataset undergoes supervised classification where the percent accuracy and kappa value of Spectral Angle Mapper and minimum distance algorithms were compared. In-situ data verification was done to assess how the model varies from the real world. Based on the result, the 3D reality meshes model of the site can be developed by integrating aerial and terrestrial photographs, where the accuracy of the model is highly dependent on the coverage of data. In producing 3D models, the BIM software has limitation in modelling site with space constraints and object with incoherent GPS tags. Next, the supervised classification of both algorithms shows that minimum distance algorithm gives higher percent accuracy and kappa value. The in-situ data verification of horizontal features obtains an average accuracy of 0.60 cm or 0.18% of actual measurement, where the difference of both control and model measurements is not significant.