Pixel Based Classification of Images

Nowadays, automation became a basic theme in all the fields of activity. Reducing the operator tasks from a workflow led to the need of research extensions in automation. In Photogrammetry we can start this process fromthedata collection and going to the final product generation. The main goal of this article is to facilitate the updating process of the National Geospatial Database and to ensure support for different productsof our institution. The study was conducted on two types of coverage: extraurban and urban areas, both in Bihor County. The segmentation and classification process used the Single Feature Probability method, consideringattributes like spectral signature,texture and shape. The results depend on the initial data, the workflow used and the samples collected.