ABSTRACT
A Line object extraction from satellite imagery is presented in this thesis. The extraction technique can be significant to both application and development of digital remote sensing methodology the analysis proceeds from satellite imagery to the detailed analysis of specific regions. Smoothing, Laplacian Transform, Line detection and Threshold techniques are used to extract a line object from satellite imagery. This extraction technique may eventually be extended to recognize different patterns for examples, rivers, runway or highway because there are many kind of linear feature. The algorithms for locating linear features are of general interest. A data from National Institute for Aeronautic and Space (LAPAN) was used which cover portions of Cengkareng and Jakarta Area. Its found that result will always contain a significant amount of noise and we can get information about the object early.