Veselin Pizurica

Learn More
In this paper we p r e s e n t a model-based approach to the automatic extraction of linear features, like roads and paths, from aerial optical images. The proposed method consists of two steps. The rst step utilizes local information related to the geometry and radiometry of the structures to be extracted. It consists of a series of morphological ltering(More)
Experiments on mine detection with short range ground-based sensors (GPR, dynamic and static IR imaging, US reeectometry) are reviewed. The emphasis is on the processing and pattern recognition algorithms, although these cannot be discussed without taking into account s o m e o f the physical principles of the sensors and data acquisition circumstances. The(More)
We investigate the applicability of image analysis to the problem of minefield detection using airborne remote sensing images. We have applied different image processing methods for the detection of minefields, minefield indicators and individual mines. The proposed algorithms involve the extraction of linear features (roads, paths, fences, wires,…),(More)
Experiments on airborne mineeeld identiication are reviewed in "Part 1". "Part 2", in these same proceedings, describes ground based experiments. The emphasis in this part is on the processing and pattern recognition algorithms. Preliminary results are shown, further reenement and redesign are needed before going to real life situations.
We investigate the applicability of image analysis to the problem of minefield detection using airborne remote sensing images. We have applied different image processing methods for the detection of minefields, minefield indicators and individual mines. The proposed algorithms involve the extraction of linear features (roads, paths, fences, wires,…),(More)
  • 1