Title :
Robust recognition of buildings in compressed large aerial scenes
Author :
Azencott, Robert ; Durbin, François ; Paumard, José
Author_Institution :
DIAM-CMLA, Ecole Normale Superieure, Cachan, France
Abstract :
This paper shows how it is possible to recognize and localize objects in compressed images. The compression method we choose is based on the extraction of the quincunx multiscale edges. The edges of the object and the scene are both computed, and then matched using the censored Hausdorff distance. This distance is computed by double truncation of the classical Hausdorff distance. The localization is based on a coarse-to-fine method. Robustness to noise and possible occlusions of the objects is shown. This algorithm is fast on a workstation and we have implemented it on a massively parallel computer, demonstrating real-time feasibility
Keywords :
building; data compression; edge detection; feature extraction; image coding; image matching; object recognition; algorithm; buildings; censored Hausdorff distance; classical Hausdorff distance; coarse-to-fine method; compressed images; compressed large aerial scenes; compression method; double truncation; image matching; massively parallel computer; noise robustness; object localization; object occlusions; object recognition; quincunx multiscale edge extraction; real-time feasibility; robust recognition; workstation; Concurrent computing; Image coding; Image edge detection; Image recognition; Image resolution; Layout; Noise robustness; Pixel; Smoothing methods; Workstations;
Conference_Titel :
Image Processing, 1996. Proceedings., International Conference on
Conference_Location :
Lausanne
Print_ISBN :
0-7803-3259-8
DOI :
10.1109/ICIP.1996.560947
