TECHNICAL DETAILS: Recently, scientists have begun to understand the visual cortex of small mammals. This understanding has led to the creation of new algorithms that are achieving new levels of sophistication in electronic image processing. Pulse-Coupled Neural Networks (PCNN) became, recently, of major interest. The PCNN is an image transform that removes "unimportant" details while improving the overall quality of the image. The thesis will address on the development of an algorithm for automatic urban shadow detection and removal by using PCNN in very-high spatial resolution optical imagery (about 60 cm).
FRAMEWORK: Optical satellite images over urban areas are usually affected
by the shadows of the buildings. If these shadowed areas can be detected
and removed, objects in the images will appear more evident, so that they
can be recognized correctly, allowing, therefore, a better understanding
of the urban environment, as required by a variety of engineering applications.
OFFERED BENEFITS: The selected student will learn to analyze very-high
spatial resolution satellite imagery over urban areas. Moreover, he/she
will work in an international context, and will acquire a good knowledge
of the IDL programming language.
