Classification of urban multi-angular image sequences by aligning their manifolds
| Autor: | Maxime Trolliet, Michele Volpi, Devis Tuia |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2013 |
| Předmět: |
Similarity (geometry)
010504 meteorology & atmospheric sciences Contextual image classification Computer science business.industry 0211 other engineering and technologies Nonlinear dimensionality reduction 02 engineering and technology 01 natural sciences Manifold Data acquisition Reflection (physics) Radiance Life Science Computer vision Bidirectional reflectance distribution function Artificial intelligence business 021101 geological & geomatics engineering 0105 earth and related environmental sciences |
| Zdroj: | Joint Urban Remote Sensing Event 2013, JURSE 2013. IEEE computer society Joint Urban Remote Sensing Event 2013, JURSE 2013 |
| Popis: | When dealing with multi-angular image sequences, problems of reflectance changes due either to illumination and acquisition geometry, or to interactions with the atmosphere, naturally arise. These phenomena interplay with the scene and lead to a modification of the measured radiance: for example, according to the angle of acquisition, tall objects may be seen from top or from the side and different light scatterings may affect the surfaces. This results in shifts in the acquired radiance, that make the problem of multi-angular classification harder and might lead to catastrophic results, since surfaces with the same reflectance return significantly different signals. In this paper, rather than performing atmospheric or bi-directional reflection distribution function (BRDF) correction, a non-linear manifold learning approach is used to align data structures. This method maximizes the similarity between the different acquisitions by deforming their manifold, thus enhancing the transferability of classification models among the images of the sequence. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|