|
Descripción
|
|
|---|---|
| The analysis of fragmented agricultural scenes has been approached by two methodologies. The first one focuses on providing a framework for multi-scale segmentation and, at the same time, a way to identify the best scale according to criteria of spectral variability of the regions at each scale. While this methodology provides spectrally similar regions at all scales, the variability of agricultural covers hinders to correctly establish the plot boundaries. This issue is further discussed in later approaches. The second methodology is aimed at generating thematic maps by combining only two scales which correspond to segments generated by a segmentation based on edges and the other based on superpixels. For mapping lands covers, it has combined the results of classifying the superpixels through a supervised classifier and the segments based on edges by a set of rules. The combination of both scales has yielded results with an accuracy better than those obtained on both scales separately. Delineation of agricultural plots has been approached by two different methodologies. The first approach uses a supervised classification method to segment the image by agglomerating superpixels. This methodology represents an alternative to traditional methods of segmentation, which is based on learning how agricultural plots in a similar way as a human operator, as it does. In this regard, unlike conventional methods that require to find a suitable combination of segmentation parameters, the proposed method has the advantage that the classifier is able to find relationships in a multidimensional space to facilitate an adequate segmentation. The main drawback of this method is that it requires large amounts of annotated information for training the classifier. In this regard, a second method which uses only the image information for the delimitation has been proposed. From the basis that segmentations (superpixels) obtained by different parameters can capture various phenomena at different scales of an image, the proposed methodology allows to exploit its edges to obtain by consensus a segmentation of agricultural plots. It was found that superpixels can reduce the image noise, while most of the edges corresponding to the plots are kept in different segmentations. Both approaches are complementary, and depending on the availability of annotated data one or the other can be used, the first method can be used when there is ground truth, and the second one when there is no such information. Finally, to capture the internal variability of agricultural parcels, it has proposed a method based on superpixels that considers spatial and temporal components of multiple images to find temporarily homogeneous regions within plots. Further analysis of the behavior of generated superpixels through the different dates provided the information about the spatio-temporal variability inside the plots. In this regard, three behaviors have been considered to determine the variability: (i) no variability, (ii) low-persistence variability, and (iii) high-persistence variability. This considerations allowed to create a map of spatio-temporal dynamics that provides an overview of the variability inside plots. | |
|
Internacional
|
Si |
|
ISBN
|
|
|
Tipo de Tesis
|
Doctoral |
|
Calificación
|
Sobresaliente cum laude |
|
Fecha
|
13/10/2016 |