Advances in Kernel Machines for Image Classification and Biophysical Parameter Retrieval
Remote sensing data analysis is knowing an unprecedented upswing fostered by the activities of the public and private sectors of geospatial and environmental data analysis. Modern imaging sensors offer the necessary spatial and spectral information to tackle a wide range problems through Earth Observation, such as land cover and use updating, urban dynamics, or vegetation and crop monitoring. In the upcoming years even richer information will be available: more sophisticated hyperspectral sensors with high spectral resolution, multispectral sensors with sub-metric spatial detail or drones that can be deployed in very short time lapses. Besides such opportunities, these new and wealthy infor…
Unsupervised change detection with kernels
In this paper an unsupervised approach to change detection relying on kernels is introduced. Kernel based clustering is used to partition a selected subset of pixels representing both changed and unchanged areas. Once the optimal clustering is obtained the estimated representatives (centroids) of each group are used to assign the class membership to all others pixels composing the multitemporal scenes. Different approaches of considering the multitemporal information are considered with accent on the computation of the difference image directly in the feature spaces. For this purpose a difference kernel approach is successfully adopted. Finally an effective way to cope with the estimation o…
Spectral alignment of multi-temporal cross-sensor images with automated kernel canonical correlation analysis
In this paper we present an approach to perform relative spectral alignment between optical cross-sensor acquisitions. The proposed method aims at projecting the images from two different and possibly disjoint input spaces into a common latent space, in which standard change detection algorithms can be applied. The system relies on the regularized kernel canonical correlation analysis transformation (kCCA), which can accommodate nonlinear dependencies between pixels by means of kernel functions. To learn the projections, the method employs a subset of samples belonging to the unchanged areas or to uninteresting radiometric differences. Since the availability of ground truth information to p…
A survey of active learning algorithms for supervised remote sensing image classification
Defining an efficient training set is one of the most delicate phases for the success of remote sensing image classification routines. The complexity of the problem, the limited temporal and financial resources, as well as the high intraclass variance can make an algorithm fail if it is trained with a suboptimal dataset. Active learning aims at building efficient training sets by iteratively improving the model performance through sampling. A user-defined heuristic ranks the unlabeled pixels according to a function of the uncertainty of their class membership and then the user is asked to provide labels for the most uncertain pixels. This paper reviews and tests the main families of active …