Title :
Multitask Remote Sensing Data Classification
Author :
Leiva-Murillo, Jose M. ; Gómez-Chova, Luis ; Camps-Valls, Gustavo
Author_Institution :
Dept. of Signal Theor. & Commun., Univ. Carlos III de Madrid, Leganés, Spain
Abstract :
Many remote sensing data processing problems are inherently constituted by several tasks that can be solved either individually or jointly. For instance, each image in a multitemporal classification setting could be taken as an individual task. Here, the relation to previous acquisitions should be properly considered because of the nonstationary behavior of temporal, spatial, and angular image features which gives rise to distribution changes. This phenomenon is known as covariate shift. Additionally, when labeled data are scarce or expensive to obtain, the small sample-set problem arises, which makes solving the problems independently in each domain difficult. Multitask learning (MTL) aims at jointly solving a set of prediction problems by sharing information across tasks. This paper introduces MTL in remote sensing data classification. The proposed methods alleviate the data set shift by imposing cross-information in the classifiers through matrix regularization. We consider the support vector machine (SVM) as the core learner and two different regularization schemes: 1) the inclusion of relational operators between tasks and 2) the pairwise Euclidean distance of the predictors in the Hilbert space. These methods rely on simple and intuitive modifications of the kernel used in the standard SVM. Experiments are conducted in three challenging remote sensing problems: cloud screening from multispectral images, land-mine detection using radar data, and multitemporal and multisource image classification. The pairwise method consistently outperforms standard independent and aggregate approaches by about +2% to 4% in all problems at no additional cost. Also, the solutions found give us information about the distribution shift among tasks.
Keywords :
geophysical image processing; image classification; learning (artificial intelligence); matrix decomposition; remote sensing; support vector machines; Hilbert space pairwise predictor Euclidean distance; MTL; SVM; angular image features; cloud screening; covariate shift; cross information; data processing problems; image feature nonstationary behavior; land mine detection; matrix regularization; multisource image classification; multispectral images; multitask learning; multitask remote sensing data classification; multitemporal classification; multitemporal image classification; radar data; regularization schemes; relational operators; small sample set problem; spatial image features; support vector machine; temporal image features; Aggregates; Kernel; Machine learning; Remote sensing; Standards; Support vector machines; Training; Cloud screening; covariate shift (CS); data set bias; domain adaptation; image classification; land-mine detection; multitask learning (MTL); support vector machine (SVM); urban monitoring;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
DOI :
10.1109/TGRS.2012.2200043